(Torque and Speed Coupling) Hybrid Drive Train

(Torque and Speed Coupling) Hybrid Drive Train
CRC Press
Taylor & Francis Group
6000 Broken Sound Parkway NW, Suite 300
Boca Raton, FL 33487-2742
© 2010 by Taylor and Francis Group, LLC
CRC Press is an imprint of Taylor & Francis Group, an Informa business
No claim to original U.S. Government works
Printed in the United States of America on acid-free paper
10 9 8 7 6 5 4 3 2 1
International Standard Book Number: 978-1-4200-5398-2 (Hardback)
This book contains information obtained from authentic and highly regarded sources. Reasonable
efforts have been made to publish reliable data and information, but the author and publisher cannot
assume responsibility for the validity of all materials or the consequences of their use. The authors and
publishers have attempted to trace the copyright holders of all material reproduced in this publication
and apologize to copyright holders if permission to publish in this form has not been obtained. If any
copyright material has not been acknowledged please write and let us know so we may rectify in any
future reprint.
Except as permitted under U.S. Copyright Law, no part of this book may be reprinted, reproduced,
transmitted, or utilized in any form by any electronic, mechanical, or other means, now known or
hereafter invented, including photocopying, microfilming, and recording, or in any information storage or retrieval system, without written permission from the publishers.
For permission to photocopy or use material electronically from this work, please access www.copyright.com (http://www.copyright.com/) or contact the Copyright Clearance Center, Inc. (CCC), 222
Rosewood Drive, Danvers, MA 01923, 978-750-8400. CCC is a not-for-profit organization that provides licenses and registration for a variety of users. For organizations that have been granted a photocopy license by the CCC, a separate system of payment has been arranged.
Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are
used only for identification and explanation without intent to infringe.
Visit the Taylor & Francis Web site at
http://www.taylorandfrancis.com
and the CRC Press Web site at
http://www.crcpress.com
Contents
Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xv
Authors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xix
1
2
Environmental Impact and History of Modern Transportation
1.1 Air Pollution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.1.1 Nitrogen Oxides . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.1.2 Carbon Monoxide . . . . . . . . . . . . . . . . . . . . . . . . . .
1.1.3 Unburned HCs . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.1.4 Other Pollutants . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.2 Global Warming . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.3 Petroleum Resources . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.4 Induced Costs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.5 Importance of Different Transportation
Development Strategies to Future Oil Supply . . . . . . . . . .
1.6 History of EVs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.7 History of HEVs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.8 History of Fuel Cell Vehicles . . . . . . . . . . . . . . . . . . . . . . .
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
1
1
2
2
3
3
3
5
8
.
.
.
.
.
.
.
.
.
.
9
12
14
17
18
Fundamentals of Vehicle Propulsion and Brake . . . . . . . .
2.1 General Description of Vehicle Movement . . . . . . . . .
2.2 Vehicle Resistance . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.2.1 Rolling Resistance . . . . . . . . . . . . . . . . . . . . . .
2.2.2 Aerodynamic Drag . . . . . . . . . . . . . . . . . . . . .
2.2.3 Grading Resistance . . . . . . . . . . . . . . . . . . . . .
2.3 Dynamic Equation . . . . . . . . . . . . . . . . . . . . . . . . . .
2.4 Tire–Ground Adhesion and Maximum Tractive Effort
2.5 Power Train Tractive Effort and Vehicle Speed . . . . . .
2.6 Vehicle Power Plant and Transmission Characteristics
2.6.1 Power Plant Characteristics . . . . . . . . . . . . . . .
2.6.2 Transmission Characteristics . . . . . . . . . . . . . .
2.6.3 Manual Gear Transmission . . . . . . . . . . . . . . .
2.6.3.1 Hydrodynamic Transmission . . . . . . . .
2.6.3.2 Continuously Variable Transmission . .
2.7 Vehicle Performance . . . . . . . . . . . . . . . . . . . . . . . . .
2.7.1 Maximum Speed of a Vehicle . . . . . . . . . . . . . .
2.7.2 Gradeability . . . . . . . . . . . . . . . . . . . . . . . . . .
2.7.3 Acceleration Performance . . . . . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
19
19
20
20
23
24
26
28
30
32
32
35
35
38
42
43
43
44
45
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
v
vi
Contents
2.8 Operating Fuel Economy . . . . . . . . . . . . . . . . . . . . . . . . .
2.8.1 Fuel Economy Characteristics of IC Engines . . . . . .
2.8.2 Computation of Vehicle Fuel Economy . . . . . . . . . .
2.8.3 Basic Techniques to Improve Vehicle Fuel Economy
2.9 Brake Performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.9.1 Braking Force . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.9.2 Braking Distribution on Front and Rear Axles . . . .
2.9.3 Braking Regulation and Braking
Performance Analysis . . . . . . . . . . . . . . . . . . . . . .
2.9.3.1 Braking Regulation . . . . . . . . . . . . . . . . . .
2.9.3.2 Braking Performance Analysis . . . . . . . . . .
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
48
48
49
51
53
53
55
.
.
.
.
.
.
.
.
.
.
.
.
61
61
62
65
Internal Combustion Engines . . . . . . . . . . . . . . . . . . . . . . . . . .
3.1 4S, Spark-Ignited IC Engines . . . . . . . . . . . . . . . . . . . . . . . .
3.1.1 Operating Principles . . . . . . . . . . . . . . . . . . . . . . . . .
3.1.2 Operation Parameters . . . . . . . . . . . . . . . . . . . . . . . .
3.1.2.1 Rating Values of Engines . . . . . . . . . . . . . . . .
3.1.2.2 Indicated Work per Cycles and Mean
Effective Pressure . . . . . . . . . . . . . . . . . . . . .
3.1.2.3 Mechanical Efficiency . . . . . . . . . . . . . . . . . .
3.1.2.4 Specific Fuel Consumption and Efficiency . . .
3.1.2.5 Specific Emissions . . . . . . . . . . . . . . . . . . . . .
3.1.2.6 Fuel/Air and Air/Fuel Ratios . . . . . . . . . . . .
3.1.2.7 Volumetric Efficiency . . . . . . . . . . . . . . . . . . .
3.1.3 Relationships between Operation
and Performance Parameters . . . . . . . . . . . . . . . . . . .
3.1.4 Engine Operation Characteristics . . . . . . . . . . . . . . .
3.1.4.1 Engine Performance Parameters . . . . . . . . . .
3.1.4.2 Indicated and Brake Power and Torque . . . . .
3.1.4.3 Fuel Consumption Characteristics . . . . . . . . .
3.1.5 Design and Operating Variables Affecting
SI Engine Performance, Efficiency, and Emission
Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.1.5.1 Compression Ratio . . . . . . . . . . . . . . . . . . . .
3.1.5.2 Spark Timing . . . . . . . . . . . . . . . . . . . . . . . . .
3.1.5.3 Fuel/Air Equivalent Ratio . . . . . . . . . . . . . . .
3.1.6 Emission Control . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.1.7 Basic Techniques for Improving Engine Performance,
Efficiency, and Emissions . . . . . . . . . . . . . . . . . . . . . .
3.1.7.1 Forced Induction . . . . . . . . . . . . . . . . . . . . . .
3.1.7.2 Gasoline Direct Injection
and Lean-Burn Engines . . . . . . . . . . . . . . . . .
3.1.7.3 Multi- and Variable-Valve Timing . . . . . . . . .
3.1.7.4 Throttle-Less Torque Control . . . . . . . . . . . . .
3.1.7.5 Variable Compression Ratio . . . . . . . . . . . . . .
.
.
.
.
.
67
67
67
69
69
.
.
.
.
.
.
69
71
72
73
73
74
.
.
.
.
.
75
76
76
77
78
.
.
.
.
.
78
79
80
82
84
.
.
85
85
.
.
.
.
86
86
87
87
vii
Contents
3.1.7.6 Exhaust Gas Recirculation
3.1.7.7 Intelligent Ignition . . . . . .
3.1.7.8 New Engine Materials . . .
3.2 4S, Compression-Ignition IC Engines . . .
3.3 2S Engines . . . . . . . . . . . . . . . . . . . . . .
3.4 Wankel Rotary Engines . . . . . . . . . . . . .
3.5 Stirling Engines . . . . . . . . . . . . . . . . . . .
3.6 Gas Turbine Engines . . . . . . . . . . . . . . .
3.7 Quasi-Isothermal Brayton Cycle Engines
References . . . . . . . . . . . . . . . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
. 87
. 87
. 87
. 88
. 89
. 93
. 95
. 100
. 103
. 104
4
Electric Vehicles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.1 Configurations of EVs . . . . . . . . . . . . . . . . . . . . . . .
4.2 Performance of EVs . . . . . . . . . . . . . . . . . . . . . . . . .
4.2.1 Traction Motor Characteristics . . . . . . . . . . . .
4.2.2 Tractive Effort and Transmission Requirement
4.2.3 Vehicle Performance . . . . . . . . . . . . . . . . . . .
4.3 Tractive Effort in Normal Driving . . . . . . . . . . . . . .
4.4 Energy Consumption . . . . . . . . . . . . . . . . . . . . . . .
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
5
Hybrid Electric Vehicles . . . . . . . . . . . . . . . . . . . . . . . . . .
5.1 Concept of Hybrid Electric Drive Trains . . . . . . . . . . . .
5.2 Architectures of Hybrid Electric Drive Trains . . . . . . . .
5.2.1 Series Hybrid Electric Drive Trains
(Electrical Coupling) . . . . . . . . . . . . . . . . . . . . .
5.2.2 Parallel Hybrid Electric Drive Trains
(Mechanical Coupling) . . . . . . . . . . . . . . . . . . .
5.2.2.1 Parallel Hybrid Drive Train with Torque
Coupling . . . . . . . . . . . . . . . . . . . . . . . .
5.2.2.2 Parallel Hybrid Drive Train with Speed
Coupling . . . . . . . . . . . . . . . . . . . . . . . .
5.2.2.3 Hybrid Drive Trains with Both Torque
and Speed Coupling . . . . . . . . . . . . . . .
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
6
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
105
105
108
108
109
112
115
120
122
. . . . . 123
. . . . . 123
. . . . . 126
. . . . . 128
. . . . . 130
. . . . . 132
. . . . . 138
. . . . . 144
. . . . . 149
Electric Propulsion Systems . . . . . . . . . . . . . . . . . . . . . . . . . . .
6.1 DC Motor Drives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
6.1.1 Principle of Operation and Performance . . . . . . . . . .
6.1.2 Combined Armature Voltage and Field Control . . . .
6.1.3 Chopper Control of DC Motors . . . . . . . . . . . . . . . .
6.1.4 Multi-Quadrant Control of Chopper-Fed DC
Motor Drives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
6.1.4.1 Two-Quadrant Control of Forward Motoring
and Regenerative Braking . . . . . . . . . . . . . .
6.1.4.2 Four-Quadrant Operation . . . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
151
154
154
158
158
. . 163
. . 164
. . 167
viii
Contents
6.2 Induction Motor Drives . . . . . . . . . . . . . . . . . . . . . . . . . .
6.2.1 Basic Operation Principles of Induction Motors . . .
6.2.2 Steady-State Performance . . . . . . . . . . . . . . . . . . .
6.2.3 Constant Volt/Hertz Control . . . . . . . . . . . . . . . . .
6.2.4 Power Electronic Control . . . . . . . . . . . . . . . . . . . .
6.2.5 Field Orientation Control . . . . . . . . . . . . . . . . . . . .
6.2.5.1 Field Orientation Principles . . . . . . . . . . . .
6.2.5.2 Control . . . . . . . . . . . . . . . . . . . . . . . . . . .
6.2.5.3 Direction Rotor Flux Orientation Scheme . .
6.2.5.4 Indirect Rotor Flux Orientation Scheme . . .
6.2.6 Voltage Source Inverter for FOC . . . . . . . . . . . . . . .
6.2.6.1 Voltage Control in Voltage Source Inverter .
6.2.6.2 Current Control in Voltage Source Inverter .
6.3 Permanent Magnetic BLDC Motor Drives . . . . . . . . . . . .
6.3.1 Basic Principles of BLDC Motor Drives . . . . . . . . .
6.3.2 BLDC Machine Construction and Classification . . .
6.3.3 Properties of PM Materials . . . . . . . . . . . . . . . . . . .
6.3.3.1 Alnico . . . . . . . . . . . . . . . . . . . . . . . . . . . .
6.3.3.2 Ferrites . . . . . . . . . . . . . . . . . . . . . . . . . . .
6.3.3.3 Rare-Earth PMs . . . . . . . . . . . . . . . . . . . . .
6.3.4 Performance Analysis and Control
of BLDC Machines . . . . . . . . . . . . . . . . . . . . . . . . .
6.3.4.1 Performance Analysis . . . . . . . . . . . . . . . .
6.3.4.2 Control of BLDC Motor Drives . . . . . . . . .
6.3.5 Extend Speed Technology . . . . . . . . . . . . . . . . . . .
6.3.6 Sensorless Techniques . . . . . . . . . . . . . . . . . . . . . .
6.3.6.1 Methods Using Measurables and Math . . .
6.3.6.2 Methods Using Observers . . . . . . . . . . . . .
6.3.6.3 Methods Using Back EMF Sensing . . . . . . .
6.3.6.4 Unique Sensorless Techniques . . . . . . . . . .
6.4 SRM Drives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
6.4.1 Basic Magnetic Structure . . . . . . . . . . . . . . . . . . . .
6.4.2 Torque Production . . . . . . . . . . . . . . . . . . . . . . . . .
6.4.3 SRM Drive Converter . . . . . . . . . . . . . . . . . . . . . .
6.4.4 Modes of Operation . . . . . . . . . . . . . . . . . . . . . . . .
6.4.5 Generating Mode of Operation
(Regenerative Braking) . . . . . . . . . . . . . . . . . . . . .
6.4.6 Sensorless Control . . . . . . . . . . . . . . . . . . . . . . . . .
6.4.6.1 Phase Flux Linkage-Based Method . . . . . .
6.4.6.2 Phase Inductance-Based Method . . . . . . . .
6.4.6.3 Modulated Signal Injection Methods . . . . .
6.4.6.4 Mutual-Induced Voltage-Based Method . . .
6.4.6.5 Observer-Based Methods . . . . . . . . . . . . .
6.4.7 Self-Tuning Techniques of SRM Drives . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
168
169
172
174
176
179
179
187
189
192
193
195
198
200
203
203
205
206
208
208
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
208
209
211
213
213
214
215
215
216
217
218
222
224
226
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
227
230
231
232
233
236
236
236
ix
Contents
6.4.7.1 Self-Tuning with Arithmetic Method
6.4.7.2 Self-Tuning Using an ANN . . . . . . .
6.4.8 Vibration and Acoustic Noise in SRM . . . . . .
6.4.9 SRM Design . . . . . . . . . . . . . . . . . . . . . . . .
6.4.9.1 Number of Stator and Rotor Poles . .
6.4.9.2 Stator Outer Diameter . . . . . . . . . . .
6.4.9.3 Rotor Outer Diameter . . . . . . . . . . .
6.4.9.4 Air Gap . . . . . . . . . . . . . . . . . . . . . .
6.4.9.5 Stator Arc . . . . . . . . . . . . . . . . . . . .
6.4.9.6 Stator Back Iron . . . . . . . . . . . . . . . .
6.4.9.7 Performance Prediction . . . . . . . . . .
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
7
8
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
237
238
240
243
243
244
244
245
245
245
246
247
Design Principle of Series (Electrical Coupling)
Hybrid Electric Drive Train . . . . . . . . . . . . . . . . . . . . . . .
7.1 Operation Patterns . . . . . . . . . . . . . . . . . . . . . . . . . .
7.2 Control Strategies . . . . . . . . . . . . . . . . . . . . . . . . . . .
7.2.1 Max. SOC-of-PPS Control Strategy . . . . . . . . .
7.2.2 Engine On–Off or Thermostat Control Strategy
7.3 Design Principles of a Series (Electrical Coupling)
Hybrid Drive Train . . . . . . . . . . . . . . . . . . . . . . . . . .
7.3.1 Electrical Coupling Device . . . . . . . . . . . . . . . .
7.3.2 Power Rating Design of the Traction Motor . . .
7.3.3 Power Rating Design of the Engine/Generator
7.3.4 Design of PPS . . . . . . . . . . . . . . . . . . . . . . . . .
7.3.4.1 Power Capacity of PPS . . . . . . . . . . . .
7.3.4.2 Energy Capacity of PPS . . . . . . . . . . . .
7.4 Design Example . . . . . . . . . . . . . . . . . . . . . . . . . . . .
7.4.1 Design of Traction Motor Size . . . . . . . . . . . . .
7.4.2 Design of the Gear Ratio . . . . . . . . . . . . . . . . .
7.4.3 Verification of Acceleration Performance . . . . .
7.4.4 Verification of Gradeability . . . . . . . . . . . . . . .
7.4.5 Design of Engine/Generator Size . . . . . . . . . .
7.4.6 Design of the Power Capacity of PPS . . . . . . . .
7.4.7 Design of the Energy Capacity of PPS . . . . . . .
7.4.8 Fuel Consumption . . . . . . . . . . . . . . . . . . . . .
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
253
254
256
256
257
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
259
259
264
267
270
271
271
272
272
272
273
274
275
277
277
279
279
Parallel (Mechanically Coupled) Hybrid
Electric Drive Train Design . . . . . . . . . . . . . . . . . . . . . .
8.1 Drive Train Configuration and Design Objectives . . .
8.2 Control Strategies . . . . . . . . . . . . . . . . . . . . . . . . . .
8.2.1 Max. SOC-of-PPS Control Strategy . . . . . . . . .
8.2.2 Engine On–Off (Thermostat) Control Strategy
8.2.3 Constrained Engine On–Off Control Strategy .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
281
281
283
284
287
288
.
.
.
.
.
.
x
Contents
8.2.4 Fuzzy Logic Control Technique . . .
8.2.5 Dynamic Programming Technique .
8.3 Parametric Design of a Drive Train . . . . .
8.3.1 Engine Power Design . . . . . . . . . .
8.3.2 Transmission Design . . . . . . . . . . .
8.3.3 Electric Motor Drive Power Design
8.3.4 PPS Design . . . . . . . . . . . . . . . . . .
8.4 Simulations . . . . . . . . . . . . . . . . . . . . . . .
References . . . . . . . . . . . . . . . . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
290
292
295
295
298
299
302
305
307
Design and Control Methodology of Series–Parallel
(Torque and Speed Coupling) Hybrid Drive Train . .
9.1 Drive Train Configuration . . . . . . . . . . . . . . . . .
9.1.1 Speed-Coupling Analysis . . . . . . . . . . . . .
9.1.2 Drive Train Configuration . . . . . . . . . . . .
9.2 Drive Train Control Methodology . . . . . . . . . . .
9.2.1 Control System . . . . . . . . . . . . . . . . . . . .
9.2.2 Engine Speed Control Approach . . . . . . .
9.2.3 Traction Torque Control Approach . . . . . .
9.2.4 Drive Train Control Strategies . . . . . . . . .
9.2.4.1 Engine Speed Control Strategy . . .
9.2.4.2 Traction Torque Control Strategy .
9.2.4.3 Regenerative Braking Control . . .
9.3 Drive Train Parameters Design . . . . . . . . . . . . .
9.4 Simulation of an Example Vehicle . . . . . . . . . . .
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
309
309
309
313
320
320
320
321
323
323
325
328
328
329
332
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
333
333
335
335
341
346
351
11 Mild Hybrid Electric Drive Train Design . . . . . . . . .
11.1 Energy Consumed in Braking and Transmission
11.2 Parallel Mild Hybrid Electric Drive Train . . . . .
11.2.1 Configuration . . . . . . . . . . . . . . . . . . . .
11.2.2 Operating Modes and Control Strategy .
11.2.3 Drive Train Design . . . . . . . . . . . . . . . .
11.2.4 Performance . . . . . . . . . . . . . . . . . . . . .
11.3 Series–Parallel Mild Hybrid Electric Drive Train
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
353
353
355
355
355
356
360
365
9
10 Design and Control Principles of Plug-In
Hybrid Electric Vehicles . . . . . . . . . . . . . .
10.1 Statistics of Daily Driving Distance . . .
10.2 Energy Management Strategy . . . . . . .
10.2.1 AER-Focused Control Strategy
10.2.2 Blended Control Strategy . . . .
10.3 Energy Storage Design . . . . . . . . . . . .
References . . . . . . . . . . . . . . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
xi
Contents
11.3.1 Configuration of the Drive Train
with a Planetary Gear Unit . . . . . . . . . . . .
11.3.2 Operating Modes and Control . . . . . . . . . .
11.3.2.1 Speed-Coupling Operating Mode
11.3.2.2 Torque-Coupling Operating Mode
11.3.2.3 Engine-Alone Traction Mode . . . .
11.3.2.4 Motor-Alone Traction Mode . . . .
11.3.2.5 Regenerative Braking Mode . . . . .
11.3.2.6 Engine Starting . . . . . . . . . . . . . .
11.3.3 Control Strategy . . . . . . . . . . . . . . . . . . . .
11.3.4 Drive Train with a Floating-Stator Motor . .
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
365
367
367
368
369
369
370
370
370
371
372
12 Peaking Power Sources and Energy Storages . . . . . . . . . . . .
12.1 Electrochemical Batteries . . . . . . . . . . . . . . . . . . . . . . .
12.1.1 Electrochemical Reactions . . . . . . . . . . . . . . . . .
12.1.2 Thermodynamic Voltage . . . . . . . . . . . . . . . . . .
12.1.3 Specific Energy . . . . . . . . . . . . . . . . . . . . . . . . .
12.1.4 Specific Power . . . . . . . . . . . . . . . . . . . . . . . . . .
12.1.5 Energy Efficiency . . . . . . . . . . . . . . . . . . . . . . .
12.1.6 Battery Technologies . . . . . . . . . . . . . . . . . . . . .
12.1.6.1 Lead–Acid Battery . . . . . . . . . . . . . . . .
12.1.6.2 Nickel-Based Batteries . . . . . . . . . . . . .
12.1.6.3 Lithium-Based Batteries . . . . . . . . . . . .
12.2 Ultracapacitors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
12.2.1 Features of Ultracapacitors . . . . . . . . . . . . . . . .
12.2.2 Basic Principles of Ultracapacitors . . . . . . . . . . .
12.2.3 Performance of Ultracapacitors . . . . . . . . . . . . .
12.2.4 Ultracapacitor Technologies . . . . . . . . . . . . . . . .
12.3 Ultra-High-Speed Flywheels . . . . . . . . . . . . . . . . . . . .
12.3.1 Operation Principles of Flywheels . . . . . . . . . . .
12.3.2 Power Capacity of Flywheel Systems . . . . . . . . .
12.3.3 Flywheel Technologies . . . . . . . . . . . . . . . . . . .
12.4 Hybridization of Energy Storages . . . . . . . . . . . . . . . . .
12.4.1 Concept of Hybrid Energy Storage . . . . . . . . . .
12.4.2 Passive and Active Hybrid Energy Storage with
Battery and Ultracapacitor . . . . . . . . . . . . . . . . .
12.4.3 Battery and Ultracapacitor Size Design . . . . . . .
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
375
375
378
379
380
382
384
385
385
386
388
390
390
391
392
396
397
397
400
402
404
404
. . . . 404
. . . . 406
. . . . 410
13 Fundamentals of Regenerative Breaking . . . . . . .
13.1 Braking Energy Consumed in Urban Driving
13.2 Braking Energy versus Vehicle Speed . . . . . .
13.3 Braking Energy versus Braking Power . . . . .
13.4 Braking Power versus Vehicle Speed . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
411
411
413
416
416
xii
Contents
13.5 Braking Energy versus Vehicle Deceleration Rate . . . . . . . . .
13.6 Braking Energy on Front and Rear Axles . . . . . . . . . . . . . . . .
13.7 Brake System of EV, HEV, and FCV . . . . . . . . . . . . . . . . . . . .
13.7.1 Parallel Hybrid Braking System . . . . . . . . . . . . . . . . .
13.7.1.1 Design and Control Principles with Fixed
Ratios between Electric and Mechanical
Braking Forces . . . . . . . . . . . . . . . . . . . . . . .
13.7.1.2 Design and Control Principles for Maximum
Regenerative Braking . . . . . . . . . . . . . . . . .
13.7.2 Fully Controllable Hybrid Brake System . . . . . . . . . .
13.7.2.1 Control Strategy for Optimal Braking
Performance . . . . . . . . . . . . . . . . . . . . . . . .
13.7.2.2 Control Strategy for Optimal Energy
Recovery . . . . . . . . . . . . . . . . . . . . . . . . . . .
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
14 Fuel Cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
14.1 Operating Principles of Fuel Cells . . . . . . . . . . .
14.2 Electrode Potential and Current–Voltage Curve
14.3 Fuel and Oxidant Consumption . . . . . . . . . . . .
14.4 Fuel Cell System Characteristics . . . . . . . . . . . .
14.5 Fuel Cell Technologies . . . . . . . . . . . . . . . . . . .
14.5.1 Proton Exchange Membrane Fuel Cells .
14.5.2 Alkaline Fuel Cells . . . . . . . . . . . . . . . .
14.5.3 Phosphoric Acid Fuel Cells . . . . . . . . . .
14.5.4 Molten Carbonate Fuel Cells . . . . . . . . .
14.5.5 Solid Oxide Fuel Cells . . . . . . . . . . . . . .
14.5.6 Direct Methanol Fuel Cells . . . . . . . . . .
14.6 Fuel Supply . . . . . . . . . . . . . . . . . . . . . . . . . . .
14.6.1 Hydrogen Storage . . . . . . . . . . . . . . . . .
14.6.1.1 Compressed Hydrogen . . . . . .
14.6.1.2 Cryogenic Liquid Hydrogen . .
14.6.1.3 Metal Hydrides . . . . . . . . . . . .
14.6.2 Hydrogen Production . . . . . . . . . . . . . .
14.6.2.1 Steam Reforming . . . . . . . . . . .
14.6.2.2 POX Reforming . . . . . . . . . . . .
14.6.2.3 Autothermal Reforming . . . . . .
14.6.3 Ammonia as Hydrogen Carrier . . . . . . .
14.7 Non-Hydrogen Fuel Cells . . . . . . . . . . . . . . . .
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
15 Fuel Cell Hybrid Electric Drive Train Design
15.1 Configuration . . . . . . . . . . . . . . . . . . . .
15.2 Control Strategy . . . . . . . . . . . . . . . . . . .
15.3 Parametric Design . . . . . . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
417
419
420
420
420
422
426
427
429
431
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
433
433
437
440
441
443
443
444
446
447
448
449
450
450
450
452
453
454
454
455
456
457
457
458
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
459
459
461
463
xiii
Contents
15.3.1 Motor Power Design . . . . . . . . . . . . . . . .
15.3.2 Power Design of the Fuel Cell System . . .
15.3.3 Design of the Power and Energy Capacity
of the PPS . . . . . . . . . . . . . . . . . . . . . . . .
15.3.3.1 Power Capacity of the PPS . . . . .
15.3.3.2 Energy Capacity of the PPS . . . .
15.4 Design Example . . . . . . . . . . . . . . . . . . . . . . . . .
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . 463
. . . . . . . . . 464
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
465
465
465
466
469
16 Design of Series Hybrid Drive Train for Off-Road Vehicles . .
16.1 Motion Resistance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
16.1.1 Motion Resistance Caused by Terrain Compaction .
16.1.2 Motion Resistance Caused by Terrain Bulldozing . .
16.1.3 Internal Resistance of the Running Gear . . . . . . . .
16.1.4 Tractive Effort of a Terrain . . . . . . . . . . . . . . . . . . .
16.1.5 Drawbar Pull . . . . . . . . . . . . . . . . . . . . . . . . . . . .
16.2 Tracked Series Hybrid Vehicle Drive Train Architecture . . .
16.3 Parametric Design of the Drive Train . . . . . . . . . . . . . . . .
16.3.1 Traction Motor Power Design . . . . . . . . . . . . . . . .
16.3.1.1 Vehicle Thrust versus Speed . . . . . . . . . .
16.3.1.2 Motor Power and Acceleration
Performance . . . . . . . . . . . . . . . . . . . . . .
16.3.1.3 Motor Power and Gradeability . . . . . . . .
16.3.1.4 Steering Maneuver of a Tracked Vehicle . .
16.4 Engine/Generator Power Design . . . . . . . . . . . . . . . . . . .
16.5 Power and Energy Design of Energy Storage . . . . . . . . . .
16.5.1 Peaking Power for Traction . . . . . . . . . . . . . . . . . .
16.5.2 Peaking Power for Nontraction . . . . . . . . . . . . . . .
16.5.3 Energy Design of Batteries/Ultracapacitors . . . . . .
16.5.4 Combination of Batteries and Ultracapacitors . . . .
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
471
471
472
475
476
476
477
478
479
480
480
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
481
482
485
489
490
491
491
494
494
496
Appendix Technical Overview of Toyota Prius . . . . . . . . . . . . . .
A.1 Vehicle Performance . . . . . . . . . . . . . . . . . . . . . . .
A.2 Overview of Prius Hybrid Power Train
and Control Systems . . . . . . . . . . . . . . . . . . . . . . .
A.3 Major Components . . . . . . . . . . . . . . . . . . . . . . . .
A.3.1 Engine . . . . . . . . . . . . . . . . . . . . . . . . . . . .
A.3.2 Hybrid Transaxle . . . . . . . . . . . . . . . . . . .
A.3.3 HV Battery . . . . . . . . . . . . . . . . . . . . . . . .
A.3.4 Inverter Assembly . . . . . . . . . . . . . . . . . . .
A.3.4.1 Booster Converter (2004 and Later)
A.3.4.2 Inverter . . . . . . . . . . . . . . . . . . . .
A.3.4.3 DC–DC Converter . . . . . . . . . . . .
A.3.4.4 AC Inverter . . . . . . . . . . . . . . . . .
. . 499
. . 499
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
499
501
501
501
502
506
506
506
507
507
xiv
Contents
A.3.5 Brake System . . . . . . . . . . . . . . . . . . . . . . . .
A.3.5.1 Regenerative Brake Cooperative
Control . . . . . . . . . . . . . . . . . . . . . . .
A.3.5.2 Electronic Brake Distribution Control
(2004 and Later Models) . . . . . . . . . .
A.3.5.3 Brake Assist System (2004 and Later
Models) . . . . . . . . . . . . . . . . . . . . . .
A.3.6 Electric Power Steering . . . . . . . . . . . . . . . . .
A.3.7 Enhanced Vehicle Stability Control (VSC)
System (2004 and Later Prius) . . . . . . . . . . . .
A.4 Hybrid System Control Modes . . . . . . . . . . . . . . . . .
507
509
509
510
510
512
512
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 519
Preface
The development of internal combustion engine automobiles is one of the
greatest achievements of modern technology. However, the highly developed
automotive industry and the increasingly large number of automobiles in
use around the world are causing serious problems for the environment and
hydrocarbon resources. The deteriorating air quality, global warming issues,
and depleting petroleum resources are becoming serious threats to modern
life. Progressively more rigorous emissions and fuel efficiency standards are
stimulating the aggressive development of safer, cleaner, and more efficient
vehicles. It is now well recognized that electric, hybrid electric, and fuel-cellpowered drive train technologies are the most promising vehicle solutions
for the foreseeable future.
To meet this challenge, an increasing number of engineering schools, in the
United States and around the world, have initiated academic programs in
advanced energy and vehicle technologies at the undergraduate and graduate levels. We started our first graduate course, in 1998, on “Advanced
Vehicle Technologies—Design Methodology of Electric and Hybrid Electric
Vehicles” for students in mechanical and electrical engineering at Texas
A&M University. While preparing the lectures for this course, we found that
although there is a wealth of information in the form of technical papers
and reports, there was no rigorous and comprehensive textbook for students
and professors who may wish to offer such a course. Furthermore, practicing engineers also needed a systematic reference book to fully understand
the essentials of this new technology. The first edition of this book was our
attempt to fill this need. The second edition introduces newer topics and
deeper treatments than the first edition.
The book deals with the fundamentals, theoretical bases, and design
methodologies of conventional internal combustion engine (ICE) vehicles,
electric vehicles (EVs), hybrid electric vehicles (HEVs), and fuel cell vehicles (FCVs). It comprehensively covers vehicle performance characteristics,
configurations, control strategies, design methodologies, modeling, and
simulations for modern vehicles with mathematical rigor. It includes drive
train architecture analysis, ICE-based drive trains, EV and HEV configurations, electric propulsion systems, series/parallel/mild hybrid electric drive
train design methodologies, energy storage systems, regenerative braking,
fuel cells and their applications in vehicles, and fuel cell hybrid electric
drive train design. The book’s perspective is from the overall drive train
system and not just individual components. The design methodology is
xv
xvi
Preface
described in mathematical terms, step by step. Furthermore, in explaining
the design methodology of each drive train, design examples are presented
with simulation results.
More specifically, the second edition contains many corrections and
updates of the material in the first edition. Three new chapters and one
appendix have been added. They are Chapter 9: Design and Control Methodology of Series–Parallel (Torque and Speed Coupling) Hybrid Drive Train;
Chapter 10: Design and Control Principles of Plug-In Hybrid Electric Vehicles; Chapter 16: Design of Series Hybrid Drive Train for Off-Road Vehicles,
and Appendix: Technical Overview of Toyota Prius. Chapter 13: Fundamentals of Regenerative Braking has been completely rewritten, based on our
new research. In addition, plenty of new materials have been added to the
old chapters. All these new contributions to the second edition make it more
complete and useful to the reader.
This book consists of 16 chapters and one appendix. In Chapter 1, the social
and environmental importances of modern transportation is discussed. This
mainly includes the air pollution, global warming, and petroleum resource
depletion issues associated with the development of modern transportation.
In this chapter, the impact of future vehicle technologies on oil supplies is
analyzed. The results are helpful for the development strategies of the next
generation of vehicles. In addition, the development history of EVs, HEVs,
and FCVs is briefly reviewed.
In Chapter 2, basic understandings of vehicle performance, power plant
characteristics, transmission characteristics, and the equations used to
describe vehicle performance are introduced. The main purpose of this chapter is to provide the basic knowledge that is necessary for vehicle drive train
design. As an improvement to the first edition, material on the brake system
and its design and performance has been strengthened in order to provide a
more solid base for the hybrid brake system designs in EVs, HEVs, and FCVs.
In Chapter 3, major operating characteristics of different heat engines are
introduced. As the primary power plant, the engine is the most important
subsystem in conventional and hybrid drive train systems. Full understanding of the characteristics of engine is necessary for the design and control of
conventional as well as HEVs.
In Chapter 4, EVs are introduced. This chapter mainly includes the design
of the electric propulsion system and its energy storage device, the design of
the traction motor and its transmission, methodology of prediction of vehicle
performance, and system simulation results.
In Chapter 5, the basic concept of hybrid traction is established first. Then,
various configurations of HEVs are discussed. These include series hybrid,
parallel hybrid, torque-coupling and speed-coupling hybrids, and other
configurations. The main operating characteristics of these configurations are
also presented.
In Chapter 6, several electric power plants are introduced. These include
DC, AC, permanent magnet brushless DC, and switched reluctance motor
Preface
xvii
drives. Their basic structure, operating principles, control and operational
characteristics are described from a traction system point of view.
In Chapter 7, the design methodology of series hybrid electric drive trains
is presented. This chapter focuses on the system-oriented design of the engine and the energy storage, the traction motor, the transmission, the control
strategy, and the power converters. A design example is also provided. As
an improvement to the first edition, various power converter configurations
have been added.
In Chapter 8, a design methodology of parallel hybrid electric drive trains is
provided. This chapter includes driving patterns and driving mode analysis;
control strategy; design of the major components, for example, the engine, the
energy storage, and the transmission; and vehicle performance simulation. In
addition to the material covered in the first edition, a constrained engine on
and off control strategy, fuzzy logic control strategy, and the concept of control
optimization based on dynamic programming have been added.
In Chapter 9, the operating characteristics, design methodology, and control
strategies of a series–parallel hybrid drive train are presented. This is a new
chapter in the second edition.
In Chapter 10, the design and control principles of the plug-in hybrid vehicle
are introduced. This chapter mainly addresses the charge sustaining hybrid
drive train with regard to the drive train control strategy, energy storage
design, and electric motor design. This is also a new chapter.
In Chapter 11, a design methodology of mild hybrid drive trains is
introduced with two major configurations of parallel torque coupling and
series–parallel, torque–speed coupling. This chapter focuses on operational
analysis, control system development, and system simulation.
In Chapter 12, different energy storage technologies are introduced,
including batteries, ultracapacitors, and flywheels. The discussion focuses
on power and energy capacities. The concept of hybrid energy storage is also
introduced in this chapter.
In Chapter 13, the design and control principles of hybrid brake systems
are introduced. Brake safety and recoverable energy are the main concerns.
The available braking energy characteristics, with regard to vehicle speed,
and the braking power in typical driving cycles are investigated. The brake
force distribution on the front and rear wheels is discussed for guaranteeing
the vehicle braking performance for safety. Furthermore, this chapter discusses the important issue of distributing the total braking force between
the mechanical and the electrical regenerative brakes. Two advanced hybrid
brake systems, including their control strategies, are introduced. This chapter
has been rewritten based on our recent research.
In Chapter 14, different fuel cell systems are described, with a focus on their
operating principles and characteristics, various technologies, and their fuels.
Specifically, vehicle applications of fuel cells are explained.
In Chapter 15, a systematic design of fuel cell hybrid drive trains is introduced. First, the concept of fuel cell hybrid vehicles is established. Then, their
xviii
Preface
operating principles and drive train control systems are analyzed. Lastly, a
design methodology is provided, focusing on the system designs of the fuel
cell, the electric propulsion system, and the energy storage system. A design
example and its corresponding performance simulation results are provided.
In Chapter 16, a design methodology of an off-road tracked series hybrid
vehicle is developed. The discussion focuses on the motion resistance calculation on soft grounds, traction motor system design, the engine/generator
system design, and the peaking power source system design. This is a new
chapter for the second edition.
A case study appendix has been added to the second edition. This is an
overview of the Toyota Prius hybrid system. The purpose is to give the reader
a practical example of the architecture, operational modes, control system,
among other things, of a commercial hybrid electric drive train.
This book is suitable for a graduate or senior-level undergraduate course in
advanced vehicles. Depending on the backgrounds of the students in different
disciplines such as mechanical or electrical engineering, course instructors
have the flexibility of choosing the specialized material to suit their lectures.
This text has been used at Texas A&M University in a graduate-level course
for many years. The manuscript of this text has been revised many times and
over many years, based on the comments and feedback from the students in
our course. We are grateful to our students for their help.
This book is also an in-depth resource and a comprehensive reference in
modern automotive systems for engineers, students, researchers, and other
professionals who are working in automotive-related industries, as well as
those in government and academia.
In addition to the work by others, many of the technologies and advances
presented in this book are the collection of many years of research and
development by the authors and other members of the Advanced Vehicle
Systems Research Program at Texas A&M University. We are grateful to all
the dedicated staff of the Advanced Vehicle Systems Research group and the
Power Electronics and Motor Drives group at Texas A&M, who made great
contributions to this book.
We would also like to express our sincere thanks to Mr. Glenn C. Krell,
whose proofreading and corrections have improved this text. In addition, we
would like to acknowledge the efforts and assistance of the staff of CRC Press,
LLC, especially Ms. Nora Konopka. Last but not least, we thank our families
for their patience and support during the long effort in the writing of this
book.
Mehrdad Ehsani
Yimin Gao
Ali Emadi
Authors
Mehrdad Ehsani received his BS and MS from
the University of Texas at Austin in 1973 and
1974, respectively, and his PhD from the University of Wisconsin–Madison in 1981, all in electrical
engineering.
From 1974 to 1977 he was with the Fusion
Research Center, University of Texas, as a research
engineer. From 1977 to 1981 he was with the
Argonne National Laboratory, Argonne, Illinois,
as a resident research associate, while simultaneously doing the doctoral work at the University
of Wisconsin–Madison in energy systems and
control systems. Since 1981 Dr. Ehsani has been
at Texas A&M University, College Station, where
he is now a professor of electrical engineering and
director of the Advanced Vehicle Systems Research Program and the Power
Electronics and Motor Drives Laboratory. He is the recipient of the Prize
Paper Awards in Static Power Converters and motor drives at the IEEEIndustry Applications Society 1985, 1987, and 1992 annual meetings, as well as
numerous other honors and recognitions. In 1984, Dr. Ehsani was named the
Outstanding Young Engineer of the Year by the Brazos chapter of the Texas
Society of Professional Engineers. In 1992, he was named the Halliburton
Professor in the College of Engineering at Texas A&M. In 1994, he was also
named the Dresser Industries Professor in the same college. In 2001, he was
selected as the Ruth & William Neely/Dow Chemical Faculty Fellow of the
College of Engineering for 2001–2002, for “contributions to the Engineering
Program at Texas A&M, including classroom instruction, scholarly activities, and professional service.” In 2003, he received the BP Amoco Faculty
Award for Teaching Excellence in the College of Engineering. He was awarded
the IEEE Vehicular Society 2001 Avant Garde Award for “contributions to
the theory and design of hybrid electric vehicles.” In 2003, Dr. Ehsani was
awarded the IEEE Undergraduate Teaching Award “for outstanding contributions to advanced curriculum development and teaching of power electronics
and drives.” In 2004, he was elected to the Robert M. Kennedy endowed Chair
in Electrical Engineering at Texas A&M University. In 2005, he was elected as
the Fellow of Society of Automotive Engineers. Dr. Ehsani is the author of over
300 publications in pulsed-power supplies, high-voltage engineering, power
xix
xx
Authors
electronics, motor drives, and advanced vehicle systems, and is the coauthor
of 12 books on power electronics, motor drives, and advanced vehicle systems, including Vehicular Electric Power Systems, Marcel Dekker, Inc. 2003 and
Modern Electric Hybrid Vehicles and Fuel Cell Vehicles—Fundamentals, Theory,
and Design, CRC Press, 2004. He has over 23 granted or pending U.S. and
EC patents. His current research work is in power electronics, motor drives,
hybrid vehicles and their control systems.
Dr. Ehsani has been a member of the IEEE Power Electronics Society (PELS)
AdCom, past chairman of the PELS Educational Affairs Committee, past
chairman of the IEEE-IAS Industrial Power Converter Committee, and past
chairman of the IEEE Myron Zucker Student–Faculty Grant program. He was
the general chair of the IEEE Power Electronics Specialist Conference for 1990.
He is the founder of the IEEE Power and Propulsion Conference, the founding chairman of the IEEE VTS Vehicle Power and Propulsion Committee,
and chairman of the Convergence Fellowship Committees. In 2002, he was
elected to the board of governors of VTS. He also serves on the editorial board
of several technical journals and is the associate editor of IEEE Transactions
on Industrial Electronics and IEEE Transactions on Vehicular Technology. He is
a fellow of IEEE, an IEEE Industrial Electronics Society and Vehicular Technology Society Distinguished Speaker, and an IEEE Industry Applications
Society and Power Engineering Society Distinguished Lecturer. He is also a
registered professional engineer in the state of Texas.
Yimin Gao received his BS, MS, and PhD in
mechanical engineering (major in development,
design, and manufacturing of automotive systems) in 1982, 1986, and 1991, respectively, all from
Jilin University of Technology, Changchun, Jilin,
China. From 1982 to 1983, he worked as a vehicle
design engineer for the DongFeng Motor Company, Shiyan, Hubei, China. He finished a layout
design of a 5-ton truck (EQ144) and participated
in prototyping and testing. From 1983 to 1986, he
was a graduate student in the Automotive Engineering College of Jilin University of Technology,
Changchun, Jilin, China. His working field was the improvement of vehicle
fuel economy by optimal matching of engine and transmission.
From 1987 to 1992, he was a PhD student in the Automotive Engineering
College of Jilin University of Technology, Changchun, Jilin, China. During this
period, he worked on research and development of legged vehicles, which
can potentially operate in harsh environments, where mobility is difficult for
wheeled vehicles. From 1991 to 1995, Dr. Gao was an associate professor and
automotive design engineer in the Automotive Engineering College of Jilin
Authors
xxi
University of Technology. During this period, he taught undergraduate students in a course entitled Automotive Theory and Design for several semesters
and graduate students in a course entitled Automotive Experiment Technique for
two semesters. Meanwhile, he also conducted vehicle performance, chassis,
and components analyses, and conducted automotive design including chassis design, power train design, suspension design, steering system design,
and brake design.
Dr. Gao joined the Advanced Vehicle Systems Research Program at Texas
A&M University in 1995 as a research associate. Since then, he has been
working in this program on research and development of electric and hybrid
electric vehicles. His research areas are mainly on the fundamentals, architecture, control, modeling, and design of electric and hybrid electric drive
trains and major components. He is a member of the Society of Automotive
Engineers.
Ali Emadi received his BS and MS in
Electrical Engineering with highest distinction from Sharif University of Technology,
Tehran, Iran. He also received his PhD
in Electrical Engineering from Texas A&M
University, College Station, Texas. He is currently the Harris Perlstein Endowed chair
professor of Electrical Engineering and the
director of the Electric Power and Power
Electronics Center and Grainger Laboratories at Illinois Institute of Technology (IIT)
in Chicago, where he has established research and teaching facilities as well
as courses in power electronics, motor drives, and vehicular power systems.
In addition, Dr. Emadi is the founder, president, and chief technology officer
of Hybrid Electric Vehicle Technologies, Inc. (HEVT)—a university spin-off
company of IIT.
Dr. Emadi is the recipient of numerous awards and recognitions. He has
been named a Chicago Matters Global Visionary in 2009. He was named the
Eta Kappa Nu Outstanding Young Electrical Engineer of the Year 2003 (single
international award) by virtue of his outstanding contributions to hybrid
electric vehicle conversion by the Electrical Engineering Honor Society. He
also received the 2005 Richard M. Bass Outstanding Young Power Electronics
Engineer Award from the IEEE Power Electronics Society. In 2005, he was
selected as the Best Professor of the Year by the students at IIT. Dr. Emadi is
the recipient of the 2002 University Excellence in Teaching Award from IIT as
well as the 2004 Sigma Xi/IIT Award for Excellence in University Research.
He directed a team of students to design and build a novel motor drive, which
won the First Place Overall Award of the 2003 IEEE/DOE/DOD International
Future Energy Challenge for Motor Competition.
xxii
Authors
Dr. Emadi is the principal author and coauthor of over 250 journals and
conference papers as well as several books including Vehicular Electric Power
Systems: Land, Sea, Air, and Space Vehicles, Marcel Dekker, 2003; Energy Efficient
Electric Motors, Marcel Dekker, 2004; Uninterruptible Power Supplies and Active
Filters, CRC Press, 2004; Modern Electric, Hybrid Electric, and Fuel Cell Vehicles: Fundamentals, Theory, and Design, CRC Press, 2004; and Integrated Power
Electronic Converters and Digital Control, CRC Press, 2009. Dr. Emadi is also
the editor of the Handbook of Automotive Power Electronics and Motor Drives,
Marcel Dekker, 2005.
Dr. Emadi was the founding general chair of the 1st IEEE Vehicle Power
and Propulsion Conference (VPPC’05), which was colocated under his
chairmanship with the SAE International Future Transportation Technology
Conference. He is currently the chair of the IEEE Vehicle Power and Propulsion Steering Committee, chair of the Technical Committee on Transportation
Power Electronics of the IEEE Power Electronics Society, and Chair of the
Power Electronics Technical Committee of the IEEE Industrial Electronics
Society. He has also served as the Chair of the 2007 IEEE International Future
Energy Challenge.
Dr. Emadi is the editor (North America) of the International Journal of Electric
and Hybrid Vehicles. He has been the guest editor-in-chief of the Special Issue on
Automotive Power Electronics and Motor Drives, IEEE Transactions on Power
Electronics. He has also been the guest editor of the Special Section on Hybrid
Electric and Fuel Cell Vehicles, IEEE Transactions on Vehicular Technology and
guest editor of the Special Section on Automotive Electronics and Electrical
Drives, IEEE Transactions on Industrial Electronics. He has served as an associate
editor of the IEEE Transactions on Vehicular Technology, IEEE Transactions on
Power Electronics, and IEEE Transactions on Industrial Electronics.
1
Environmental Impact and History of
Modern Transportation
The development of internal combustion (IC) engine vehicles, and especially
automobiles, is one of the greatest achievements of modern technology. Automobiles have made great contributions to the growth of modern society by
satisfying many of the needs for mobility in everyday life. The rapid development of the automotive industry, unlike that of any other industry, has
prompted the progress of human beings from a primitive security to a highly
developed industrial one. The automobile industry and the other industries
that serve it constitute the backbone of the world’s economy and employ the
greatest share of the working population.
However, the large number of automobiles in use around the world has
caused and continues to cause serious problems for environment and human
life. Air pollution, global warming, and the rapid depletion of the Earth’s
petroleum resources are now problems of paramount concern.
In recent decades, the research and development activities related to transportation have emphasized the development of high-efficiency, clean, and
safe transportation. Electric vehicles (EVs), hybrid electric vehicles (HEVs),
and fuel cell vehicles have been typically proposed to replace conventional
vehicles in the near future.
This chapter reviews the problems of air pollution, gas emissions causing
global warming, and petroleum resource depletion. It also gives a brief review
of the history of EVs, HEVs, and fuel cell technology.
1.1 Air Pollution
At present, all vehicles rely on the combustion of hydrocarbon (HC) fuels
to derive the energy necessary for their propulsion. Combustion is a reaction between the fuel and the air that releases heat and combustion products.
The heat is converted to mechanical power by an engine and the combustion
1
2
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
products are released to the atmosphere. An HC is a chemical compound with
molecules made up of carbon and hydrogen atoms. Ideally, the combustion
of an HC yields only carbon dioxide and water, which do not harm the
environment. Indeed, green plants “digest” carbon dioxide by photosynthesis. Carbon dioxide is a necessary ingredient in vegetal life. Animals do not
suffer from breathing carbon dioxide unless its concentration in air is such
that oxygen is almost absent.
Actually, the combustion of HC fuel in combustion engines is never ideal.
Besides carbon dioxide and water, the combustion products contain a certain
amount of nitrogen oxides (NOx ), carbon monoxides (CO), and unburned
HCs, all of which are toxic to human health.
1.1.1
Nitrogen Oxides
Nitrogen oxides (NOx ) result from the reaction between nitrogen in the air and
oxygen. Theoretically, nitrogen is an inert gas. However, the high temperatures and pressures in engines create favorable conditions for the formation of
nitrogen oxides. Temperature is by far the most important parameter in nitrogen oxide formation. The most commonly found nitrogen oxide is nitric oxide
(NO), although small amounts of nitric dioxide (NO2 ) and traces of nitrous
oxide (N2 O) are present. Once released into the atmosphere, NO reacts with
the oxygen to form NO2 . This is later decomposed by the Sun’s ultraviolet
radiation back to NO and highly reactive oxygen atoms that attack the membranes of living cells. Nitrogen dioxide is partly responsible for smog; its
brownish color makes smog visible. It also reacts with atmospheric water to
form nitric acid (HNO3 ), which dilutes in rain. This phenomenon is referred
to as “acid rain” and is responsible for the destruction of forests in industrialized countries.1 Acid rain also contributes to the degradation of historical
monuments made of marble.1
1.1.2
Carbon Monoxide
Carbon monoxide results from the incomplete combustion of HCs due
to a lack of oxygen.1 It is a poison to human beings and animals who
inhale/breathe it. Once carbon monoxide reaches the blood cells, it fixes to
the hemoglobin in place of oxygen, thus diminishing the quantity of oxygen that reaches the organs and reducing the physical and mental abilities of
affected living beings.1 Dizziness is the first symptom of carbon monoxide
poisoning, which can rapidly lead to death. Carbon monoxide binds more
strongly to hemoglobin than oxygen. The bonds are so strong that normal
body functions cannot break them. People intoxicated by carbon monoxide
must be treated in pressurized chambers, where the pressure makes it easier
to break the carbon monoxide–hemoglobin bonds.
Environmental Impact and History of Modern Transportation
1.1.3
3
Unburned HCs
Unburned HCs are a result of the incomplete combustion of HCs.1,2 Depending on their nature, unburned HCs may be harmful to living beings.2 Some of
these unburned HCs may be direct poisons or carcinogenic chemicals such as
particulates, benzene, or others. Unburned HCs are also responsible for smog:
the Sun’s ultraviolet radiations interact with the unburned HCs and NO in the
atmosphere to form ozone and other products. Ozone is a molecule formed
of three oxygen atoms. It is colorless but very dangerous, and is poisonous
because as it attacks the membranes of living cells, causing them to age prematurely or die. Toddlers, older people, and asthmatics suffer greatly from
exposure to high ozone concentrations. Annually, deaths from high ozone
peaks in polluted cities have been reported.3
1.1.4
Other Pollutants
Impurities in fuels result in the emission of pollutants. The major impurity
is sulfur: mostly found in diesel and jet fuel, but also in gasoline and natural gas.1 The combustion of sulfur (or sulfur compounds such as hydrogen
sulfide) with oxygen releases sulfur oxides (SOx ). Sulfur dioxide (SO2 ) is the
major product of this combustion. On contact with air, it forms sulfur trioxide,
which later reacts with water to form sulfuric acid, a major component of acid
rain. It should be noted that sulfur oxide emissions originate from transportation sources but also largely from the combustion of coal in power plants
and steel factories. In addition, there is debate over the exact contribution of
natural sources such as volcanoes.
Petroleum companies add chemical compounds to their fuels in order
to improve the performance or lifetime of engines.1 Tetraethyl lead, often
referred to simply as “lead,” was used to improve the knock resistance of gasoline and therefore allow better engine performance. However, the combustion
of this chemical releases lead metal, which is responsible for a neurological disease called “saturnism.” Its use is now forbidden in most developed countries
and it has been replaced by other chemicals.1
1.2 Global Warming
Global warming is a result of the “greenhouse effect” induced by the presence
of carbon dioxide and other gases, such as methane, in the atmosphere. These
gases trap the Sun’s infrared radiation reflected by the ground, thus retaining
the energy in the atmosphere and increasing the temperature. An increased
Earth temperature results in major ecological damages to its ecosystems and
in many natural disasters that affect human populations.2
4
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Considering the ecological damages induced by global warming, the
disappearance of some endangered species is a concern because this destabilizes the natural resources that feed some populations. There are also
concerns about the migration of some species from warm seas to previously colder northern seas, where they can potentially destroy indigenous
species and the economies that live off those species. This may be happening
in the Mediterranean Sea, where barracudas from the Red Sea have been
observed.
Natural disasters command our attention more than ecological disasters
because of the amplitude of the damages they cause. Global warming is
believed to have induced meteorological phenomena such as “El Niño,”
which disturbs the South Pacific region and regularly causes tornadoes, inundations, and dryness. The melting of the polar icecaps, another major result of
global warming, raises the sea level and can cause the permanent inundation
of coastal regions and sometimes of entire countries.
Carbon dioxide is the result of the combustion of HCs and coal. Transportation accounts for a large share (32% from 1980 to 1999) of carbon
dioxide emissions. The distribution of carbon dioxide emissions is shown
in Figure 1.1.4
Figure 1.2 shows the trend in carbon dioxide emissions. The transportation
sector is clearly now the major contributor to carbon dioxide emissions. It
should be noted that developing countries are rapidly increasing their transportation sector, and these countries represent a very large share of the world
population. Further discussion is provided in the next subsection.
The large amounts of carbon dioxide released into the atmosphere by
human activities are believed to be largely responsible for the increase in
the global Earth temperature observed during the last decades (Figure 1.3).
It is important to note that carbon dioxide is indeed digested by plants and
sequestrated by oceans in the form of carbonates. However, these natural
Residential
19%
Transportation
32%
Commercial
15%
Industrial
34%
FIGURE 1.1 Carbon dioxide emission distribution from 1980 to 1999.
5
Environmental Impact and History of Modern Transportation
CO2 emission in million metric tons
2000
Industrial
1800
1600
1400
Transportation
1200
Residential
1000
800
600
Commercial
400
200
0
1975
1980
1985
1990
1995
2000
Year
FIGURE 1.2 Evolution of CO2 emission.
0.33
0.4
0.22
0.11
0.2
DFo
0
0
–0.11
–0.2
–0.6
–0.22
–0.33
–0.8
–0.44
–1
–0.56
18
61
18
71
18
81
18
91
19
01
19
11
19
21
19
31
19
41
19
51
19
61
19
71
19
81
19
91
–0.4
DCo
Global temperature changes (1861–1996) EPA
0.6
Year
FIGURE 1.3 Global Earth atmospheric temperature. (Source: IPCC (1995) updated.)
assimilation processes are limited and cannot assimilate all of the emitted carbon dioxide, resulting in an accumulation of carbon dioxide in the
atmosphere.
1.3 Petroleum Resources
The vast majority of fuels for transportation are liquid fuels originating
from petroleum. Petroleum is a fossil fuel, resulting from the decomposition of living matters that were imprisoned millions of years ago (Ordovician,
6
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
600–400 million years ago) in geologically stable layers. The process is roughly
the following: living matters (mostly plants) die and are slowly covered by
sediments. Over time, these accumulating sediments form thick layers and
transform to rock. The living matters are trapped in a closed space, where they
encounter high pressures and temperatures and slowly transform into either
HCs or coal, depending on their nature. This process takes millions of years
to accomplish. This is what makes the Earth’s resources in fossil fuels finite.
Proved reserves are “those quantities that geological and engineering information indicates with reasonable certainty can be recovered in the future
from known reservoirs under existing economic and operating conditions.”5
Therefore, they do not constitute an indicator of the Earth’s total reserves. The
proved reserves, as they are given in the British Petroleum 2001 estimate,5 are
given in billion tons in Table 1.1. The R/P ratio is the number of years that the
proved reserves would last if the production were to continue at its current
level. This ratio is also given in Table 1.1 for each region.5
The oil extracted nowadays is the easily extractable oil that lies close to
the surface, in regions where the climate does not pose major problems. It is
believed that far more oil lies underneath the Earth’s crust in regions such as
Siberia, or the American and Canadian Arctic. In these regions, the climate
and ecological concerns are major obstacles to extracting or prospecting for
oil. The estimation of the total Earth’s reserves is a difficult task for political
and technical reasons. A 2000 estimation of the undiscovered oil resources
by the US Geological Survey is given in Table 1.2.6
Although the R/P ratio does not include future discoveries, it is significant. Indeed, it is based on proved reserves, which are easily accessible to
this day. The amount of future oil discoveries is hypothetical, and the newly
discovered oil will not be easily accessible. The R/P ratio is also based on the
hypothesis that the production will remain constant. It is obvious, however,
that consumption (and therefore production) is increasing yearly to keep up
with the growth of developed and developing economies. Consumption is
likely to increase in gigantic proportions with the rapid development of some
TABLE 1.1
Proved Petroleum Reserves in 2000
Region
North America
South and Central America
Europe
Africa
Middle East
Former USSR
Asia Pacific
Total world
Proved Reserves in 2000 in Billion Tons
R/P Ratio
8.5
13.6
2.5
10
92.5
9.0
6.0
13.8
39
7.7
26.8
83.6
22.7
15.9
142.1
39.9
7
Environmental Impact and History of Modern Transportation
TABLE 1.2
U.S. Geological Survey Estimate of Undiscovered Oil in 2000
Region
Undiscovered Oil in 2000 in Billion Tons
North America
South and Central America
Europe
Sub-Saharan Africa and Antarctic
Middle East and North Africa
Former USSR
Asia Pacific
World (potential growth)
19.8
14.9
3.0
9.7
31.2
15.7
4.0
98.3 (91.5)
largely populated countries, particularly in the Asia-Pacific region. Figure 1.4
shows the trend in oil consumption over the last 20 years.7 Oil consumption
is given in thousand barrels per day (one barrel is about 8 metric tons).
Despite the drop in oil consumption for Eastern Europe and the former
USSR, the world trend is clearly increasing, as shown in Figure 1.5. The fastestgrowing region is Asia Pacific, where most of the world’s population lives. An
25,000
Oil consumption in thousand barrels per day
North America
20,000
Asia Pacific
15,000
Western Europe
10,000
Eastern Europe
and former USSR
5000
South & Central America
Middle East
Africa
FIGURE 1.4 Oil consumption per region.
6
98
19
19
9
94
19
92
19
8
19
90
19
8
19
86
4
19
8
82
19
19
80
0
8
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Oil consumption in thousand barrels per day
80,000
70,000
60,000
50,000
40,000
30,000
20,000
10,000
98
19
96
19
94
19
92
19
90
19
88
19
86
19
84
19
82
19
19
80
0
Year
FIGURE 1.5 World oil consumption.
explosion in oil consumption is to be expected, with a proportional increase
in pollutant emissions and CO2 emissions.
1.4
Induced Costs
The problems associated with the frenetic combustion of fossil fuels are many:
pollution, global warming, and foreseeable exhaustion of resources, among
others. Although difficult to estimate, the costs associated with these problems
are huge and indirect,8 and may be financial, human, or both.
Costs induced by pollution include, but are not limited to, health expenses,
the cost of replanting forests devastated by acid rain, and the cost of cleaning
and fixing monuments corroded by acid rain. Health expenses probably represent the largest share of these costs, especially in developed countries with
socialized medicine or health-insured populations.
Costs associated with global warming are difficult to assess. They may
include the cost of the damages caused by hurricanes, lost crops due to dryness, damaged properties due to floods, and international aid to relieve the
affected populations. The amount is potentially huge.
Most of the petroleum-producing countries are not the largest petroleumconsuming countries. Most of the production is located in the Middle East,
while most of the consumption is located in Europe, North America, and
Asia Pacific. As a result, consumers have to import their oil and depend on
the producing countries. This issue is particularly sensitive in the Middle
Environmental Impact and History of Modern Transportation
9
East, where political turmoil affected the oil delivery to Western countries
in 1973 and 1977. The Gulf War, the Iran–Iraq war, and the constant surveillance of the area by the United States and allied forces come at a cost that is
both human and financial. The dependency of Western economies on a fluctuating oil supply is potentially expensive. Indeed, a shortage in oil supply
causes a serious slowdown of the economy, resulting in damaged perishable goods, lost business opportunities, and the eventual impossibility to run
businesses.
In searching for a solution to the problems associated with oil consumption,
one has to take into account those induced costs. This is difficult because the
cost is not necessarily asserted where it is generated. Many of the induced
costs cannot be counted in asserting the benefits of an eventual solution. The
solution to these problems will have to be economically sustainable and commercially viable without government subsidies in order to sustain itself in the
long run. Nevertheless, it remains clear that any solution to these problems—
even if it is only a partial solution—will indeed result in cost savings, which
will benefit the payers.
1.5 Importance of Different Transportation
Development Strategies to Future Oil Supply
The number of years that oil resources of the Earth can support our oil supply
completely depends on the new discovery of oil reserves and cumulative oil
production (as well as cumulative oil consumption). Historical data show
that the new discovery of oil reserves grows slowly. On the other hand, the
consumption shows a high growth rate, as shown in Figure 1.6. If oil discovery
and consumption follow the current trends, the world oil resource will be used
up by about 2038.9,10
It is becoming more and more difficult to discover new reserves of
petroleum in the Earth. The cost of exploring new oil fields is becoming higher
and higher. It is believed that the scenario of oil supply will not change much
if the consumption rate cannot be significantly reduced.
As shown in Figure 1.7, the transportation sector is the primary user of
petroleum, consuming 49% of the oil used in the world in 1997. The patterns
of consumption of industrialized and developing countries are quite different, however. In the heat and power segments of the markets in industrialized
countries, nonpetroleum energy sources were able to compete with and substitute for oil throughout the 1980s; by 1990, the oil consumption in other
sectors was less than that in the transportation sector.
Most of the gains in worldwide oil use occur in the transportation sector.
Of the total increase (11.4 million barrels per day) projected for industrialized
countries from 1997 to 2020, 10.7 million barrels per day are attributed to the
10
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Remaining reserves, total reserves, and
cumulative consumption from 1970, Gb
2500
Discovered reserves
(remaining reserves +
cumulative consumption)
2000
1500
Remaining
reserves
1000
500
Cumulative
consumption
0
1970
1980
1990
2000
2010
Year
2020
2030
2040
2050
The year of oil
supply ends
FIGURE 1.6 World oil discovery, remaining reserves, and cumulative consumption.
50
Transportation
Other
45
Million barrels per day
40
35
30
25
20
15
10
5
0
90
19
97
19
05 010 015
20
2
2
Industrialized
20
20
19
90
97
19
05
20
FIGURE 1.7 World oil consumption in transportation and others.
10 015
20
2
Developing
20
20
11
Environmental Impact and History of Modern Transportation
Generalized annual oil consumption
transportation sector, where few alternatives are economical until late in the
forecast.
In developing countries, the transportation sector also shows the fastest projected growth in petroleum consumption, promising to rise nearly to the level
of nontransportation energy use by 2020. In the developing world however,
unlike in industrialized countries, oil use for purposes other than transportation is projected to contribute 42% of the total increase in petroleum consumption. The growth in nontransportation petroleum consumption in developing
countries is caused in part by the substitution of petroleum products for
noncommercial fuels (such as wood burning for home heating and cooking).
Improving the fuel economy of vehicles has a crucial impact on oil supply. So far, the most promising technologies are HEVs and fuel cell vehicles.
Hybrid vehicles, using current IC engines as their primary power source and
batteries/electric motor as the peaking power source, have a much higher
operation efficiency than those powered by IC engine alone. The hardware
and software of this technology are almost ready for industrial manufacturing. On the other hand, fuel cell vehicles, which are potentially more efficient
and cleaner than HEVs, are still in the laboratory stage and it will take a long
time to overcome technical hurdles for commercialization.
Figure 1.8 shows the generalized annual fuel consumptions of different
development strategies of next-generation vehicles. Curve a–b–c represents
the annual fuel consumption trend of current vehicles, which is assumed to
have a 1.3% annual growth rate. This annual growth rate is assumed to be
the annual growth rate of the total vehicle number. Curve a–d–e represents
a development strategy in which conventional vehicles gradually become
hybrid vehicles during the first 20 years, and after 20 years all the vehicles
will be hybrid vehicles. In this strategy, it is assumed that the hybrid vehicle
is 25% more efficient than a current conventional vehicle (25% less fuel consumption). Curve a–b–f–g represents a strategy in which, in the first 20 years,
2.2
c
2
1.8
e
1.6
1.4
b
1.2
1
a
g
d
f
0.8
0
10
20
30
40
50
60
FIGURE 1.8 Comparison of the annual fuel consumption between different development
strategies of the next-generation vehicles.
12
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Cumulative oil consumption
100
90
a-b-c
80
70
a-b-f-g
60
50
40
30
a-d-e
20
a-d-f-g
10
0
0
10
20
30
Years
40
50
60
FIGURE 1.9 Comparison of the cumulative fuel consumption between different development
strategies of the next-generation vehicles.
fuel cell vehicles are in a developing stage while current conventional vehicles are still on the market. In the second 20 years, the fuel cell vehicles will
gradually go to market, starting from point b and becoming totally fuel cell
powered at point f. In this strategy, it is assumed that 50% less fuel will be
consumed by fuel cell vehicles than by current conventional vehicles. Curve
a–d–f–g represents the strategy that the vehicles become hybrid in the first
20 years and fuel cell powered in the second 20 years.
Cumulative oil consumption is more meaningful because it involves annual
consumption and the time effect, and is directly associated with the reduction
of oil reserves as shown in Figure 1.6. Figure 1.9 shows the scenario of generalized cumulative oil consumptions of the development strategies mentioned
above. Although fuel cell vehicles are more efficient than hybrid vehicles,
the cumulative fuel consumption by strategy a–b–f–g (a fuel cell vehicle in
the second 20 years) is higher than the strategy a–d–e (a hybrid vehicle in the
first 20 years) within 45 years, due to the time effect. From Figure 1.8, it is
clear that strategy a–d–f–g (a hybrid vehicle in the first 20 years and a fuel cell
vehicle in the second 20 years) is the best. Figures 1.6 and 1.9 reveal another
important fact: that fuel cell vehicles should not rely on oil products because
of the difficulty of future oil supply 45 years later. Thus, the best development strategy of next-generation transportation would be to commercialize
HEVs immediately, and at the same time do the best to commercialize
nonpetroleum fuel cell vehicles as soon as possible.
1.6
History of EVs
The first EV was built by Frenchman Gustave Trouvé in 1881. It was a tricycle
powered by a 0.1 hp DC motor fed by lead-acid batteries. The whole vehicle
Environmental Impact and History of Modern Transportation
13
and its driver weighed approximately 160 kg. A vehicle similar to this was
built in 1883 by two British professors.11 These early realizations did not
attract much attention from the public because the technology was not mature
enough to compete with horse carriages. Speeds of 15 km/h and a range of
16 km were nothing exciting for potential customers. The 1864 Paris to Rouen
race changed it all: the 1135 km were run in 48 h and 53 min at an average
speed of 23.3 km/h. This speed was by far superior to that possible with horsedrawn carriages. The general public became interested in horseless carriages
or automobiles as these vehicles were now called.
The following 20 years were an era during which EVs competed with their
gasoline counterparts. This was particularly true in America, where there
were not many paved roads outside a few cities. The limited range of EVs
was not a problem. However, in Europe, the rapidly increasing number of
paved roads called for extended ranges, thus favoring gasoline vehicles.11
The first commercial EV was the Morris and Salom’s Electroboat. This vehicle was operated as a taxi in New York City by a company created by its inventors. The Electroboat proved to be more profitable than horse cabs despite
a higher purchase price (around $3000 vs. $1200). It could be used for three
shifts of 4 h with 90-min recharging periods in between. It was powered by two
1.5 hp motors that allowed a maximum speed of 32 km/h and a 40 km range.11
The most significant technical advance of that era was the invention of
regenerative braking by Frenchman M. A. Darracq on his 1897 coupe. This
method allows recuperating the vehicle’s kinetic energy while braking and
recharging the batteries, which greatly enhances the driving range. It is one
of the most significant contributions to electric and HEV technology as it
contributes to energy efficiency more than anything else in urban driving.
In addition, among the most significant EVs of that era was the first vehicle ever to reach 100 km. It was “La Jamais Contente” built by Frenchman
Camille Jenatzy. Note that Studebaker and Oldsmobile got started in business
by building EVs.
As gasoline automobiles became more powerful, more flexible, and above
all easier to handle, EVs started to disappear. Their high cost did not help,
but it is their limited driving range and performance that really impaired
them versus their gasoline counterparts. The last commercially significant
EVs were released around 1905. During nearly 60 years, the only EVs sold
were common golf carts and delivery vehicles.
In 1945, three researchers at Bell Laboratories invented a device that was
meant to revolutionize the world of electronics and electricity: the transistor.
It quickly replaced vacuum tubes for signal electronics and soon the thyristor
was invented, which allowed switching high currents at high voltages. This
made it possible to regulate the power fed to an electric motor without the
very inefficient rheostats and allowed the running of AC motors at variable
frequency. In 1966, General Motors (GM) built the Electrovan, which was
propelled by induction motors that were fed by inverters built with thyristors.
The most significant EV of that era was the Lunar Roving Vehicle, which
the Apollo astronauts used on the Moon. The vehicle itself weighed 209 kg
14
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
and could carry a payload of 490 kg. The range was around 65 km. The design
of this extraterrestrial vehicle, however, has very little significance down on
Earth. The absence of air and the lower gravity on the Moon, and the low
speed made it easier for engineers to reach an extended range with a limited
technology.
During the 1960s and 1970s, concerns about the environment triggered
some research on EVs. However, despite advances in battery technology and
power electronics, their range and performance were still obstacles.
The modern EV era culminated during the 1980s and early 1990s with the
release of a few realistic vehicles by firms such as GM with the EV1 and
Peugeot Société Anonyme (PSA) with the 106 Electric. Although these vehicles represented a real achievement, especially when compared with early
realizations, it became clear during the early 1990s that electric automobiles
could never compete with gasoline automobiles for range and performance.
The reason is that in batteries the energy is stored in the metal of the electrodes, which weigh far more than gasoline for the same energy content. The
automotive industry abandoned the EV to conduct research on hybrid electric
vehicles. After a few years of development, these are far closer to the assembly
line for mass production than EVs have ever been.
In the context of the development of EVs, it is the battery technology that is
the weakest, blocking the way of EVs to the market. Great effort and investment have been put into battery research, with the intention of improving
performance to meet the EV requirement. Unfortunately, progress has been
very limited. Performance is far behind the requirement, especially energy
storage capacity per unit weight and volume. This poor energy storage capability of batteries limits EVs to only some specific applications, such as at
airports, railroad stations, mail delivery routes, golf courses, and so on. In
fact, basic study12 shows that the EV will never be able to challenge the liquidfueled vehicle even with the optimistic value of battery energy capacity. Thus,
in recent years, advanced vehicle technology research has turned to HEVs as
well as fuel cell vehicles.
1.7
History of HEVs
Surprisingly, the concept of a HEV is almost as old as the automobile itself. The
primary purpose, however, was not so much to lower the fuel consumption
but rather to assist the IC engine to provide an acceptable level of performance.
Indeed, in the early days, IC engine engineering was less advanced than
electric motor engineering.
The first hybrid vehicles reported were shown at the Paris Salon of 1899.13
These were built by the Pieper establishments of Liège, Belgium and by the
Vendovelli and Priestly Electric Carriage Company, France. The Pieper vehicle was a parallel hybrid with a small air-cooled gasoline engine assisted
Environmental Impact and History of Modern Transportation
15
by an electric motor and lead-acid batteries. It is reported that the batteries
were charged by the engine when the vehicle coasted or was at a standstill.
When the driving power required was greater than the engine rating, the
electric motor provided additional power. In addition to being one of the
two first hybrid vehicles, and the first parallel hybrid vehicle, the Pieper was
undoubtedly the first electric starter.
The other hybrid vehicle introduced at the Paris Salon of 1899 was the
first series HEV and was derived from a pure EV commercially built by the
French firm Vendovelli and Priestly.13 This vehicle was a tricycle, with the two
rear wheels powered by independent motors. An additional 3/4 hp gasoline
engine coupled to a 1.1 kW generator was mounted on a trailer and could be
towed behind the vehicle to extend the range by recharging the batteries. In
the French case, the hybrid design was used to extend its range by recharging
the batteries. Also, the hybrid design was used to extend the range of an EV
and not to supply additional power to a weak IC engine
Frenchman Camille Jenatzy presented a parallel hybrid vehicle at the Paris
Salon of 1903. This vehicle combined a 6 hp gasoline engine with a 14 hp
electric machine that could either charge the batteries from the engine or
assist them later. Another Frenchman, H. Krieger, built the second reported
series hybrid vehicle in 1902. His design used two independent DC motors
driving the front wheels. They drew their energy from 44 lead-acid cells that
were recharged by a 4.5 hp alcohol spark-ignited engine coupled to a shunt
DC generator.
Other hybrid vehicles, both of the parallel and series type, were built during
a period ranging from 1899 until 1914. Although electric braking has been
used in these early designs, there is no mention of regenerative braking. It
is likely that most, possibly even all, designs used dynamic braking by short
circuiting or by placing a resistance in the armature of the traction motors. The
Lohner-Porsche vehicle of 1903 is a typical example of this approach.13 The
frequent use of magnetic clutches and magnetic couplings should be noted.
Early hybrid vehicles were built in order to assist the weak IC engines
of that time or to improve the range of EVs. They made use of the basic
electric technologies that were then available. In spite of the great creativity
that featured in their design, these early hybrid vehicles could no longer
compete with the greatly improved gasoline engines that came into use after
World War I. The gasoline engine made tremendous improvements in terms
of power density, the engines became smaller and more efficient, and there
was no longer a need to assist them with electric motors. The supplementary
cost of having an electric motor and the hazards associated with the lead-acid
batteries were key factors in the disappearance of hybrid vehicles from the
market after World War I.
However, the greatest problem that these early designs had to cope with
was the difficulty of controlling the electric machine. Power electronics did
not become available until the mid-1960s and early electric motors were controlled by mechanical switches and resistors. They had a limited operating
16
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
range incompatible with efficient operation. Only with great difficulty could
they be made compatible with the operation of a hybrid vehicle.
Dr. Victor Wouk is recognized as the modern investigator of the HEV
movement.13 In 1975, along with his colleagues, he built a parallel hybrid
version of a Buick Skylark.13 The engine was a Mazda rotary engine, coupled
to a manual transmission. It was assisted by a 15 hp separately excited DC
machine, located in front of the transmission. Eight 12 V automotive batteries
were used for energy storage. A top speed of 80 mph (129 km/h) was achieved
with acceleration from 0 to 60 mph in 16 s.
The series hybrid design was revived by Dr. Ernest H. Wakefield in 1967,
when working for Linear Alpha Inc. A small engine coupled to an AC
generator, with an output of 3 kW, was used to keep a battery pack charged.
However, the experiments were quickly stopped because of technical problems. Other approaches studied during the 1970s and early 1980s used range
extenders, similar in concept to the French Vendovelli and Priestly 1899
design. These range extenders were intended to improve the range of EVs
that never reached the market. Other prototypes of hybrid vehicles were
built by the Electric Auto Corporation in 1982 and by the Briggs & Stratton
Corporation in 1980. These were both parallel hybrid vehicles.
Despite the two oil crises of 1973 and 1977, and despite growing environmental concerns, no HEV made it to the market. The researchers’ focus was
drawn by the EV, of which many prototypes were built during the 1980s. The
lack of interest in HEVs during this period may be attributed to the lack of
practical power electronics, modern electric motor, and battery technologies.
The 1980s witnessed a reduction in conventional IC engine-powered vehicle
sizes, the introduction of catalytic converters, and the generalization of fuel
injection.
The HEV concept drew great interest during the 1990s when it became
clear that EVs would never achieve the objective of saving energy. The Ford
Motor Corporation initiated the Ford Hybrid Electric Vehicle Challenge,
which drew efforts from universities to develop hybrid versions of production
automobiles.
Automobile manufacturers around the world built prototypes that achieved
tremendous improvements in fuel economy over their IC engine-powered
counterparts. In the United States, Dodge built the Intrepid ESX 1, 2, and 3.
The ESX-1 was a series hybrid vehicle, powered by a small turbocharged threecylinder diesel engine and a battery pack. Two 100 hp electric motors were
located in the rear wheels. The U.S. government launched the Partnership for
a New Generation of Vehicles (PNGV), which included the goal of a mid-size
sedan that could achieve 80 mpg. The Ford Prodigy and GM Precept resulted
from this effort. The Prodigy and Precept vehicles were parallel HEVs powered by small turbocharged diesel engines coupled to dry clutch manual transmissions. Both of them achieved the objective but production did not follow.
Efforts in Europe are represented by the French Renault Next, a small parallel hybrid vehicle using a 750 cc spark-ignited engine and two electric motors.
Environmental Impact and History of Modern Transportation
17
This prototype achieved 29.4 km/L (70 mpg) with maximum speed and acceleration performance comparable to conventional vehicles. Volkswagen also
built a prototype, the Chico. The base was a small EV, with a nickel-metal
hydride battery pack and a three-phase induction motor. A small two-cylinder
gasoline engine was used to recharge the batteries and provide additional
power for high-speed cruising.
The most significant effort in the development and commercialization of
HEVs was made by Japanese manufacturers. In 1997, Toyota released the
Prius sedan in Japan. Honda also released its Insight and Civic Hybrid.
These vehicles are now available throughout the world. They achieve excellent figures of fuel consumption. Toyota’s Prius and Honda’s Insight vehicles
have historical value in that they are the first hybrid vehicles commercialized in the modern era to respond to the problem of personal vehicle fuel
consumption.
1.8 History of Fuel Cell Vehicles
As early as 1839, Sir William Grove (often referred to as the “Father of the Fuel
Cell”) discovered that it may be possible to generate electricity by reversing
the electrolysis of water. It was not until 1889 that two researchers, Charles
Langer and Ludwig Mond, coined the term “fuel cell” as they were trying to
engineer the first practical fuel cell using air and coal gas. Although further
attempts were made in the early 1900s to develop fuel cells that could convert
coal or carbon into electricity, the advent of IC engine temporarily quashed
any hopes of further development of the fledgling technology.
Francis Bacon developed what was perhaps the first successful fuel cell
device in 1932, with a hydrogen–oxygen cell using alkaline electrolytes and
nickel electrodes—inexpensive alternatives to the catalysts used by Mond
and Langer. Due to a substantial number of technical hurdles, it was not until
1959 that Bacon and company first demonstrated a practical 5-kW fuel cell
system. Harry Karl Ihrig presented his now-famous 20-hp fuel-cell-powered
tractor that same year.
National Aeronautics and Space Administration (NASA) also began building compact electric generators for use on space missions in the late 1950s.
NASA soon came to fund hundreds of research contracts involving fuel cell
technology. Fuel cells now have a proven role in the space program, after
supplying electricity for several space missions.
In more recent decades, a number of manufacturers—including major
automakers—and various federal agencies have supported ongoing research
into the development of fuel cell technology for use in fuel cell vehicles and
other applications.14 Hydrogen production, storage, and distribution are the
biggest challenges. Truly, fuel-cell-powered vehicles still have a long way to
go to enter the market.
18
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
References
1. C. R. Ferguson and A. T. Kirkpatrick, Internal Combustion Engines—Applied ThermoSciences, Second Edition, John Wiley & Sons, New York, 2001.
2. U.S. Environmental Protection Agency (EPA), “Automobile emissions: An
overview,” EPA 400-F-92-007, Fact Sheet OMS-5, August 1994.
3. U.S. Environmental Protection Agency (EPA), “Automobiles and ozone,” EPA
400-F-92-006, Fact Sheet OMS-4, January 1993.
4. Energy Information Administration, U.S. Department of Energy, “Carbon dioxide
emissions from energy consumption by sector, 1980–1999,” 2001, available at
http://www.eia.doe.gov/emeu/aer/txt/tab1202.htm.
5. “BP statistical review of world energy—oil,” 2001, available at http://www.bp.com/
downloads/837/global_oil_section.pdf.
6. USGS World Energy Assessment Team, “World undiscovered assessment
results summary,” U.S. Geological Survey Digital Data Series 60, available at
http://greenwood.cr.usgs.gov/energy/WorldEnergy/DDS-60/sum1.html#TOP.
7. International Energy Database, Energy Information Administration, U.S. Department of Energy, “World petroleum consumption, 1980–1999,” 2000.
8. D. Doniger, D. Friedman, R. Hwang, D. Lashof, and J. Mark, “Dangerous addiction: Ending America’s oil dependence,” National Resources Defense Council and
Union of Concerned Scientists, 2002.
9. M. Ehsani, D. Hoelscher, N. Shidore, and P. Asadi, “Impact of hybrid electric
vehicles on the world’s petroleum consumption and supply,” Society of Automotive
Engineers (SAE) Future Transportation Technology Conference, Paper No. 2003-012310, 2003.
10. J. E. Hake, “International energy outlook—2000 with projection to 2020,” available
at http://tonto.eia.doe.gov/FTPROOT/presentations/ieo2000/sld008.htm.
11. E. H. Wakefield, History of the Electric Automobile: Battery-only Powered Cars, Society
of Automotive Engineers (SAE), ISBN: 1-56091-299-5, Warrendale, PA, 1994.
12. Y. Gao and M. Ehsani, “An investigation of battery technologies for the Army’s
hybrid vehicle application,” in Proceedings of the IEEE 56th Vehicular Technology
Conference, Vancouver, British Columbia, Canada, September 2002.
13. E. H. Wakefield, History of the Electric Automobile: Hybrid Electric Vehicles, Society
of Automotive Engineers (SAE), ISBN: 0-7680-0125-0, Warrendale, PA, 1998.
14. California Fuel Cell Partnership, available at http://www.fuelcellpartnership.org/.
2
Fundamentals of Vehicle Propulsion
and Brake
Vehicle operation fundamentals mathematically describe vehicle behavior,
based on the general principles of mechanics. A vehicle, consisting of thousands of components, is a complex system. To describe its behavior fully,
sophisticated mechanical and mathematical knowledge is needed. A great
amount of literature in this field already exists. Since this book proposes to
discuss electric, hybrid electric, and fuel cell power trains, the discussion
of vehicle fundamentals will be restricted to one-dimensional movement.
This chapter will, therefore, focus on vehicle performance, such as speed,
gradeability, acceleration, fuel consumption, and braking performance.
2.1 General Description of Vehicle Movement
The movement behavior of a vehicle along its moving direction is completely
determined by all the forces acting on it in this direction. Figure 2.1 shows the
forces acting on a vehicle moving up a grade. The tractive effort, Ft , in the contact area between the tires of the driven wheels and the road surface propels
the vehicle forward. It is produced by the power plant torque and transferred
through transmission and final drive to the drive wheels. While the vehicle is
moving, there is resistance that tries to stop its movement. The resistance usually includes tire rolling resistance, aerodynamic drag, and uphill resistance.
According to Newton’s second law, vehicle acceleration can be written as
dV
=
dt
Ft − Fr
,
δM
(2.1)
where Vis the speed of the vehicle,
Ft is the total tractive effort of the
vehicle,
Fr is the total resistance, M is the total mass of the vehicle, and δ
is the mass factor that equivalently converts the rotational inertias of rotating
components into translational mass.
19
20
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
V
Fw
hw
O Mg sin
a
Tr
f
hg
Ft
W
f
Mg
cos
a
Mg
La
Tr
r
Lb
a
L
W
r
FIGURE 2.1 Forces acting on a vehicle moving uphill.
2.2
Vehicle Resistance
As shown in Figure 2.1, vehicle resistances opposing its movement include
rolling resistance of the tires, appearing in Figure 2.1 as rolling resistance
torques Trf and Trr , aerodynamic drag, Fw , and hill climbing resistance (the
term Mg sin α in Figure 2.1). All of the resistances will be discussed in detail
in the following sections.
2.2.1
Rolling Resistance
The rolling resistance of tires on hard surfaces is primarily caused by hysteresis in the tire materials. Figure 2.2 shows a tire at standstill, on which a
force, P, is acting at its center. The pressure in the contact area between the
tire and ground is distributed symmetrically to the central line and the resultant reaction force, Pz , is aligned to P. The deformation, z, versus the load, P,
in the loading and unloading process is shown in Figure 2.3. Due to hysteresis
in the deformation of rubber material, the load at loading is larger than that
at unloading at the same deformation, z, as shown in Figure 2.3. When the
tire is rolling, as shown in Figure 2.4a, the leading half of the contact area is
loading and the trailing half is unloading. Consequently, the hysteresis causes
an asymmetric distribution of the ground reaction forces. The pressure in the
leading half of the contact area is larger than that in the trailing half, as shown
in Figure 2.4a. This phenomenon results in the ground reaction force shifting forward somewhat. This forwardly shifted ground reaction force, with
the normal load acting on the wheel center, creates a moment, which opposes
21
Fundamentals of Vehicle Propulsion and Brake
P
r
z
Pz
FIGURE 2.2 Pressure distribution in contact area.
rolling of the wheel. On soft surfaces, the rolling resistance is primarily caused
by deformation of the ground surface as shown in Figure 2.4b. The ground
reaction force almost completely shifts to the leading half.
The moment produced by the forward shift of the resultant ground reaction
force is called rolling resistant moment, as shown in Figure 2.4a, and can be
expressed as
Tr = Pa.
(2.2)
To keep the wheel rolling, a force, F, acting on the center of the wheel is
required to balance this rolling resistant moment. This force is expressed as
Tr
Pa
=
= Pfr ,
rd
rd
Force, P
F=
P1
P2
Deformation, z
FIGURE 2.3 Force acting on a tire versus tire deformation in loading and unloading.
(2.3)
22
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
(a)
(b)
P
F
P
Moving
direction
F
Moving
direction
r
r
rd
z
Px
z
a
Pz
P
On hard road surface
On soft road surface
FIGURE 2.4 Tire deflection and rolling resistance on a (a) hard and (b) soft road surface.
where rd is the effective radius of the tire and fr = a/rd is called the rolling resistance coefficient. In this way, the rolling resistant moment can be equivalently
replaced by a horizontal force acting on the wheel center in the opposite movement direction of the wheel. This equivalent force is called rolling resistance
with a magnitude of
Fr = Pfr ,
(2.4)
where P is the normal load acting on the center of the rolling wheel. When a
vehicle is operated on a slope road, the normal load, P, should be replaced by
the component that is perpendicular to the road surface. That is,
Fr = Pfr cos α,
(2.5)
where α is the road angle (refer to Figure 2.1).
The rolling resistance coefficient, fr , is a function of tire material, tire structure, tire temperature, tire inflation pressure, tread geometry, road roughness,
road material, and presence or absence of liquids on the road. Typical values
of rolling resistance coefficients on various roads are given in Table 2.1.1 For
fuel saving in recent years, low-resistance tires for passenger cars have been
developed. Their rolling resistance coefficient is less than 0.01.
The values given in Table 2.1 do not take into account their variations with
speed. Based on experimental results, many empirical formulas have been
proposed for calculating the rolling resistance on a hard surface. For example,
the rolling resistance coefficient of passenger cars on a concrete road may be
calculated from the following equation:
fr = f0 + fs
V
100
2.5
,
(2.6)
23
Fundamentals of Vehicle Propulsion and Brake
TABLE 2.1
Rolling Resistance Coefficients
Conditions
Car tires on a concrete or asphalt road
Car tires on a rolled gravel road
Tar macadam road
Unpaved road
Field
Truck tire on a concrete or asphalt road
Wheel on iron rail
Rolling Resistance Coefficient
0.013
0.02
0.025
0.05
0.1–0.35
0.006–0.01
0.001–0.002
where V is vehicle speed in km/h, and f0 and fs depend on the inflation
pressure of the tire.2
In vehicle performance calculation, it is sufficient to consider the rolling
resistance coefficient as a linear function of speed. For the most common range
of inflation pressure, the following equation can be used for a passenger car
on a concrete road2 :
V
fr = 0.01 1 +
.
(2.7)
160
This equation predicts the values of fr with acceptable accuracy for speeds up
to 128 km/h.
2.2.2 Aerodynamic Drag
A vehicle traveling at a particular speed in air encounters a force resisting its
motion. This force is referred to as aerodynamic drag. It mainly results from
two components: shape drag and skin friction.
Shape drag: The forward motion of the vehicle pushes the air in front of it.
However, the air cannot instantaneously move out of the way and its pressure
is thus increased, resulting in high air pressure. In addition, the air behind
the vehicle cannot instantaneously fill the space left by the forward motion
of the vehicle. This creates a zone of low air pressure. The motion of the
vehicle, therefore, creates two zones of pressure that oppose the motion by
pushing (high pressure in front) and pulling it backwards (low pressure at the
back) as shown in Figure 2.5. The resulting force on the vehicle is the shape
drag. The name “shape drag” comes from the fact that this drag is completely
determined by the shape of the vehicle body.
Skin friction: Air close to the skin of the vehicle moves almost at the speed
of the vehicle while air away from the vehicle remains still. In between, air
molecules move at a wide range of speeds. The difference in speed between
two air molecules produces a friction that results in the second component of
aerodynamic drag.
24
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
High pressure
Low pressure
Moving direction
FIGURE 2.5 Shape drag.
Aerodynamic drag is a function of vehicle speed V, vehicle frontal area, Af ,
shape of the vehicle body, and air density, ρ:
Fw =
1
ρAf CD (V − Vw )2 ,
2
(2.8)
where CD is the aerodynamic drag coefficient that characterizes the shape of
the vehicle body and Vw is component of the wind speed on the vehicle moving direction, which has a positive sign when this component is in the same
direction of the moving vehicle and a negative sign when it is opposite to
the vehicle speed. The aerodynamic drag coefficients for typical vehicle body
shapes are shown in Figure 2.6.
2.2.3
Grading Resistance
When a vehicle goes up or down a slope, its weight produces a component
that is always directed in the downward direction, as shown in Figure 2.7.
This component either opposes the forward motion (grade climbing) or helps
the forward motion (grade descending). In vehicle performance analysis,
only uphill operation is considered. This grading force is usually called
grading resistance.
Grading resistance, referring to Figure 2.7, can be expressed as
Fg = Mg sin α.
(2.9)
To simplify the calculation, the road angle, α, is usually replaced by the grade
value, when the road angle is small. As shown in Figure 2.7, grade is defined as
i=
H
= tan α ≈ sin α.
L
(2.10)
25
Fundamentals of Vehicle Propulsion and Brake
Coefficient of aerodynamic
resistance
Vehicle
type
Open convertible
0.5º0.7
Van body
0.5º0.7
Ponton body
0.4º0.55
Wedged-shaped body; headlamps
and bumpers are integrated into
the body, covered underbody,
optimized cooling air flow
0.3º0.4
Headlamp and all wheels in
body, covered underbody
0.2º0.25
K-shaped (small breakaway section)
0.23
Optimum streamlined design
0.15º0.20
Trucks, road trains
Buses
Streamlined buses
Motorcycles
0.8º1.5
0.6º0.7
0.3º0.4
0.6º0.7
FIGURE 2.6 Indicative drag coefficients for different body shapes.
In some literature, the tire rolling resistance and grading resistance together
are called road resistance, which is expressed as
Frd = Ff + Fg = Mg( fr cos α + sin α).
(2.11)
O Mg sin
a
H
hg
a
Mg
co
sa
Mg
a
L
FIGURE 2.7 Vehicle climbing a grade.
26
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
When the road angle is small, the road resistance can be simplified as
Frd = Ff + Fg = Mg( fr + i).
2.3
(2.12)
Dynamic Equation
In the longitudinal direction, the major external forces acting on a two-axle
vehicle, as shown in Figure 2.1, include the rolling resistance of the front and
rear tires Frf and Frr , which are represented by rolling resistance moment, Trf
and Trr , aerodynamic drag, Fw , climbing resistance, Fg (Mg sin α), and tractive
effort of the front and rear tires, Ftf and Ftr . Ftf is zero for a rear-wheel-driven
vehicle, whereas Ftr is zero for a front-wheel-driven vehicle.
The dynamic equation of vehicle motion along the longitudinal direction is
expressed by
M
dV
= (Ftf + Ftr ) − (Frf + Frr + Fw + Fg ),
dt
(2.13)
where dV/dt is the linear acceleration of the vehicle along the longitudinal
direction and M is the vehicle mass. The first term on the right-hand side of
Equation 2.13 is the total tractive effort and the second term is the resistance.
To predict the maximum tractive effort that the tire–ground contact can
support, the normal loads on the front and rear axles have to be determined. By
summing the moments of all the forces about point R (center of the tire–ground
area), the normal load on the front axle Wf can be determined as
Wf =
MgLb cos α − (Trf + Trr + Fw hw + Mghg sin α + Mhg dV/dt)
. (2.14)
L
Similarly, the normal load acting on the rear axle can be expressed as
Wr =
MgLa cos α + (Trf + Trr + Fw hw + Mghg sin α + Mhg dV/dt)
. (2.15)
L
For passenger cars, the height of the center of application of aerodynamic
resistance, hw , is assumed to be near the height of the gravity center of the
vehicle, hg . Equations 2.14 and 2.15 can be simplified as
Wf =
hg
rd
Lb
dV
Mg cos α −
Fw + Fg + Mg fr cos α + M
L
L
hg
dt
(2.16)
Wr =
hg
rd
La
dV
Mg cos α +
Fw + Fg + Mg fr cos α + M
,
L
L
hg
dt
(2.17)
and
Fundamentals of Vehicle Propulsion and Brake
27
where rd is the effective radius of the wheel. Referring to Equations 2.5 and
2.13, Equations 2.16 and 2.17 can be rewritten as
hg
rd
Lb
Mg cos α −
Ft − F r 1 −
L
hg
L
(2.18)
hg
rd
La
Wr =
,
Mg cos α +
Ft − F r 1 −
L
L
hg
(2.19)
Wf =
and
where Ft = Ftf + Ftr is the total tractive effort of the vehicle and Fr is the
total rolling resistance of the vehicle. The first term on the right-hand side of
Equations 2.18 and 2.19 is the static load on the front and rear axles when the
vehicle is at rest on level ground. The second term is the dynamic component
of the normal load.
The maximum tractive effort that the tire–ground contact can support (any
small amount over this maximum tractive effort will cause the tire to spin on
the ground) is usually described by the product of the normal load and the
coefficient of road adhesion, μ, or referred to as frictional coefficient in some
of the literature (more details are given in Section 2.4). For a front-wheeldriven vehicle,
hg
rd
Lb
= μWf = μ
Mg cos α −
Ft max − Fr 1 −
L
L
hg
Ft max
and
Ft max
μMg cos α Lb + fr (hg − rd ) /L
=
,
1 + μhg /L
(2.20)
(2.21)
where fr is the coefficient of the rolling resistance. Similarly, for a rear-wheeldriven vehicle,
Ft max = μWr = μ
and
Ft max
hg
rd
La
Mg cos α +
Ft max − Fr 1 −
L
L
hg
μMg cos α La − fr (hg − rd ) /L
=
.
1 − μhg /L
(2.22)
(2.23)
In vehicle operation, the maximum tractive effort on the driven wheels, transferred from power plant through transmission, should not exceed the maximum values that are limited by the tire–ground cohesion in Equations 2.21
and 2.23. Otherwise, the driven wheels will spin on the ground, leading to
vehicle instability.
28
2.4
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Tire–Ground Adhesion and Maximum Tractive Effort
When the tractive effort of a vehicle exceeds the limitation of the maximum
tractive effort due to the adhesive capability between the tire and ground,
the driven wheels will spin on the ground. Actually, the adhesive capability
between the tire and the ground sometimes is the main limitation of vehicle
performance. This is especially true when the vehicle drives on wet, icy, snowcovered, or soft soil roads. In this case, a tractive torque on the driven wheel
would cause the wheel to slip significantly on the ground. The maximum
tractive effort on the driven wheel depends on the longitudinal force that the
adhesive capability between the tire and ground can supply, rather than the
maximum torque that an engine can supply.
Experimental results show that, on various types of ground, the maximum
tractive effort of the drive wheel closely relates to the slipping of the running
wheel. This is also true on a good paved, dry road where the slipping is very
small due to the elasticity of the tire. The slip, s, of a tire is defined in traction as
V
s= 1−
rω
× 100%,
(2.24)
where V is the translatory speed of the tire center, ω is the angular speed of
the tire, and r is the rolling radius of the free rolling tire. In traction, the speed,
V, is less than rω; therefore the slip of a tire has a positive value between 0
and 1.0. During braking, however, the slip of a tire can be defined as
rω × 100%,
s= 1−
V
(2.25)
which has a positive value between 0 and 1.0, similar to traction. The maximum traction effort of a tire corresponding to a certain tire slip is usually
expressed as
Fx = Pμ(s),
(2.26)
where P is the vertical load of the tire and μ is the tractive effort coefficient,
which is a function of tire slip. The tractive effort coefficient and the tire slip
have the relationship as shown in Figure 2.8.
In the small-slip range (section OA in Figure 2.8), the tractive effort is
almost linearly proportional to the slip value. This small slip is caused by
the elasticity of the tire rather than the relative slipping between the tire and
ground at the contact patch, as shown in Figure 2.9. When a tractive torque
is applied to the tire, a tractive force is developed at the tire–ground contact
patch. At the same time, the tire tread in front and within the contact patch
is subjected to compression. A corresponding shear deformation of the side
wall of the tire is also developed. As tread elements are compressed before
entering the contact region, the distance that the tire travels will be less than
29
Fundamentals of Vehicle Propulsion and Brake
Tractive effort coefficient
B
Longitudinal
A
mp
Lateral
O
0
15-20
ms
50
Slip
100%
FIGURE 2.8 Variation of tractive effort coefficient with longitudinal slip of a tire.
the distance in a free rolling tire. Because of the nearly linear elastic property
of the tire, the tractive effort–slip curve is almost linear. Further increase of
wheel torque and tractive force results in part of the tire tread sliding on the
ground. Under these circumstances, the relationship between tractive force
and slip is nonlinear. This corresponds to section AB of the curve, as shown
in Figure 2.8. The peak tractive effort is reached at a slip of 15–20%. Further
increase of the slip beyond that results in an unstable condition. The tractive
effort coefficient falls rapidly from the peak value to the purely sliding value
as shown in Figure 2.8. For normal driving, the slip of the tire must be limited
l
P
w
Tw
Compression
Fx
(1 – e)l
a
Longitudinal
stress
Normal
pressure
Pz
Pz
Fx
FIGURE 2.9 Behavior of a tire under the action of driving torque.
30
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
TABLE 2.2
Average Values of Tractive Effort Coefficient on Various Roads
Surface
Asphalt and concrete (dry)
Concrete (wet)
Asphalt (wet)
Gravel
Earth road (dry)
Earth road (wet)
Snow (hard packed)
Ice
Peaking Values, μp
Slipping Values, μs
0.8–0.9
0.8
0.5–0.7
0.6
0.68
0.55
0.2
0.1
0.75
0.7
0.45–0.6
0.55
0.65
0.4–0.5
0.15
0.07
in the range less than 15–20%. Table 2.2 shows the average values of tractive
effort coefficients on various roads.2
2.5
Power Train Tractive Effort and Vehicle Speed
An automotive power train, as shown in Figure 2.10, consists of a power
plant (engine or electric motor), a clutch in manual transmission or a torque
converter in automatic transmission, a gearbox (transmission), final drive,
differential, drive shaft, and driven wheels. The torque and rotating speed
from the output shaft of the power plant are transmitted to the driven
wheels through the clutch or torque converter, gearbox, final drive, differential, and drive shaft. The clutch is used in manual transmission to couple or
decouple the gearbox to the power plant. The torque converter in automatic
transmission is a hydrodynamic device, functioning as the clutch in manual
transmission with a continuously variable gear ratio (for more details, see
Engine
Clutch or
torque
converter
Transmission
Differential
Final drive
FIGURE 2.10 Conceptual illustration of an automobile power train.
Drive shaft
Driven wheel
31
Fundamentals of Vehicle Propulsion and Brake
Section 2.6). The gearbox supplies a few gear ratios from its input shaft to its
output shaft for the power plant torque–speed profile to match the requirements of the load. The final drive is usually a pair of gears that supply a
further speed reduction and distribute the torque to each wheel through the
differential.
The torque on the driven wheels, transmitted from the power plant, is
expressed as
(2.27)
Tw = ig i0 ηt Tp ,
where ig is the gear ratio of the transmission defined as ig = Nin /Nout (Nin —
input rotating speed, Nout —output rotating speed), i0 is the gear ratio of the
final drive, ηt is the efficiency of the driveline from the power plant to the
driven wheels, and Tp is the torque output from the power plant.
The tractive effort on the driven wheels, as shown in Figure 2.11, can be
expressed as
Tw
.
(2.28)
Ft =
rd
Substituting Equation 2.27 into Equation 2.28 yields the following result:
Ft =
Tp ig i0 ηt
.
rd
(2.29)
The friction in the gear teeth and bearings creates losses in the mechanical
gear transmission. The following are representative values of the mechanical
efficiency of various components:
Clutch: 99%.
Each pair of gears: 95–97%.
Bearing and joint: 98–99%.
The total mechanical efficiency of the transmission between the engine
output shaft and driven wheels is the product of the efficiencies of all the
V
Tw
Nw
rd
Ft
FIGURE 2.11 Tractive effort and torque on a driven wheel.
32
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
components in the driveline. As a first approximation, the following average
values of the overall mechanical efficiency of a manual gear-shift transmission
may be used:
Direct gear: 90%.
Other gear: 85%.
Transmission with very high reduction ratio: 75–80%.
The rotating speed (rpm) of the driven wheel can be expressed as
Nw =
Np
,
ig i0
(2.30)
where Np is the transmission rotating speed (rpm), which is equal to the
engine speed in the vehicle with a manual transmission and the turbine speed
of a torque converter in the vehicle with an automatic transmission (for more
details, see Section 2.6). The translational speed of the wheel center (vehicle
speed) can be expressed as
V=
πNw rd
(m/s).
30
(2.31)
Substituting Equation 2.30 into Equation 2.31 one obtains
V=
2.6
πNp rd
(m/s).
30ig i0
(2.32)
Vehicle Power Plant and Transmission Characteristics
There are two limiting factors to the maximum tractive effort of the vehicle.
One is the maximum tractive effort that the tire–ground contact can support
(Equation 2.21 or 2.23) and the other is the tractive effort that the maximum
torque of the power plant can produce with given driveline gear ratios (Equation 2.29). The smaller of these two factors will determine the performance
potential of the vehicle. For on-road vehicles, the performance is usually
limited by the second factor.
2.6.1
Power Plant Characteristics
For vehicular applications, the ideal performance characteristic of a power
plant is a constant power output over the full speed range. Consequently,
the torque varies with speed hyperbolically as shown in Figure 2.12. With
this ideal profile, the maximum power of the power plant will be available at
33
Fundamentals of Vehicle Propulsion and Brake
Power
Torque
Speed
FIGURE 2.12 Ideal performance characteristics for a vehicle traction power plant.
300
Torque
Power (kW)
80
240
Power
60
180
40
310
Specific fuel
consumption
20
290
270
0
1000
2000
3000
4000
Speed (rpm)
5000
FIGURE 2.13 Typical performance characteristics of gasoline engines.
Specific fuel
consumption (g/kWh)
100
Torque (Nm)
any vehicle speed, therefore yielding optimal vehicle performance. However,
in practice, the torque is constrained to be constant in low speed, so as not to
be over the maxima limited by the adhesion between the tire–ground contact
area. This constant power characteristic will provide the vehicle with high
tractive effort at low speeds where demands for acceleration, drawbar pull,
or grade climbing capability are high.
IC engines are the most commonly used power plants for land vehicles to
date. Representative characteristics of a gasoline engine in wide open throttle are shown in Figure 2.13, which has torque–speed characteristics far from
the ideal performance characteristic required by traction. It starts operating
smoothly at the idle speed. Good combustion quality and maximum torque
are reached at an intermediate engine speed. As the speed further increases,
the torque decreases due to less air induced into the cylinders, caused by
the growing losses in the air-induction manifold and grossing power losses
caused by mechanical frication and hydraulic viscosity. Power output, however, increases to its maximum at a certain higher speed. Beyond this speed,
34
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
5
1st gear
4
2nd gear
3
3rd gear
2
4th gear
1
0
0
20 40 60 80 100 120 140 160 180 200
Vehicle speed (km/h)
FIGURE 2.14 Tractive effort of an IC engine and a multigear transmission vehicle versus vehicle
speed.
the engine power starts declining. In vehicular applications, the maximum
permissible speed of the engine is usually set just a little above the speed of
the maximum power output. The IC engine has a relatively flat torque–speed
profile (as compared with an ideal power plant), as shown in Figure 2.13.
Consequently, a multigear transmission is usually employed to modify it, as
shown in Figure 2.14.
The electric motor is another candidate as a vehicle power plant, and becoming more and more important with the rapid development of electric, hybrid
electric, and fuel cell vehicles. Electric motors with good speed adjustment
control usually have a speed–torque characteristic that is much closer to the
ideal, as shown in Figure 2.15.
Generally, the electric motor starts from zero speed. As it increases to its
base speed, the voltage increases to its rated value while the flux remains
80
400
Motor power (kW)
Power
60
50
Torque
300
250
40
200
30
150
100
20
Base
speed
10
0
0
1000
2000
3000
4000
Motor (rpm)
Motor torque (Nm)
350
70
50
5000
FIGURE 2.15 Typical performance characteristics of electric motors for traction.
35
Fundamentals of Vehicle Propulsion and Brake
Tractive effort on wheel (kN)
7
6
5
4
3
2
1
0
0
50
100
150
200
Speed (km/h)
FIGURE 2.16 Tractive effort of a single-gear EV versus vehicle speed.
constant. In this speed range of zero to base speed, the electric motor produces a constant torque. Beyond the base speed, the voltage remains constant
and the flux is weakened. This results in a constant output power while the
torque declines hyperbolically with speed. Since the speed–torque profile of
an electric motor is close to the ideal, a single-gear or double-gear transmission may be employed to meet the vehicle performance requirement, as
shown in Figure 2.16.
2.6.2 Transmission Characteristics
The transmission requirements of a vehicle depend on the characteristics of
the power plant and the performance requirements of the vehicle. As mentioned previously, a well-controlled electric machine, such as the power plant
of an EV, would not need a multigear transmission. However, an IC engine
has to use multigear or continuously varying transmission to multiply its
torque at low speed. The term transmission here includes all those systems
employed for transmitting the engine power to the drive wheels. For automobile applications, there are usually two basic types of transmission: manual
gear transmission and hydrodynamic transmission.
2.6.3
Manual Gear Transmission
Manual gear transmission consists of a clutch, a gearbox, a final drive, and
a drive shaft as shown in Figure 2.10. The final drive has a constant gear
ratio. The common practice of requiring direct drive (nonreducing) in the
gearbox to be in the highest gear determines this ratio. The gearbox provides
a number of gear ratios ranging from three to five for passenger cars and
more for heavy commercial vehicles that are powered with gasoline or diesel
engines.
36
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Ideal tractive effort
Tractive effort
1st gear
2nd gear
3rd gear
4th gear
Speed
FIGURE 2.17 Tractive effort characteristics of a gasoline-engine-powered vehicle.
The maximum speed of the vehicle determines the gear ratio of the highest
gear (i.e., the smallest ratio). On the other hand, the gear ratio of the lowest
gear (i.e., the maximum ratio) is determined by the requirement of the maximum tractive effort or the gradeability. Ratios between them should be spaced
in such a way that they will provide the tractive effort–speed characteristics
as close to the ideal as possible, as shown in Figure 2.17. In the first iteration
of transmission design, gear ratios between the highest and the lowest gear
may be selected in such a way that the engine can operate in the same speed
range for all the gears. This approach would benefit the fuel economy and
performance of the vehicle. For instance, in normal driving, the proper gear
can be selected, according to vehicle speed, to operate the engine in its optimum speed range for fuel-saving purposes. In fast acceleration, the engine
can be operated in its speed range with high power output. This approach is
depicted in Figure 2.18.
For a four-speed gearbox, the following relationship can be established (see
Figure 2.18):
ig2
ig3
ig1
=
(2.33)
=
= Kg
ig2
ig3
ig4
and
Kg =
3
ig1
,
ig4
(2.34)
where ig1 , ig2 , ig3 , and ig4 are the gear ratios for the first, second, third, and
fourth gear, respectively. In the more general case, if the ratio of the highest
gear, ign (smallest gear ratio), and the ratio of the lowest gear, ig1 (largest gear
37
Fundamentals of Vehicle Propulsion and Brake
Engine operating
speed range
3rd
V3
V2
2nd
Vehicle speed
4th gear
V4
1st
V1
ne1
ne2
Engine speed
FIGURE 2.18 Demonstration of vehicle and speed ranges for each gear.
ratio), have been determined and the number of the gear ng is known, the
factor Kg can be determined as
Kg =
ng −1
ig1
,
ign
(2.35)
and each gear ratio can be obtained by
ign−1 = Kg ign ,
ign−2 = Kg2 ign ,
..
.
(2.36)
ng −1
ig2 = Kg
ign .
For passenger cars that usually use high gear in normal driving, the step
between the ratios of the upper two gears is often slightly closer than that
calculated from Equation 2.36. That is,
ig2
ig3
ig1
>
>
.
ig2
ig3
ig4
(2.37)
This, in turn, affects the selection of the ratios of the lower gears. For commercial vehicles, however, the gear ratios in the gearbox are often arranged
based on Equation 2.37.
38
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
10
1st gear
Tractive effort (kN)
8
Electric motor with
single gear transmission
6
2nd gear
4
3rd gear
2
0
0
4th gear
Gasoline engine with
four gear transmission
100
Vehicle speed (km/h)
200
FIGURE 2.19 Tractive efforts of a gasoline engine vehicle with a four-gear transmission and EV
with a single-gear transmission.
Figure 2.19 shows the tractive effort of a gasoline engine vehicle with a
four-gear transmission and that of an EV with a single-gear transmission. It is
clear that electric machines with favorable torque–speed characteristics can
satisfy the tractive effort with a simple single-gear transmission.
2.6.3.1
Hydrodynamic Transmission
Hydrodynamic transmissions use fluid to transmit power in the form of
torque and speed. Hydrodynamic transmissions are widely used in passenger
cars. They consist of a torque converter and an automatic gearbox. The torque
converter consists of at least three rotary elements known as the impeller
(pump), the turbine, and the reactor, as shown in Figure 2.20. The impeller
is connected to the engine shaft and the turbine is connected to the output
shaft of the converter, which in turn is coupled with the input shaft of the
multispeed gearbox. The reactor is coupled to external housing to provide a
reaction on the fluid circulating in the converter. The function of the reactor is
to enable the turbine to develop output torque higher than the input torque
of the converter, thus obtaining torque multiplication. The reactor is usually
mounted on a free wheel (one-way clutch) so that when the starting period
has been completed and the turbine speed is approaching that of the pump,
the reactor is in free rotation. At this point, the converter operates as a fluid
coupler with a 1-to-1 ratio of output torque to input torque.
The major advantages of hydrodynamic transmission may be summarized
as follows:
•
When properly matched, the engine will not stall.
•
It provides flexible coupling between the engine and the driven
wheels.
39
Fundamentals of Vehicle Propulsion and Brake
Impeller
(pump)
Turbine
One way
clutch
Reactor
Output shaft
FIGURE 2.20 Schematic view of a torque converter.
•
Together with a suitably selected multispeed gearbox, it provides
torque–speed characteristics that approach the ideal.
The major disadvantages of hydrodynamic transmission are its low efficiency in a stop–go driving pattern and complex structure.
The performance characteristics of a torque converter are described in terms
of the following four parameters:
Speed ratio Csr =
output_speed
,
input_speed
(2.38)
which is the reciprocal of the gear ratio mentioned before.
Torque ratio Ctr =
Efficiency ηc =
output_torque
.
input_torque
output_speed × output_torque
= Csr Ctr .
input_speed × input_torque
speed
.
Capacity factor (size factor) Ktc = torque
(2.39)
(2.40)
(2.41)
The capacity factor, Kc , is an indicator of the ability of the converter to absorb
or transmit torque, which is closely related to the size and geometric shape
of the blades.
Typical performance characteristics of the torque converter are shown in
Figure 2.21, in which torque ratio, efficiency, and input capacity factor (the
ratio of input speed to the square root of input torque) are plotted against
speed ratio. Torque ratio has the maximum value at stall condition where the
40
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
rpm
(Nm)½
Torque ratio
200
1.5
150
75
100
50
1.0
Efficiency
%
100
Efficiency
2.0
Capacity factor
Torque ratio
250
Ktc Input
0.5
50
0
0
25
0
0.2
0.4
0.6
Speed ratio
0.8
1
0
FIGURE 2.21 Performance characteristics of a torque converter.
output speed is zero. The torque ratio decreases as the speed ratio increases
(gear ratio decreases) and the converter eventually acts as a hydraulic coupling with a torque ratio of 1.0. At this point, a small difference between
the input and output speed exists because of the slip between the impeller
(pump) and the turbine. The efficiency of the torque converter is zero at stall
condition (zero speed ratio) and increases with an increase in the speed ratio.
It reaches the maximum when the converter acts as a fluid coupling (torque
ratio equal to 1.0).
To determine the actual operating condition of the torque converter, the
engine operating point has to be specified since the engine directly drives
the torque converter. To characterize the engine operating condition for the
purpose of determining the combined performance of the engine and the
converter, an engine capacity factor, Ke , is introduced and defined as
ne
Ke = √ ,
Te
(2.42)
where ne and Te are engine speed and torque, respectively. The variation of
the capacity factor with speed for a typical engine is shown in Figure 2.22. To
achieve proper matching, the engine and the torque converter should have a
similar range in the capacity factor.
As mentioned above, the engine shaft is usually directly connected to the
input shaft of the torque converter. That is,
Ke = Ktc .
(2.43)
41
Fundamentals of Vehicle Propulsion and Brake
450
400
Engine capacity factor, Ke
225
350
200
300
175
Engine
torque
250
150
200
125
150
100
100
1000
2000
3000
Engine speed (rpm)
4000
Engine capacity factor
Engine torque (Nm)
250
50
5000
FIGURE 2.22 Capacity factor of a typical engine.
The matching procedure begins by specifying engine speed and engine
torque. Knowing the engine operating point, one can determine the engine
capacity factor, Ke , using Equation 2.42 (see Figure 2.22). Since Ke = Ktc , the
input capacity factor of the torque converter corresponding to the specific
engine operating point is then known. As shown in Figure 2.21, for a particular value of the input capacity factor of the torque converter, Ktc , the converter
speed ratio, Csr , and the torque ratio, Ctr , can be determined from the torque
converter performance characteristics as shown in Figure 2.21. The output
torque and the output speed of the converter are then given by
Ttc = Te Ctr
(2.44)
ntc = ne Csr ,
(2.45)
and
where Ttc and ntc are the output torque and the output speed of the converter,
respectively.
Since the torque converter has a limited torque ratio range (usually less
than 2), a multispeed gearbox is usually connected to it. The gearbox comprises several planetary gear sets and is automatically shifted. With the gear
ratios of the gearbox, the tractive effort and speed of the vehicle can be
calculated (see Equations 2.27 and 2.32) by
Ft =
Te Ctr ig i0 ηt
r
(2.46)
42
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
and
V=
πne Csr r
ne Csr r
(m/s) = 0.377
(km/h).
30ig i0
it
(2.47)
Figure 2.23 shows the variation of the tractive effort with speed for a
passenger car equipped with a torque converter and a three-speed gearbox.
2.6.3.2
Continuously Variable Transmission
A continuously variable transmission (CVT) has a gear ratio that can be
varied continuously within a certain range, thus providing an infinite number of gears. The continuous variation allows for matching virtually any
engine speed and torque to any wheel speed and torque. It is therefore possible
to achieve an ideal torque–speed profile (constant power profile).
The commonly used CVT in automobiles uses a pulley and belt assembly.
One pulley is connected to the engine shaft, while the other is connected to
the output shaft. The belt links the two pulleys. The distance between the two
half pulleys can be varied, thus varying the effective diameter on which the
belt grips. The transmission ratio is a function of the two effective diameters:
ig =
D2
,
D1
(2.48)
where D1 and D2 are the effective diameters of the output pulley and the
input pulley, respectively.
Until recently, this implementation was affected by the limited belt–pulley
adhesive contact. The design has been improved by the use of metallic belts
that provide better solidity and improved contact. Furthermore, an interesting concept has been developed and is being used by Nissan. This concept
Tractive effort (kN)
8
6
Low
4
Intermediate
Direct
2
0
0
50
100
150
200
250
Vehicle speed (km/h)
FIGURE 2.23 Tractive effort–speed characteristics of a passenger car with automatic
transmission.
43
Fundamentals of Vehicle Propulsion and Brake
uses three friction gears: one is connected to the engine shaft, another to the
output shaft, and the third grips on the particular profile of the other two
gears. It can be rotated to grip on different effective diameters, thus achieving
variable gear ratio.
2.7 Vehicle Performance
The performance of a vehicle is usually described by its maximum cruising
speed, gradeability, and acceleration. The prediction of vehicle performance
is based on the relationship between tractive effort and vehicle speed discussed in Sections 2.5 and 2.6. For on-road vehicles, it is assumed that the
maximum tractive effort is limited by the maximum torque of the power
plant, rather than the road adhesion capability. Depicted tractive effort
(Equation 2.29 or 2.46) and resistance (Fr + Fw + Fg ) on a diagram are used
for vehicle performance analysis, as shown in Figures 2.24 and 2.25, for
a gasoline-engine-powered, four-gear manual transmission vehicle and an
electric-motor-powered, single-gear transmission vehicle, respectively.
2.7.1
Maximum Speed of a Vehicle
The maximum speed of a vehicle is defined as the constant cruising speed
that the vehicle can develop with full power plant load (full throttle of engine
or full power of motor) on a flat road. The maximum speed of the vehicle is
1st gear
Tractive effort and resistance (kN)
8
Tractive effort
Resistance on grade
30° (57.7%)
7
25° (46.6%)
6
20° (36.4%)
2nd gear
15° (26.8%)
5
4
10° (17.6%)
3rd gear
5° (8.7%)
4th gear
0° (0%)
3
2
1
0
Fr + Fw + Fg
0
50
100
150
Speed (km/h) Maximum
speed
200
FIGURE 2.24 Tractive effort of a gasoline-engine-powered vehicle with manual multispeed
transmission and its resistance.
44
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
7
Tractive effort
Resistance on grade
25° (46.6%)
Tractive effort (kN)
6
20° (36.4%)
5
15° (26.8%)
4
10° (17.6%)
3
5° (8.7%)
0° (0%)
2
1
0
Fr + Fw + Fg
0
50
100
150
speed (km/h) Maximum
speed
FIGURE 2.25 Tractive effort of an electric-motor-powered vehicle with single-speed transmission and its resistance.
determined by the equilibrium between the tractive effort of the vehicle and
the resistance or the maximum speed of the power plant and gear ratios
of the transmission. The tractive effort and resistance equilibrium can be
expressed as
Tp ig i0 ηt
1
= Mg fr cos α + ρa CD Af V 2 .
(2.49)
rd
2
This equation indicates that the vehicle reaches its maximum speed when the
tractive effort, represented by the left-hand-side term in Equation 2.49, equals
the resistance, represented by the right-hand-side terms. The intersection of
the tractive effort curve and the resistance curve is the maximum speed of the
vehicle, as shown in Figures 2.24 and 2.25.
It should be noted that for some vehicles, no intersection exists between
the tractive effort curve and the resistance curve, because of a large power
plant or a large gear ratio. In this case, the maximum speed of the vehicle is
determined by the maximum speed of the power plant. Using Equation 2.32
or 2.47, the maximum speed of the vehicle can be obtained by
Vmax =
πnp max rd
(m/s),
30i0 ig min
(2.50)
where np max and ig min are the maximum speed of the engine (or electric
motor) and the minimum gear ratio of the transmission, respectively.
2.7.2
Gradeability
Gradeability is usually defined as the grade (or grade angle) that the vehicle
can overcome at a certain constant speed, for instance the grade at a speed of
45
Fundamentals of Vehicle Propulsion and Brake
100 km/h (60 mph). For heavy commercial vehicles or off-road vehicles, the
gradeability is usually defined as the maximum grade or grade angle that the
vehicle can overcome in the whole speed range.
When the vehicle drives on a road with relative small grade and constant
speed, the tractive effort and resistance equilibrium can be written as
Tp i0 ig ηt
1
= Mg fr + ρa CD Af V 2 + Mgi.
rd
2
(2.51)
Thus,
i=
Tp i0 ig ηt /rd − Mg fr − 1/2 ρa CD Af V 2
= d − fr ,
Mg
(2.52)
Tp i0 ig ηt /rd − 1/2 ρa CD Af V 2
Ft − Fw
=
Mg
Mg
(2.53)
where
d=
is called the performance factor. When the vehicle drives on a road with a
large grade, the gradeability of the vehicle can be calculated as
d − fr 1 − d2 + fr2
.
sin α =
1 + fr2
(2.54)
The gradeability of the vehicle can also be obtained from the diagram in
Figure 2.24 or 2.25, in which the tractive effort and resistance are plotted.
2.7.3 Acceleration Performance
The acceleration performance of a vehicle is usually described by its acceleration time and distance covered from zero speed to a certain high speed (0 to
96 km/h or 60 mph, for example) on level ground. Using Newton’s second
law (Equation 2.13), the acceleration of the vehicle can be written as
a=
=
Ft − Ff − Fw
dV
=
dt
Mδ
Tp i0 ig ηt /rd − Mg fr − 1/2 ρa CD Af V 2
g
= (d − fr ),
Mδ
δ
(2.55)
where δ is called rotational inertia factor, considering the equivalent mass
increase due to the angular moments of the rotating components. The mass
factor can be written as
δ=1+
Iw
Mrd2
+
i02 ig2 Ip
Mr2
,
(2.56)
46
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
4
Acceleration time (m/s2)
1st gear
3
2nd gear
2
3rd gear
1
0
4th gear
0
50
100
150
Vehicle speed (km/h)
200
FIGURE 2.26 Acceleration rate of a gasoline-engine-powered vehicle with four-gear
transmission.
where Iw is the total angular inertial moment of the wheels and Ip is the
total angular inertial moment of the rotating components associated with the
power plant. Calculation of the mass factor, δ, requires knowing the values
of the mass moments of inertia of all the rotating parts. In the case where
these values are not known, the rotational inertia factor, δ, for a passenger car
would be estimated using the following empirical relation:
δ = 1 + δ1 + δ2 ig2 i02 ,
(2.57)
where δ1 represents the second term on the right-hand side of Equation 2.56,
with a reasonable estimate value of 0.04, and δ2 represents the effect of the
power-plant-associated rotating parts, with a reasonable estimate value of
0.0025.
Figures 2.26 and 2.27 show the acceleration rate along with vehicle speed
for a gasoline-engine-powered vehicle with a four-gear transmission and an
electric-motor-powered vehicle with a single-gear transmission.
From Equation 2.55, the acceleration time, ta , and distance, Sa , from lowspeed V1 to high-speed V2 can be written, respectively, as
ta =
and
V1
Sa =
V2
V2
V1
Mδ
dV
Tp ig i0 ηt /rd − Mg fr − 1/2ρa CD Af V 2
(2.58)
MδV
dV.
Tp ig i0 ηt /r0 − Mg fr − 1/2 ρa CD Af V 2
(2.59)
In Equations 2.58 and 2.59, the torque of the power plant Tp is a function of
speed (see Figures 2.13 and 2.14), which in turn is a function of vehicle speed
47
Fundamentals of Vehicle Propulsion and Brake
4
Acceleration (m/s2)
3
2
1
0
0
50
100
150
200
Vehicle speed (km/h)
30
600
25
500
20
400
15
300
Time
200
10
5
0
Distance
0
50
100
Vehicle speed (km/h)
100
Acceleration distance (m)
Acceleration time (s)
FIGURE 2.27 Acceleration rate of an electric-machine-powered vehicle with a single-gear
transmission.
0
150
25
500
20
400
300
15
Time
200
10
100
5
Distance
0
0
50
100
Vehicle speed (km/h)
0
150
Acceleration distance (m)
Acceleration time (s)
FIGURE 2.28 Acceleration time and distance along with vehicle speed for a gasoline-enginepowered passenger car with four-gear transmission.
FIGURE 2.29 Acceleration time and distance along with vehicle speed for an electric-machinepowered passenger car with single-gear transmission.
48
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
(see Equations 2.23 and 2.37) and gear ratio of the transmission. This makes
it difficult to solve Equations 2.58 and 2.59 analytically. Numeral methods
are usually used. Figures 2.28 and 2.29 show the acceleration time and distance along with vehicle speed for a gasoline-engine-powered vehicle and an
electric-machine-powered EV, respectively.
2.8
Operating Fuel Economy
The fuel economy of a vehicle is evaluated by the amount of fuel consumption
per 100 km traveling distance (liters/100 km) or mileage per gallon fuel consumption (miles/gallon), which is currently used in the United States. The
operating fuel economy of a vehicle depends on a number of factors, including fuel consumption characteristics of the engine, gear number and ratios,
vehicle resistance, vehicle speed, and traffic conditions.
2.8.1
Fuel Economy Characteristics of IC Engines
The fuel economy characteristic of an IC engine is evaluated by the amount of
fuel per kWh energy output, which is referred to as the specific fuel consumption (g/kWh). The typical fuel economy characteristic of a gasoline engine is
shown in Figure 2.30. The fuel consumption is quite different from one operating point to another. The optimum operating points are close to the points
of full load (wide open throttle). The speed of the engine also has a significant
influence on the fuel economy. With a given power output, the fuel consumption is usually lower at low speed than at high speed. For instance, when
Maximum engine power
Optimum
operation
line
60
*5
25
26
5
28
80
5
100
0
32 50
3
0
40
40
20
0
1000
0
50
600
700 0
80
Engine specific fuel
consumption (g/kWh)
2000
3000
4000
Engine speed (rpm)
5000
FIGURE 2.30 Fuel economy characteristics of a typical gasoline engine.
Fundamentals of Vehicle Propulsion and Brake
49
the engine shown in Figure 2.30 has a power output of 40 kW, its minimum
specific fuel consumption would be 270 g/kWh, at a speed of 2080 rpm.
For a given power output at a given vehicle speed, the engine operating
point is determined by the gear ratio of the transmission (refer to Equations
2.32 and 2.47). Ideally, a continuous variable transmission can choose the gear
ratio, in a given driving condition, to operate the engine at its optimum operating point. This advantage has stimulated the development of a variety of
continuous variable transmissions, including frictional drive, hydrodynamic
drive, hydrostatic drive, and hydro-mechanical variable drive.
2.8.2
Computation of Vehicle Fuel Economy
Vehicle fuel economy can be calculated by finding the load power and speed,
and thus the specific fuel consumption of the engine. The engine power output is always equal to the resistance power plus the dynamic power for
acceleration of the vehicle, that is,
Pe =
V
dV
Ff + F w + F g + M v δ
.
ηt
dt
(2.60)
Equation (2.60) can be written as
V
1
dV
2
Pe =
Mg fr cos α + ρa CD Af V + Mg sin α + Mδ
(kW). (2.61)
1000ηt
2
dt
The engine speed, related to vehicle speed and gear ratio, can be expressed as
Ne =
30Vig i0
.
πrd
(2.62)
After determination of the engine power and speed by Equations 2.60 and
2.61, the value of the specific fuel consumption, ge , can be found in the graph
of the engine fuel economy characteristics as shown in Figure 2.30. The time
rate of fuel consumption can be calculated by
Qfr =
Pe ge
(l/h),
1000γf
(2.63)
where ge is the specific fuel consumption of the engine in g/kWh and γf is
the mass density of the fuel in kg/L. The total fuel consumption within the
total distance, S, at a constant cruising speed, V, is obtained by
Qs =
Pe ge S
.
1000γf V
(2.64)
50
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Highest gear
2nd highest gear
Fuel consumption
L/100 km
30
25
mpg
60
50
20
L/100 km
15
40
30
10
20
mpg
5
10
0
50
0
150
100
Vehicle speed (km/h)
FIGURE 2.31 Fuel economy characteristics of a typical vehicle at constant speed.
Figure 2.31 shows an example of the fuel economy characteristics of a gasoline
vehicle at a constant cruising speed on level ground. This figure indicates that
with high speed, the fuel consumption increases because the aerodynamic
resistance power increases with the speed cubed. This figure also indicates
that with a high-speed gear (small gear ratio), the fuel economy of the vehicle
can be enhanced due to the reduced engine speed at a given vehicle speed
and decreased gear ratio.
Figure 2.32 shows the operating points of the engine at a constant vehicle
speed, with the highest gear and the second highest gear. It indicates that the
engine has a much lower operating efficiency in low gear than in high gear.
This is the reason why the fuel economy of a vehicle can be improved with
more gear transmission and continuous variable transmission.
Maximum engine
power
40
30
20
10
0
1000
Highest
gear
5
0
32
50
5
25
5
60
Optimum
economy
line
26
70
28
Engine power output (kW)
80
0
35
0
40
0
50
0
60 0
70 00
8
Specific fuel
consumption
2nd highest
g/kWh
gear
2000
3000
4000
5000
Engine speed (rpm)
FIGURE 2.32 Operating point of the engine at a constant speed with highest gear and second
highest gear.
51
Speed (km/h)
Speed (km/h)
Fundamentals of Vehicle Propulsion and Brake
100
Urban driving
50
0
0
200
400
600
800
1000
1200
1400
100
50
0
Highway
driving
0
100
200
300
400
500
600
700
800
Driving time (s)
FIGURE 2.33 EPA FTP75 urban and highway drive cycles.
It should be noted that because of the complexity of vehicle operation in the
real world, fuel consumption at a constant speed (as shown in Figure 2.12)
cannot accurately represent fuel consumption for a vehicle under real driving
conditions. Thus, various drive cycles have been developed to simulate real
driving conditions, such as EPA FTP75 urban and highway, LA92, ECE-15,
Japan1015, and so on. The drive cycles are usually represented by the speed
of the vehicle along with the driving time. Figure 2.33 shows the urban and
highway drive cycles of EPA FTP75 used in the United States.
To calculate fuel consumption in a drive cycle, the total fuel consumption can be obtained by the summation of fuel consumption in each time
interval, Δti ,
Pei gei
Δti
(2.65)
Qtc =
1000γf
i
where Pei is the average power of the engine in the ith time interval in kW,
gei is the average specific fuel consumption of the engine in the ith time interval in g/kWh, and Δti is the ith time interval in h. This calculation can be
performed with a numerical method using a computer program. Figures 2.34
and 2.35 show examples of the fuel economy and engine operating points in
EPA FTP75 urban and highway drive cycle, respectively.
2.8.3
Basic Techniques to Improve Vehicle Fuel Economy
The effort to improve the fuel economy of vehicles has always been an ongoing process in the automobile industry. Fundamentally, the techniques used
mainly include the following aspects:
1. Reducing vehicle resistance: Using light materials and advanced manufacturing technologies can reduce the weight of vehicles, in turn
52
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Fuel economy:
12.8 L/100 km (18.5 mpg)
Maximum engine
power
Optimum
operating
line
Actual
60 operating
points
5
25
26
5
28
5
32
0
Engine power output (kW)
80
40
350 0
40
20
500
600
700
800
Specific fuel
consumption (g/kWh)
0
1000
2000
3000
4000
Engine speed (rpm)
5000
6000
FIGURE 2.34 Fuel economy and engine operating points in EPA FTP75 urban drive cycle
overlapped on engine fuel consumption characteristics map.
Engine power output (kW)
reducing the rolling resistance and inertial resistance in acceleration, and therefore reducing the demanded power on the engine. The
use of advanced technologies in tire production is another important
method in reducing the rolling resistance of vehicles. For instance, steel
wire plied radial tires have a much lower rolling resistance coefficient
than conventional bias ply tires. Reducing aerodynamic resistance is
also quite important at high speeds. This can be achieved by using
a flow-shaped body style, a smooth body surface, and other techniques. Furthermore, improving transmission efficiency can reduce
energy losses in the transmission. Proper transmission construction,
good lubrication, proper adjustment and tightening of moving parts
in the transmission, and so on will achieve this purpose.
Fuel economy:
10.5 L/100 km (22.5 mpg)
80
Optimum
Actual operating
line
60 operating
points
Maximum engine
power
5
25
40
5
26 285
0
32 50
3
0
40
0
50
20
0
60 00
7 0
80
Specific fuel
consumption (g/kWh)
0
1000
2000
3000
4000
5000
6000
Engine speed (rpm)
FIGURE 2.35 Fuel economy and engine operating points in EPA FTP75 highway drive cycle
overlapped on engine fuel consumption characteristics map.
Fundamentals of Vehicle Propulsion and Brake
53
2. Improving engine operation efficiency: Improving engine operation efficiency has great potential to contribute to the improvement of vehicle fuel economy. There are many effective advanced techniques,
such as accurate air/fuel ratio control with computer-controlled fuel
injection, high thermal isolated materials for reducing thermal loss,
varying ignition-timing techniques, active controlled valve and port,
and so on.
3. Properly matched transmission: Parameters of the transmission, especially gear number and gear ratios, have much influence on operating
fuel economy as described previously. In the design of the transmission, the parameters should be constructed so that the engine will
operate close to its fuel optimum region.
4. Advanced drive trains: Advanced drive trains developed in recent years,
such as new power plants, various hybrid drive trains, etc., can greatly
improve the fuel economy of vehicles. Fuel cells have higher efficiency
and lower emissions than conventional IC engines. Hybridization of
a conventional combustion engine with an advanced electric motor
drive may greatly enhance the overall efficiency of vehicles.
2.9 Brake Performance
The braking performance of a vehicle is undoubtedly one of the most important concerns that affect vehicle safety. In urban area driving, a significant
amount of energy is consumed in braking. In recent years, more and more
electric drives have been involved in vehicle traction, such as EVs, HEVs,
and fuel-cell-powered vehicles; the electrification of the vehicle drive train
makes it feasible to recover some of the energy lost in braking. Nevertheless,
braking performance is still the first concern in the design of the vehicle brake system. Actually, when electric braking is introduced for braking
energy recovery, mechanical braking by using frication is still required to
ensure that the vehicle is stopped quickly. Consequently, a hybrid braking system is established. The design and control objectives of such a hybrid braking
system are (1) sufficient braking force to quickly reduce the vehicle speed; (2)
proper braking force distribution on the front and rear wheels to ensure vehicle stability during braking; and (3) recovery of as much braking energy as
possible. This chapter discusses only the design principle of the vehicle brake
system from the braking performance point of view. Regenerative braking
will be discussed in Chapter 13.
2.9.1
Braking Force
The function of the vehicle brake system is to quickly reduce the vehicle
speed while keeping the vehicle traveling direction stable and controllable
54
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
(a)
(b)
P
Fb
Tb
w
V
Fb max
O
rd
Fb
Tb
W
FIGURE 2.36 (a) Braking torque and braking force; (b) relationship between braking torque
and braking force.
under various road conditions. These requirements are satisfied by applying
sufficient braking force on the wheels and properly allocate the total braking
force on the front and rear wheels.
Figure 2.36a shows a wheel during braking. The brake pad is pressed against
the brake plate hydraulically or pneumatically, thus developing a frictional
torque on the brake plate. This braking torque results in a braking force in
the tire–ground contact area. It is just this braking force that tries to stop the
vehicle. The braking force can be expressed as
Fb =
Tb
.
rd
(2.66)
The braking force increases with an increase in the braking torque. However, when the braking force reaches the maximum braking force that the
tire–ground adhesion can support, it will not increase further, although the
braking torque may still increase as shown in Figure 2.36b. This maximum
braking force limited by the adhesive capability can be expressed as
Fb max = μW,
(2.67)
where μ is the adhesive coefficient of the tire–ground contact. Similar to the
traction case, the adhesive coefficient varies with the slipping of the tire as
shown in Figure 2.37. However, the slip is defined in braking as
rω s= 1−
× 100%,
V
(2.68)
55
Fundamentals of Vehicle Propulsion and Brake
B
Adhesive coefficient (m)
O
Longitudinal
A
mp
Lateral
0
15-20
50
ms
100%
Slip
FIGURE 2.37 Variation of tractive effort coefficient with longitudinal slip of a tire.
where V is the vehicle translatory speed, ω is the wheel rotation speed, and r
is the wheel radius. In this definition, when ω = 0, that is, the wheel is completely locked, s = 100%. Figure 2.37 shows the typical relationship between
adhesive coefficient and wheel slip. There exists a maximum value in the slip
range of 15–20%, and somewhat declining at 100% slip. Table 2.3 shows the
average values of tractive effort coefficients on various roads.2
2.9.2
Braking Distribution on Front and Rear Axles
Figure 2.38 shows the forces acting on a vehicle during braking on a flat road.
Rolling resistance and aerodynamic drag are ignored in this figure, because
they are quite small compared to the braking forces. j is the deceleration of
TABLE 2.3
Average Values of Tractive Effort Coefficient on Various Roads
Surface
Asphalt and concrete (dry)
Concrete (wet)
Asphalt (wet)
Gravel
Earth road (dry)
Earth road (wet)
Snow (hard packed)
Ice
Peaking Values, μp
0.8–0.9
0.8
0.5–0.7
0.6
0.68
0.55
0.2
0.1
Slipping Values, μs
0.75
0.7
0.45–0.6
0.55
0.65
0.4–0.5
0.15
0.07
56
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Mj
hg
Mg
Fbf A
Lb
La
Wf
Fbr
L
B
Wr
FIGURE 2.38 Force acting on a vehicle during braking on a flat road.
the vehicle during braking, which can be easily expressed as
j=
Fbf + Fbr
,
M
(2.69)
where Fbf and Fbr are the braking forces acting on the front and rear wheels,
respectively.
The maximum braking force is limited by the tire–ground adhesion and is
proportional to the normal load acting on the tire. The actual braking force
developed by the brake torque should also be proportional to the normal
load so that both the front and the rear wheels obtain their maximum braking
force at the same time. During braking, there is load transfer from the rear
axle to the front axle. By considering the equilibrium of moments about the
front and rear tire–ground contact points A and B, as shown in Figure 2.38, the
normal loads on the front and rear axles Wf and Wr , with a vehicle deceleration
rate, j, can be expressed as
j
Mg
Wf =
Lb + h g
(2.70)
L
g
and
Wr =
j
Mg
La − h g
.
L
g
(2.71)
The braking forces applied on the front axle and the rear axle should be
proportional to their normal load, respectively; thus, one obtains
Lb + hg j/g
Fbf
Wf
=
=
.
Fbr
Wr
La − hg j/g
(2.72)
Combining Equations 2.69 and 2.72, the ideal braking force distribution on the
front and rear axles can be obtained as shown in Figure 2.39. When the braking is strong, both the front and rear wheels obtain their maximum ground
57
Rear axle braking force ratio to vehicle weight (Fbr/Mg)
Fundamentals of Vehicle Propulsion and Brake
1
0.9
0.8
0.7
j=
0.6
j=
j=
0.5
j=
j=
0.4
j=
0.3
j=
0.5
g,
g,
0.4
g,
m=
0.3
g
0.4
0.2 , m =
g,
0
.
m=
3
j=
0.2
0.1 m = 0.1g,
0.1
j=
0.2
m=
m=
g,
0.8
g,
0.7
g,
0.6
0.9
m=
m=
m=
Ideal braking force
distribution curve
j = 0.9g, m = 0.9
j = 0.8g, m = 0.8
j = 0.7g, m = 0.7
j = 0.6g, m = 0.6
j = 0.5g, m = 0.5
j = 0.4g, m = 0.4
j = 0.3g, m = 0.3
j = 0.2g, m = 0.2
j = 0.1g, m = 0.1
0.9
0.8
0.7
0.6
0.5
0
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
Front axle braking force ratio to vehicle weight (Fbf /Mg)
FIGURE 2.39 Ideal braking force distribution curve on the front and rear axles.
braking force, which is limited by the capability of the tire–ground adhesion
[wheel lock for non-antilock brake system (ABS), or ABS starting to function
to limit the braking force rising to avoid the wheels being locked]. In this case,
the vehicle achieves its maximum deceleration rate as
jmax = Fbf- max + Fbr- max = (Wf + Wr )μ = gμ.
μ
M
M
(2.73)
The ideal braking force distribution curve (simply, I curve), as shown in
Figure 2.39, is a nonlinear hyperbolic curve. If it is desired for the front and
rear wheels to lock or the ABS to function at the same time on any road, the
braking force on the front and rear axles must exactly follow this curve.
Completely following the I curve for the braking force distribution makes
the system very complex in structure and control. However, with rapid
advancement of electronics and microcontrol technologies, electric braking
systems (EBSs) are being developed, which can greatly improve the braking performance compared with the traditional design used in most vehicles
at the present. This technology will be briefly described in the regenerative
braking chapter (Chapter 13).
Traditionally, the actual braking forces applied to the front and rear axles
by the brake system are usually designed to have a fixed linear proportion.
This proportion is represented by the ratio of the front axle braking force to
58
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
the total braking force of the vehicle, that is,
β=
Fbf
,
Fb
(2.74)
where Fb is the total braking force of the vehicle (Fb = Fbf + Fbr ). β depends
only upon the braking system design, such as the diameters of wheel cylinders in the front and rear wheels, and has nothing to do with the vehicle
parameters. With a value of β, the actual braking forces on the front and rear
axles, produced by the brake system, can be expressed as
Fbf = βFb
(2.75)
Fbr = (1 − β)Fb .
(2.76)
Fbf
β
=
.
Fbr
1−β
(2.77)
and
Thus, one obtains
Figure 2.40 shows the ideal and actual braking force distribution curves
(labeled I and β curve). It is obvious that only one intersection point exists,
at which the front and rear axles lock at the same time. This point represents
Rear axle braking force ratio to vehicle
weight (Fbr / Mg)
1
0.9
0.8
0.7
m
0.6
0.5
=
0.8
0.4
b line
0.3
0.2
l curve
0.1
0
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
Front axle braking force ratio to vehicle weight (Fbf /Mg)
FIGURE 2.40 Ideal and actual braking force distribution curves.
1
Fundamentals of Vehicle Propulsion and Brake
59
one specific road adhesive coefficient, μ0 . Referring to Equation 2.72 in which
j/g is replaced by μ0 and Equation 2.77, one obtains
L b + μ 0 hg
β
.
=
1−β
L a − μ 0 hg
(2.78)
From Equation (2.78) one can obtain μ0 and β by
Lβ − Lb
hg
(2.79)
μ0 hg + Lb
.
L
(2.80)
μ0 =
and
β=
During braking on the roads with the adhesive coefficient less than μ0 (the
region in which the β curve is below the I curve), the front wheels lock first,
whereas when the road adhesive coefficient is larger than μ0 (the region in
which the β curve is above the I curve), the rear wheels lock first.
When the rear wheels lock first, the vehicle will lose directional stability,
as shown in Figure 2.41, which shows the top view of a two-axle vehicle acted upon by the braking force and the inertia force. When the rear
wheels lock, the capability of the rear tires to resist lateral forces is reduced
to zero (refer to Figure 2.37). If some slight lateral movement of the rear
wheels is initiated by side wind, road camber, or centrifugal force, a yawing moment due to the inertia force about the yaw center of the front
axle will be developed. As the yaw motion progresses, the moment arm of
the inertia force increases, resulting in an increase in yaw acceleration. As
the rear end of the vehicle swings around 90◦ , the moment arm gradually
decreases and eventually the vehicle rotates 180◦ with the rear end leading
the front end.
o
O--yaw center of front axle
FIGURE 2.41 Loss of directional stability due to the lockup of rear wheels.
60
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Clockwise
80
Steering wheels help fixed in straight
ahead position
Initial speed 64.4 km/h, 40 mph
Southbound runs
Northbound runs
60
40
20
Counter clockwise
Angular deviation (deg)
100
0
20
40
0.5
Front wheel
locked first
Rear wheel
locked first
0
0.5
1.0
1.5
Time (s)
FIGURE 2.42 Angular deviation of a car when all four wheels do not lock at the same instant.
The lockup of front wheels will cause a loss of directional control, and
the driver will no longer be able to exercise effective steering. It should be
pointed out, however, that front wheel lockup does not cause directional
instability. This is because whenever the lateral movement of the front wheels
occurs, a self-correcting moment due to the inertial force of the vehicle about
the yaw center of the rear axle will be developed. Consequently, it tends to
bring the vehicle back to a straight line path. Figure 2.42 shows the measured
angular deviation of a vehicle when the front and rear wheels do not lock at
the same instant.2
Loss of steering control may be detected more readily by the driver and
control may be regained by release or partial release of the brakes. Contrary
to the case of front wheel lockup, when rear wheels lock and the angular
deviation of the vehicle exceeds a certain level, control cannot be regained
even by complete release of the brakes and by the most skilful driving. This
suggests that rear wheel lockup is a more critical situation, particularly on a
road with a low adhesive coefficient. Because the value of the braking force is
low on slippery surfaces, the kinetic energy of the vehicle will dissipate at a
low rate and the vehicle will experience a serious loss of directional stability
over a considerable distance. Therefore, designers of vehicle brake systems
must ensure that the rear wheels do not lock first.
The ABS, developed in recent years, can effectively prevent the wheels
from lockup. This system employs speed sensors to detect the wheel rotating
speed. When a wheel lockup is detected, the braking pressure control system
reduces the pressure and brings the wheel back to its rotation.3
Fundamentals of Vehicle Propulsion and Brake
2.9.3
2.9.3.1
61
Braking Regulation and Braking Performance Analysis
Braking Regulation
As described above, if the real braking force distribution line β is below the
ideal braking force distribution curve I as shown in Figure 2.40, the front
wheels will be locked earlier than the rear wheels. This situation leads to
stable behavior of the vehicle. This usually is the design, especially for passenger cars, which run at high speed. However, when the β line is much below
the I curve, most of the braking force will be applied to the front wheel and
a very small force to the rear wheels. This design will cause the problem of
reduced utilization of road adhesive capability. That is, when the front wheels
are locked and the rear wheels are not locked, the maximum braking force
on the rear wheels will never be used. In order to avoid this situation, some
brake design regulations have been developed. A typical one is ECE brake
regulation.
The ECE brake design regulation for passenger cars is expressed by
Fbf
Fbr
≥
.
Wf
Wr
(2.81)
Equation 2.81 shows that the rear wheels are never locked before the front
wheels. In other words, the real braking force distribution curve β is always
below the I curve. ECE also dictates the minimum braking force on rear
wheels, as expressed by
j
≥ 0.1 + 0.85(μ − 0.2),
g
(2.82)
where j is the deceleration rate of the vehicle when the front wheels are
locked on the road with adhesive coefficient, μ. The physical meaning of this
equation is that when the front wheels are locked, the rear braking force must
be large enough to make the vehicle yield a deceleration rate not smaller than
the value dictated by Equation 2.82.
Braking forces of the front and rear wheels on the boundary of ECE
regulation described by Equation 2.82 can be calculated as follows.
ECE regulation stipulates that the condition for Equation 2.81 is the front
wheels being locked. Thus the braking force on the front wheels on the road
with the adhesive coefficient, μ, is
Fbf = Wf μ,
(2.83)
where Wf is the vertical loading on the front wheels, which is expressed at
the vehicle deceleration rate by Equation 2.70, and the total braking force
of the vehicle at the deceleration rate, j, is expressed by Equations 2.69 and
2.82 related to μ. Using all the equations mentioned above, the front and rear
braking force can be calculated as shown in Figure 2.43. It must be noted that
62
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
g
1.0
.8
=0
.9
=0
.0
=1
,m
,m
.7
=0
.6
=0
I
ECE regulation
A
0
0.1
0.2
0.3
0.4
0.5
b
,m
g
0.9
,m
,m
.5
=0
.2
=0
g,
0.1
j = 0.1
m=
.3
=0
.4
=0
,m
,m
0.05
0
g
0.8
g
0.7
,m
,m
g
0.3
g
0.2
0.15
g
0.5
g
0.4
j=
0.2
g
0.6
j=
j=
0.25
j=
j=
j=
0.3
0.1
j=
j=
0.35
j=
The ratio of braking force on rear wheels to total
vehicle weight (Fbr/Mg)
0.4
0.6
0.7
0.8
0.9
1
The ratio of braking force on front wheels to total
vehicle weight (Fbf / Mg)
FIGURE 2.43 Minimum braking force on rear wheels stipulated by ECE regulation.
in Figure 2.43, the front and rear wheel braking forces on the ECE regulation
curve at a deceleration rate j (point A with j = 0.6g, for example) does not
mean that the road adhesive coefficient is μ = 0.6, but larger than it, due to
the unlocked rear wheels.
Obviously, the real braking force distribution achieved by brake system
design must fall into the area between the I curve and the ECE regulation
curve.
2.9.3.2
Braking Performance Analysis
As mentioned above, the vehicle with the traditional brake system design
of a straight line real braking force distribution locks up its front and rear
wheels simultaneously only on one kind of road with adhesive coefficient
μ0 . On other roads, the front or the rear wheels will lock up before the
other one. For fully understanding the braking force scenarios on the front
and rear wheels after the front or rear wheels lock, further analysis is introduced. This is helpful for the design of advanced braking systems for electric,
hybrid electric, and fuel cell vehicles, which not only need to meet the braking
performance requirements, but also are capable of recapturing braking energy
as much as possible.
1. The case of locked front wheels and unlocked rear wheels: When the
front wheels are locked, the braking force on them is expressed as
63
Fundamentals of Vehicle Propulsion and Brake
Equation 2.83. With the vehicle deceleration rate, j, the vertical load
on the front wheels is expressed as Equation 2.70. Thus, the braking
force on the front wheels can be expressed as
Fbf
Mgμ
j
=
L b + hg .
L
g
(2.84)
Since
Fbf + Fbr = Mj,
thus
Fbf =
(2.85)
Mgμ
Fbf + Fbr
Lb +
hg .
L
Mg
(2.86)
L − μhg
MgLb
Fbf −
.
μhg
hg
(2.87)
Finally, we obtain
Fbr =
With different road adhesive coefficient, μ, Equation 2.87 generates a
group of lines (referred to as f lines) to represent the relationship of
the braking forces on the front and rear wheels when the front wheels
are locked and the rear wheels are not locked, as shown in Figure 2.44.
2. The case of locked rear wheels and unlocked front wheels: Similarly, when
the rear wheels are locked and the front wheels are not, the braking
force on the rear wheels against the braking force on the front wheels
on roads with different adhesive coefficients can be expressed as
Fbr =
−μhg
μMgLa
Fbf +
.
L + μhg
L + μhg
(2.88)
The lines generated by Equation 2.88 on roads with different adhesive
coefficients are referred to as r lines as shown in Figure 2.44.
3. Braking process analysis: Using the diagram in which the I curve, the β
line, the ECE regulation curve, the f lines, and the r lines are plotted,
as shown in Figure 2.44, the detailed braking process can be analyzed
as follows:
(a) On a road with μ < μ0 (μ = 0.5 and μ0 = 0.8 in Figure 2.44 as the
analyzing case) On a road with μ < μ0 , the braking forces on
the front and rear wheels are increased, as the brake pedal is
depressed, along the real braking distribution line β, until point
a: the intersection point of the β line and the f line with μ = 0.5,
where the front wheels are locked, but the rear wheels are not.
Further depressing the brake pedal will cause faster increase in
64
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
f-lines
r-lines
m=
g
0.2
0.1
m = 0.3
g
0.1
m = 0.4
j=
0
a
t
0.05 0.1
0
0.1
d
m=
0.9
r
I
e
m = 1.0
m = 0.9
c
m = 0.8
b
m
0. 2
0.8
m = 0.7
m = 0.4
m = 0.3
m=
f
m = 0.5
g
.5
m=0
.7
m = 0.6
0 .2
m=0
0.15
0.1
m=0
.6
j=
0.2
b
g
0 .3
0.25
1 .0
g
j=
0.3
j=
g
0 .9
0 .8
j=
j=
g
g
0 .7
g
0 .6
j=
j=
g
0 .5
0 .4
0.35
j=
j=
Ratio of braking force on rear wheels to total vehicle
weight (Fbr /Mg)
0.4
g
ECE-regulation
s
n
k
p
u
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
Ratio of braking force on front wheels to total vehicle
weight (Fbf /Mg)
0.9
FIGURE 2.44 Braking process analysis on roads with different adhesive coefficients.
the braking force on the rear wheels and slow increase in the
braking force on the front wheels, along the f line of μ = 0.5,
until we reach point b, where the rear wheels are also locked and
the vehicle achieves its maximum deceleration of j = μg = 0.5g.
This case will not cause vehicle instability.
(b) On a road with μ > μ0 (μ = 1.0 and μ0 = 0.8 in Figure 2.44 as the
analyzing case) Similarly, when the brake pedal is depressed,
the braking forces on the front and rear wheels rise along the β
line until point d: the intersection point of the β line and the
r line with μ = 1.0, where the rear wheels are locked but the
front wheels are not. Further depressing the brake pedal will
cause the braking forces to develop along the r line with μ = 1.0,
finally reaching point e, where the front wheels are locked and
the vehicle achieves its maximum deceleration rate of j = μg =
1.0g. In this process, the braking force on the rear wheels has
slightly decreased due to the load shifting from the rear to the
front wheels. This case will cause vehicle instability.
(c) On a road with μ = μ0 (μ = 0.8 and μ0 = 0.8 in Figure 2.44 as the
analyzing case) Obviously, the front and rear wheels will be
locked simultaneously at point c, where the vehicle achieves its
1
Fundamentals of Vehicle Propulsion and Brake
65
maximum deceleration rate of j = μg = 0.8g. This is the ideal
case.
4. Maximum available braking force on front wheels: In electric, hybrid electric, and fuel cell passenger cars, electric motors are mostly employed
to drive the front wheels. This means that regenerative braking is only
available for the front wheels. In the braking system design and control
(mechanical and electrical), more braking energy should be allocated
to the front wheels so as to increase the braking energy that is available
for recovering, under the conditions of meeting brake regulation.
As shown in Figure 2.44, when the commanded braking deceleration rate,
j, is smaller than μg, the braking forces on the front and rear wheels can be
varied in a range rather than a specified point. For example, when μ = 0.7 and
j = 0.6g, this range is between point f and point g specified by the heavy solid
line in Figure 2.44. Obviously, the maximum braking force on the front wheels
is dictated by point g. However, if less deceleration rate is commanded on the
same road, the variation range of the braking force will be larger. For instance,
when j = 0.5g and μ = 0.7, the range is from point b to point p. However, it
violates the ECE regulation, and therefore the maximum braking force on
the front wheels is dictated by point n. Similarly, when μ = 0.5 and j = 0.4g,
the maximum braking on the front wheels is specified by point r, rather than
s, and when j = 0.3 by point u. It is obvious that, with a small difference
between deceleration rate ( g) and road adhesive coefficient, the maximum
braking force is usually dictated by the f lines, and naturally meeting the ECE
regulation. However, when the deceleration rate ( g) is much smaller than
the road adhesive coefficient (e.g., slight braking on a good road), the ECE
regulation will dictate the maximum braking force on the front wheels.
The analysis above provides the basis for design and control of the hybrid
brake system (mechanical plus electrical) of electric, hybrid electric, and
fuel cell vehicles. More details will be discussed in the regenerative braking
chapter.
References
1. J. Y. Wong, Theory of Ground Vehicles, John Wiley & Sons, New York, 1978.
2. Bosch, Automotive Handbook, Robert Bosch GmbH, Karlsruhe, Germany, 2000.
3. S. Mizutani, Car Electronics, Sankaido Co., Minato-Ku, Tokyo, Japan, 1992.
3
Internal Combustion Engines
The IC engine is the most popular power plant for motor vehicles. In the
foreseeable future, it will still be the dominant vehicular power plant. In
HEVs, the IC engine will also be the first selection as a primary power source.
However, the operation of an HEV differs from that of a conventional motor
vehicle. The engine in an HEV runs for a longer time at high power and does
not require its power to be changed frequently. A specifically designed and
controlled engine for hybrid vehicle applications has not been fully developed. In this chapter, the commonly used four-stroke (4S) IC engine and
other types of engines, which can be possibly used in hybrid vehicles, such
as two-stroke (2S) engines, rotary engines, Stirling engines, and gas turbine
engines, are reviewed. For more details, readers may consult the relevant
literature.
3.1 4S, Spark-Ignited IC Engines
3.1.1
Operating Principles
A 4S, spark-ignited (SI), IC engine is illustrated in Figure 3.1. It consists
of subsystems including powering (crankshaft, connection rod, pistons and
cylinders), intake and exhaust (air filter, throttle, inlet and exhaust manifolds,
inlet and exhaust valves, and valve control cams), fuel supply [fuel tank
(not shown), fuel pump (not shown), and fuel injectors], ignition [battery
(not shown), ignition coils (not shown), distributor (not shown), and spark
plugs], cooling [coolant, water pump (not shown), radiator (not shown)], and
lubricating (not shown).
The combustion of the air/fuel mixture formed within the inlet manifold
and trapped in the cylinder produces heat, so that the temperature and pressure in the cylinder increase quickly. Thus, the piston is forced to move down.
The connection rod transfers the linear movement of the piston into rotary
motion of the crankshaft.
A 4S, SI engine has four instinctive processes corresponding to the four
strokes of each piston1,2 (Figure 3.2).
67
68
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Exhaust
valve
Exhaust
manifold
Inlet
manifold
Cylinder
Inlet
valve
Coolant
Piston
Crank case
Connection rod
Crankshaft
Sump
FIGURE 3.1 4S, SI gasoline engine.
1. Induction stroke (cylinder-filling process): The inlet valve is open, the
exhaust valve is closed, and the piston travels down the cylinder,
drawing in a charge of the air/fuel mixture formed within the inlet
manifold.
2. Compression stroke: Both inlet and exhaust valves are closed, and
the piston goes up the cylinder, compressing the fuel/air mixture in
the cylinder. As the piston approaches the top dead center (TDC), the
spark plug produces a spark to ignite the air/fuel mixture.
Induction
Compression
Ignition
FIGURE 3.2 Four strokes of a spark-ignition engine.
Expansion
Exhaust
Internal Combustion Engines
69
3. Expansion stroke (power producing or working process): There is combustion propagation through the charge, raising the temperature and
pressure in the cylinder, moving the piston down. At the end of the
expansion stroke, the exhaust valve opens and irreversible expansion
of the exhaust gases blows out the exhaust valve, which is termed
“blow-down.”
4. Exhaust stroke: The exhaust valve remains open, and as the piston travels up the cylinder, the remaining gases in the cylinder are expelled. At
the end of the exhaust stroke, the exhaust valve closes. However, some
exhaust gas residuals will be left. This exhaust dilutes the next charge.
Following this stroke, the induction stroke of the next cycle starts.
Since the cycle is completed only once every two revolutions of the
crankshaft, the gear driven camshaft (for opening and closing the valves)
has to be driven by the mechanism operating at half crankshaft speed (engine
speed). Some of the power from the expansion stroke is stored in the flywheel
to provide the energy for another three strokes.
3.1.2
3.1.2.1
Operation Parameters
Rating Values of Engines
The most common parameters for engine performance are as follows:
Maximum rated power: The highest power that an engine is allowed to
develop for a short period of operation.
Normal rated power: The highest power that an engine is allowed to
develop in continuous operation.
Rated speed: The rotational speed of the crankshaft, at which the rated
power is developed.
For vehicle application, engine performance is more precisely defined by
the following:
1. The maximum power (or maximum torque) available at each speed
within the useful engine operating range.
2. The range of speed and the power over which engine operation is
satisfactory.
3.1.2.2
Indicated Work per Cycles and Mean Effective Pressure
The torque performance of the 4S SI engine is determined by the pressure
within the cylinder, as shown in Figure 3.3. In the induction stroke (g–h–a),
the pressure in the cylinder is usually lower than the atmospheric pressure
70
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
TDC
BDC
c
Pressure in cylinder
Expansion
Compression
A
b
d
Inlet
Exhaust
*
Ignition
e
f
g
B
a
h
V1
V2
Volume
FIGURE 3.3 Diagram of pressure versus volume.
because of the resistance of the airflow into the cylinder. In the compression stroke (a–b–c), the pressure increases with the upward movement of
the piston. When the piston approaches the TDC, the spark plug produces a
spark to ignite the air/fuel mixture trapped in the cylinder, and the pressure
increases quickly. In the expansion stroke (c–d–e), the high-pressure gases in
the cylinder push the piston downward, producing torque on the crankshaft.
In the exhaust stroke (e–f–g), the gases in the cylinder are propelled out of the
cylinder with a higher pressure than in the induction stroke.
The torque performance is usually evaluated by the gross work done in one
cycle, usually called gross indicated work, Wc,in . The gross indicated work
can be calculated by
p dV −
p dV,
(3.1)
Wc,in =
area A
area B
where p is the pressure in the cylinder and V is the volume. The work done
in area B is negative, because the pressure in the induction stroke is lower
than that in the exhaust stroke. In order to achieve much work in one cycle,
area A should be made as large as possible by increasing the pressure in the
expansion stroke, and area B should be made as small as possible by increasing
the pressure in the induction stroke and decreasing it in the exhaust stroke.
71
Internal Combustion Engines
When the pressure in the induction stroke is greater than that in the exhaust
stroke, the work in this area will become positive. This is the case of the
supercharged engine.
The torque of an engine depends on engine size [engine displacement,
which is defined as the volume that the piston sweeps from TDC to bottom dead center (BDC)]. A more useful relative performance measure is the
mean effective pressure (mep), which is defined as the work per cycle per
displacement:
mep =
work per cycle
.
displacement of cylinder
(3.2)
The mean effective pressure can be expressed in terms of torque as
mep (kPa) =
2πnR T (Nm)
Vd (dm3 )
,
(3.3)
where nR is the number of revolutions of the crankshaft for each power stroke
per cylinder (nR = 2 for 4S engines and nR = 1 for 2S engines), T is the torque
in N m, and Vd is the displacement of the engine. The torque of an engine
depends only on the mean effective pressure in the cylinder and the engine
displacement Vd . For a given engine size, increasing the mean effective pressure is the only method of increasing the engine torque. It should be noted that
when dealing with mean effective pressure, one must clearly state whether it
is indicated mep (imep), which is measured within the cylinder, or brake mep
(bmep), which is measured on the crankshaft. The difference between them is
that the imep includes the engine mechanical loss (it is gross) and bmep does
not include the engine mechanical loss (it is net). Mechanical loss is discussed
in the following section.
3.1.2.3
Mechanical Efficiency
Not all the power produced in the cylinder (indicated power) is available on
the crankshaft. Part of it is used to drive engine accessories and overcome
the frictions inside the engine. All of these power requirements are grouped
together and called friction power Pf ; thus
Pig = Pb + Pf ,
(3.4)
where Pb is brake power (useful power on the crankshaft). It is quite difficult to determine the friction accurately. In practice, one common approach
for automotive engines is to drive or motor the engine on a dynamometer
(operate the engine without firing it) and measure the power supplied by the
dynamometer.
72
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
The ratio of brake power (useful power on the crankshaft) to indicated
power is called mechanical efficiency, ηm :
ηm =
Pb
Pf
=1−
.
Pig
Pig
(3.5)
The mechanical efficiency of an engine depends on the throttle position as
well as on the design and engine speed. Typical values for modern automotive engines with wide-open throttles are 90% at speeds below about
1800–2400 rpm, decreasing to 75% at maximum rated speed. As the engine
is throttled, the mechanical efficiency decreases eventually to zero at idle
operation.
By removing the engine mechanical loss from the indicated work or imep,
one can obtain the network or bmep that is measured on the crankshaft. The
maximum bmep of good engine designs is well established, and is essentially
constant over a large range of engine sizes. Typical values for bmep are as
follows. For naturally aspirated spark-ignition engines, maximum values are
in the range 850–1059 kPa (125–150 psi) at the engine speed where maximum
torque is produced. At the maximum rated power, bmep values are 10–15%
lower. For turbocharged automotive spark-ignition engines, the maximum
bmep is in the range 1250–1700 kPa (180–250 psi). At the maximum rated
power, bmep is in the range of 900–1400 kPa (130–200 psi).
3.1.2.4
Specific Fuel Consumption and Efficiency
In engine tests, fuel consumption is measured as a flow rate—mass flow per
unit time, ṁf . A more useful parameter is the specific fuel consumption (sfc)—
the fuel flow rate per useful power output. It measures how efficiently an
engine is using the fuel supplied to produce work:
sfc =
ṁf
,
P
(3.6)
where ṁf is fuel flow rate and P is engine power. If the engine power P is
measured as the net power from the crankshaft, the specific fuel consumption
is called brake specific fuel consumption (bsfc). The sfc or bsfc is usually measured
in SI units by the gram numbers of fuel consumed per kW power output per
hour (g/kWh). Low values of sfc (bsfc) are obviously desirable. For SI engines,
typical best values of bsfc are about 250–270 g/kWh.
Normally, a dimensionless parameter that relates the desirable engine
output (work per cycle or power) to the necessary input (fuel flow) would
have more fundamental value. The ratio of the work produced per cycle
to the amount of fuel energy supplied per cycle that can be released in the
combustion process is commonly used for this purpose. It is a measure of the
73
Internal Combustion Engines
engine efficiency (fuel conversion efficiency) given as
ηf =
Wc
P
=
,
ṁf QHV
mf QHV
(3.7)
where Wc is the work done in one cycle, mf is the fuel mass consumed per
cycle, and QHV is the heating value of the fuel, which is defined as the heat
released from unit fuel with complete combustion at standard conditions
and the combustion products cooling down to their original temperature.
Typical heating values for commercial HC fuels used in engines are in the
range of 42–44 MJ/kg (11.7–12.2 kWh/kg). The dimensionless efficiency can
be expressed by sfc as
1
.
(3.8)
ηf =
sfc QHV
3.1.2.5
Specific Emissions
The level of emission of oxides of nitrogen [nitric oxide (NO) and nitrogen
dioxide (NO2 ) usually grouped together as NOx ], carbon monoxide (CO),
unburned HCs, and particulates are important engine operating characteristics. The concentrations of gaseous emissions in engine exhaust are usually
measured in parts per million or percent by volume (mole fraction). Specific
emissions are the flow rate of pollutant per power output:
ṁNOx
,
P
ṁCO
sCO =
,
P
ṁHC
sHC =
,
P
ṁpart
spart =
.
P
sNOx =
(3.9)
(3.10)
(3.11)
(3.12)
Alternatively, emission rates can be normalized by the fuel flow rate; an
emissions index (EI) is commonly used as
EINOx =
ṁNOx (g/s)
,
ṁf (kg/s)
(3.13)
with similar expressions for CO, HC, and particulates.
3.1.2.6
Fuel/Air and Air/Fuel Ratios
In engine testing, both the air mass flow rate, ṁa , and fuel mass flow rate,
ṁf , are normally measured. The ratio of these flow rates is useful in defining
74
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
engine operating conditions:
fuel/air ratio (F/A) =
ṁf
ṁa
and air/fuel ratio (A/F) =
ṁa
.
ṁf
(3.14)
The stoichiometric fuel/air ratio is defined as the mass ratio of fuel to air such
that, with this mass ratio, the combustion can be chemically completed. For
gasoline, the stoichiometric fuel/air ratio is 0.0685 (air/fuel ratio: 14.6). More
conveniently, fuel/air equivalent ratio, φ, and air/fuel equivalent ratio, λ, are
commonly used. The fuel/air equivalent ratio is defined as
φ=
(F/A)actual
(F/A)s
(3.15)
λ=
(A/F)actual
.
(A/F)s
(3.16)
and
For fuel-rich mixtures: φ > 1, λ < 1.
For stoichiometric mixtures: φ = λ = 1.
For fuel-lean mixtures: φ < 1, λ > 1.
The normal operating range for a conventional SI engine using gasoline is
0.82 ≤ φ ≤ 1.23 or 0.056 ≤ F/A ≤ 0.083.
3.1.2.7 Volumetric Efficiency
The intake system—air filter, intake manifold, throttle plate, intake port, and
intake valve—restricts the amount of air that an engine of a given displacement can induct. The parameter used to measure the effectiveness of an
engine’s induction process is volumetric efficiency, ηv . Volumetric efficiency
is defined as the volume flow rate of air into the intake system divided by the
rate at which volume is displaced by the piston:
ηv =
2ṁa
,
ρa,i Vd N
(3.17)
where ρa,i is the inlet air density and N is the rpm of the engine. An alternative
equivalent definition for volumetric efficiency is
ηv =
ma
,
ρa,i Vd
(3.18)
where ma is the mass of air induced into the cylinder per cycle.
The inlet density may be taken either as the atmospheric air density (in
which case ηv measures the pumping performance of the entire inlet system)
or as the density in the inlet manifold (in which case ηv measures the pumping
75
Internal Combustion Engines
performance of the inlet port and valve only). Typical maximum values of ηv
for naturally aspirated engines are in the range of 80–90%. The volumetric
efficiency for diesel is somewhat higher than for SI engines.
3.1.3
Relationships between Operation and Performance Parameters
The importance of the above-discussed parameters to engine performance
becomes evident when power, torque, and mean effective pressure are
expressed in terms of these parameters. For power P,
P=
ηf ma NQHV (F/A)
.
nR
(3.19)
For 4S engines, volumetric efficacy can be introduced, and Equation 3.19 can
be expressed as
ηf ηv NVd QHV ρa,i (F/A)
.
nR
(3.20)
ηf ηv Vd QHV ρa,i (F/A)
.
4π
(3.21)
mep = ηf ηv QHV ρa,i (F/A).
(3.22)
P=
For torque T,
T=
For mean effective pressure,
The power per unit piston area, often called specific power, is a measure of
the engine designer’s success in using the available piston area regardless of
cylinder size. From Equation 3.20, the specific power is
ηf ηv NLQHV ρa,i (F/A)
P
.
=
Ap
2
(3.23)
Mean piston speed can be introduced using Equation 3.23:
ηf ηv Sp QHV ρa,i (F/A)
P
=
,
Ap
4
(3.24)
where Sp is the mean piston speed.
These relationships illustrate the direct importance of the following parameters to engine performance:
1. High fuel conversion efficiency.
2. High volumetric efficiency.
76
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
3. Increasing the output of a given displacement engine by increasing
the inlet air density.
4. Maximum fuel/air ratio that can be usefully burned in the engine.
5. High mean piston speed.
3.1.4
3.1.4.1
Engine Operation Characteristics
Engine Performance Parameters
The practical engine performance parameters of interest are power, torque,
specific fuel consumption, and specific emissions. The power of the 4S engine
can be expressed as
P=
mepAp S̄p
,
4
(3.25)
where mep is mean effective pressure, Ap is area of the piston head, and S̄p
is mean piston speed. The torque, T, is given by
T=
mepVd
.
4π
(3.26)
Thus for well-designed engines, the maximum values of mean effective pressure and piston speed are either flow limited (in naturally aspirated engines)
or stress limited (in turbocharged engines). Power is proportional to piston area and torque to displaced volume. For 4S engines, the mean effective
pressure can be expressed as
F
.
mep = ηf ηv QLHV ρa,i
A
(3.27)
The importance of high fuel conversion efficiency, volumetric efficiency
(breathing capacity), and inlet air density is clear. Specific fuel consumption
is related to the fuel conversion efficiency by
sfc =
1
.
ηf QLHV
(3.28)
These parameters have both brake and indicated values. The difference
between these two quantities is the engine’s friction (and pumping) requirements and their ratio is the mechanical efficiency (see Equation 3.5).
The relative importance of these parameters varies over an engine’s operation speed and load range. The maximum rated or normal rated brake power
and quantities such as bmep define an engine’s full potential. The maximum
brake torque (MBT) (and bmep derived from it) over the full speed range
77
Internal Combustion Engines
indicates the engine’s ability to obtain a high airflow through itself over the
full speed range and to use that air effectively. Over the whole operating
range—and most especially in those parts of that range where the engine will
operate for a long period of time—engine fuel consumption, efficiency, and
engine emissions are important.
3.1.4.2
Indicated and Brake Power and Torque
The wide-open throttle operating characteristics of an SI engine are shown
in Figure 3.4. Indicated power is the average rate of work transfer from
gases in the cylinders to the piston during the compression and expansion
strokes. Brake power is obtained by subtracting friction power from indicated power. The brake power shows a maximum value at about a speed
slightly less than the maximum speed of the engine. Indicated torque shows
a maximum value in the mid-speed range, which approximately corresponds
to the speed at which volumetric efficiency has the maximum value. Brake
torque decreases more than indicated torque at high speed because of more
friction loss.
At partial load and fixed throttle position, these parameters behave similarly; however, at high speeds, torque decreases more rapidly than at full
Indicated
torque
Indicated
power
120
240
100
300
Brake
power
Brake
torque
180
120
60
450
40
20
Indicated sfc
350
Brake sfc
0
1000
2000
4000
3000
Speed (rpm)
5000
150
250
0
Sfc (g/kWh)
Power (kW)
80
Torque (Nm)
140
6000
FIGURE 3.4 Indicated and brake powers, torques, and specific fuel consumptions varying with
engine speed.
78
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Throttle angle
240
Engine torque (Nm)
210
180
150
q=
120
q=
90
q=
60
30
q=
q=
15
q=
20
º
25
q=
º
30
35
40
q=
90
º
q=
70
º
q=
60
º
50
º
º
º
º
º
0
500 1000 1500 2000 2500 3000 3500 4000 4500 5000 5500
Engine speed (rpm)
FIGURE 3.5 Torque characteristics with engine throttle opening and engine speed.
load as shown in Figure 3.5. The partially opened throttle causes more
resistance to flowing air at higher speed and volumetric efficiency decreases.
The pumping components of total friction also increase as the engine is
throttled.
3.1.4.3
Fuel Consumption Characteristics
The fuel consumption characteristics (brake specific fuel consumption) of an
engine vary widely with engine speed and load, as shown in Figure 3.6. Generally, an engine has its optimal operating region, in which the fuel consumption
is minimized. This region usually locates in the middle of the speed range,
corresponding to the maximum torque, where the losses in the induction and
exhaust strokes are minimized. On the other hand, this region is close to full
load operation (wide-open throttle), where the percentage of losses to the
total indicated power is small. In vehicle design, the operating points of the
engine should be close to this region in order to achieve high operating fuel
economy.
3.1.5
Design and Operating Variables Affecting SI Engine Performance,
Efficiency, and Emission Characteristics
The major design and operating variables that affect SI engine performance,
efficiency, and emission characteristics are compression ratio, spark timing,
valve timing, air/fuel ratio, and fraction of exhaust gases that are recycled for
NOx emission control.
79
Internal Combustion Engines
10
250
20
30
40
50
Brake torque (Nm)
200
60
70
80
90 Power (kW)
Torque with wide
open throttle
982
bsfc, g/kWh
(efficiency)
250 (34.3%)
785
260 (33.3%)
)
270 (31.8%
)
280 (30.6%
150
589
.7%)
(27
310
100
350
)
.5%
(24
393
%)
1.4
0 (2
40
)
17.1%
.3%)
600 (14 (12.4%)
700 (10.7%)
800
1000 (8.57%)
500 (
50
0
0
Brake mean pressuure (kPa)
(a)
196
0
1000 1500 2000 2500 3000 3500 4000 4500 5000 5500
Engine speed (rpm)
500
(b)
90
80
Engine power (KW)
70
60
50
Power with wide
open throttle
bsfc, g/kWh
(efficiency)
)
) ) %) 0.6%
.3%33% 31.80 (3
4
)
3 ( ( 8
.7%
0 ( 60 70 2
25 2 2
(27
0
)
31
.5%
24
)
(
0
.4%
35
21
(
)
0
40
.1%
(17
)
.3% 2%)
)
500
(14
.
.7%
600 00 (12 00 (10 .57%)
7
8 0 (8
100
40
30
20
10
0
500
1000
1500
2000
2500 3000 3500
Engine speed (rpm)
4000
4500
5000
5500
FIGURE 3.6 Fuel consumption characteristics of a typical SI engine: (a) along with engine speed
and torque and (b) along with engine speed and power.
3.1.5.1
Compression Ratio
The compression ratio of an engine is defined as the ratio of the total volume of
a cylinder (V2 , as shown in Figure 3.3) to the volume when the piston moves
to the TDC (V1 , as shown in Figure 3.3). Generally, a higher compression
80
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
ratio yields high fuel conversion efficiency. The highest compression ratio
of an SI engine is restricted by the octane number of the fuel. A high octane
number allows higher compression ratio. For automotive gasoline SI engines,
the compression ratio is in the range of 8–10.1 Some additives are usually
added to gasoline to enhance its octane number. Typical additives are lead
and tetraethyl lead, which are very effective in enhancing the octane number
of gasoline and have been widely used. However, lead and tetraethyl lead are
very poisonous and have been prohibited in most places in the world. Other
gasoline additives include methyl tert-butyl ether (MTBE), tert-amyl methyl
ether and ethyl tert-butyl ether. Alcohols, such as methanol and ethanol, are
also used as gasoline additives to enhance the octane number of gasoline.
3.1.5.2
Spark Timing
For SI engines under normal operating conditions, towards the end of the
compression stroke, the spark plug produces a spark to ignite the mixture in
the cylinder and combustion is initiated. Because propagation of the flame in
the cylinder needs time, ignition must start before the end of the compression
stroke. The degree of the crankshaft angle, from ignition starting to the end
of the compression stroke, is called spark advance in degrees of crankshaft.
Spark timing has a considerable effect on engine performance, efficiency,
and exhaust emission.
Figure 3.7a shows the gas pressure in the cylinder with the crankshaft angle
at different spark timings. Starting the ignition too soon (50◦ before TDC in
Figure 3.7a) results in a high gas pressure acting on the piston in the compression stroke. Consequently, the negative work in the compression stroke
increases and the positive work in the expansion stroke decreases, resulting in
low average torque. Sometimes, too-early ignition causes abnormal combustion in the cylinder, which usually causes piston “knocking.” “Knocking” is
the most serious abnormal combustion phenomenon, which usually causes
damage to mechanical parts such as crankshaft, bearings, and connection
rods. Too-late ignition (10◦ before TDC in Figure 3.7a) results in low gas pressure in the cylinder, and thus low effective work. There exists an optimal spark
timing at which the average engine torque reaches its maximum value. This
optimal spark timing is called MBT timing as shown in Figure 3.7b. Spark
timing, which is advanced or retarded from MBT timing, gives lower average torque. MBT timing is associated with the rate of flame development and
propagation, the length of the flame travel path across the combustion chamber, and the details of the flame termination process after it reaches the wall.
These depend on engine design, operating conditions, properties of fuel, air,
burned gas mixture, etc. With a given design, engine speed has a great influence on MBT timing, and therefore the sparking timing should be adjusted
with engine speed.
The time from ignition starting to reaching the maximum gas pressure in the
cylinder changes slightly. Thus, with increasing engine speed, the sparking
81
Internal Combustion Engines
(a)
Spark advance
angle (crankshaft)
1—50°
2—30°
3—10°
Pressure in cylinder
1
2
3
– 60
3
2
1
–40
–20
0
20
TDC
40
60
80
100
Crankshaft angle (°)
(b)
Relative torque
1.0
0.9
MBT
10
20
30
Spark advance (°)
40
MBT timing (° BTC)
(c)
40
30
20
1200
1700
2200
2700
3200
3700
4200 4700
Engine speed (rpm)
FIGURE 3.7 Effect of spark advance on engine performance: (a) cylinder pressure, (b) torque,
and (c) MBT timing with engine speed.
82
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
advance should be increased as shown in Figure 3.7c. Correct spark timing
is important, because NO and HC emissions vary significantly with spark
timing.
3.1.5.3
Fuel/Air Equivalent Ratio
Proper fuel/air (or air/fuel) ratio in the fuel/air mixture is a crucial factor
that affects the performance, efficiency, and emission characteristics of an
engine, as shown in Figure 3.8. The mean effective pressure peaks at slightly
rich stoichiometry (between φ = 1 and φ = 1.2). However, the fuel conversion
efficiency decreases as the mixture is enriched above the stoichiometry (φ > 1)
because part of the fuel is left after the combustion. When φ decreases, the
fuel conversion efficiency increases because there exists sufficient oxygen in
the cylinder to oxidize the fuel and the energy of the fuel is almost completely
released as thermal energy. In SI engines, too lean mixtures will cause misfire. φ = 0.8 would be the bottom limit. With lean mixtures, the combustion
produces lower temperature in the cylinder and thus results in low mean
effective pressure.
The fuel/air ratio has a dominant effect on emissions of SI engines as
shown in Figure 3.9. Leaner mixtures produce lower emissions (NOx , CO,
and HC) until combustion becomes poor (and eventually misfire occurs).
1100
900
38
800
36
isfc (g/kWh)
700
34
ηf,i
320
32
300
30
280
28
ηf,i
imep (kpa)
imep
1000
26
260
isfc
240
220
0.6
0.8
Lean mixture
1.0
1.2
1.4
Rich mixture
Fuel/air equivalent ratio
FIGURE 3.8 Fuel/air ratio effect on imep, fuel conversion efficiency, and isfc.
83
NOx, XO and HC concentration (not to scale)
Internal Combustion Engines
NOx
HC
CO
0.7
0.8
0.9
1.0
Lean mixture
1.1
1.2
1.3
Rich mixture
Fuel/air equivalence ratio
FIGURE 3.9 Fuel/air ratio effect on emissions.
When the fuel/air ratio is lower than a certain point (0.85 in Figure 3.9),
HC emissions rise sharply and engine operation becomes erratic. The NOx
emissions have a different curve shape from those of CO and HC. they
have a peak value close to the stoichiometric mixture, because NOx tends
to form at high temperature and pressure, which is the case of stoichiometric
combustion.
Figure 3.9 indicates the complexities of emission control. In a cold engine
with slow fuel vaporization, fuel flow is increased to provide an easily combustible fuel-rich mixture in the cylinder, until the engine warms up and
this enrichment is removed. At part-load conditions, a lean mixture in the
cylinder, which would produce lower HC and CO emissions and moderate NOx emissions, could be used. The use of recycled exhaust to dilute the
engine intake mixture lowers NOx level (decreasing the temperature in the
cylinder), and also deteriorates combustion quality. In practice, the fuel/air
ratio varies with engine speed and intake mass flow rate as shown in
Figure 3.10. At low mass flow rate (small throttle opening), the mixture is
close to stoichiometric or richer, especially in an engine with carburetor fuel
metering (the upper curve in Figure 3.10), due to less fuel atomization in the
84
Fuel/air equivalent ratio
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
1.2
Low
Mid
High
speed
1.0
0.8
0
Intake air mass flow rate
100%
FIGURE 3.10 Fuel/air ratio varies with engine speed and intake mass flow.
mixture and for smooth engine operation. With a fuel injection system, fuel
atomization is better than with a carburetor. With a high intake mass flow
rate, the mixture becomes richer for high torque output. In the middle, the
mixture is lean for better efficiency and emissions.
3.1.6
Emission Control
Today emissions are treated at three points: fuel, engine, and exhaust (aftertreatment). Gasoline is modified by means of additives and refining processes
to adapt its composition in order to reduce the formation of pollutant species
and facilitate after-treatment. For example, a reformulated gasoline may contain MTBE, which increases the fuel octane ratio in replacement of lead. MTBE
is less harmful to the atmosphere than lead compounds used previously; however, it is no longer used because of its danger to water sources in case of spill.
Gasoline is also treated for its sulfur content. Some after-treatment techniques
can only tolerate very small amounts of sulfur compounds in gases.
A good example of engine design techniques that limit the formation of
species is exterior air intake. Previously, the engine would breathe air from
underneath the hood. This air is much hotter than ambient air, thus resulting
in higher temperatures during compression and combustion and therefore
resulting in high nitric oxide formation. Today, the engine breathes air from
the outside, which is much cooler and therefore results in lower nitric oxide
formation. Other design or operation parameters that affect the formation
of polluting species include the following: high-pressure fuel injection and
turbulent intake flows that improve the homogeneity of the mixture; combustion chamber design; aluminum engine blocks that allow running the engine
cooler; sodium-filled exhaust valves that can therefore be cooled more easily;
exhaust gas oxygen sensing that tells the engine controller what the fuel/air
ratio really is, etc.
The most powerful means of controlling the emission of polluting species
is exhaust after-treatment. The most widespread technique, which is now
mandatory in most Western countries, is through the catalytic converter. It
Internal Combustion Engines
85
is composed of a mesh of catalytic material such as platinum and rhodium.
These catalysts react with the polluting species and transform them to harmless species. Nitric oxides are reduced to oxygen and air, carbon monoxide
is oxidized to carbon dioxide, and unburned HCs are combusted to carbon
dioxide and water. A catalytic converter that can take care of all three species
at the same time is referred to as a three-way catalytic converter.
Catalytic converters need a slight excess of oxygen in order to oxidize CO
and HC species. Therefore, the fuel/air ratio must be precisely controlled in
order to maintain a constant oxygen concentration in the exhaust. This is done
by means of exhaust oxygen sensing and fuel injections. Catalytic converters
are the primary reason for the replacement of carburetors by fuel injection.
The catalysts operate efficiently only at high temperature. They do not work
for about 1 min when the engine is cold-started. It is during this lapse of
time that the engine emits most of its pollutants. It should be noted that some
emission control techniques do impair fuel consumption. This is especially
true for the catalytic converter that acts as a restriction in the exhaust flow.
3.1.7
3.1.7.1
Basic Techniques for Improving Engine Performance,
Efficiency, and Emissions
Forced Induction
The amount of torque produced in an IC engine depends on the amount of
air induced into its cylinders. An easy way of increasing the amount of air
induced is to increase the pressure in the intake manifold. This can be done
by three means: variable intake manifold, supercharging, or turbocharging.
The intake manifold is like a wind instrument: it has resonant frequencies.
A variable intake manifold tunes itself according to engine speed in order
to exploit those resonant frequencies. If the tuning is proper, the amount of
air induced into the cylinders can be optimized because the pressure in the
intake manifold is increased. This technique improves the “breathing” of the
engine but does not result in a very large increase of torque output.
A supercharger is an air compressor turned by the engine crankshaft. The
air thus compressed is fed to the intake manifold. The advantage of a supercharger is that it can significantly increase the pressure in the intake manifold,
even at low speed. The most significant disadvantage is that the supercharging power is taken from the engine crankshaft. This reduces the engine output
and harms fuel consumption.
A turbocharger consists of a turbine driven by exhaust gases and of a compressor turned by the turbine. A turbocharger has the great advantage of
taking its energy from the exhaust gases, which are normally wasted. Therefore, the efficiency of the engine does not suffer from the addition of the
turbocharger. Turbocharging can tremendously increase the power output of
the engine, especially if coupled to a charge cooling system. It also significantly improves the efficiency because the higher intake pressure reduces
86
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
the negative work associated with the induction stroke. The disadvantages of
turbocharging include slow response, little or no effectiveness at low engine
speed, and high rotational speed for the turbocharger, which increases the
cost of forced induction.
Supercharging and turbocharging both suffer from two disadvantages:
knock and emissions. Compressing the intake air also increases its temperature. An increased temperature means a greater risk of auto-ignition and
knocks for the mixture, and increased nitric oxide emissions. The solution
to this problem consists in cooling down the intake air after compression
by means of an intercooler or heat exchanger. The compressed air is passed
through a radiator, while the ambient air or water is passed on the exterior of
the radiator, removing the heat from the charge. The temperature of induced
air can be reduced sufficiently to avoid auto-ignition and knock. Nitric oxide
emissions are also reduced. It should be noted that an engine designed for
forced induction has a lower compression ratio than an engine that is designed
for normal induction. Cooling the induced air is also beneficial for torque production, because cooler air is denser air. Therefore, more air can be induced
into the cylinder if it is cooler.
3.1.7.2
Gasoline Direct Injection and Lean-Burn Engines
HC and CO emissions can be reduced if the engine burns a lean mixture. If
a SI engine could run on extremely lean mixtures, then the emissions would
be very significantly reduced. However, ultralean mixtures pose problems
because the flame has trouble propagating.
Gasoline direct injection is one means of achieving a very efficient mixing. Because the injector is located in the cylinder, it must inject the fuel at
high pressure, which reduces the size of the fuel droplets. The fuel can be
injected close to the spark plug, thus enriching locally the mixture and allowing better ignition of the intake air. Additional advantages of gasoline direct
injection include the cooling of intake air, which reduces knock and allows
operating at a higher compression ratio. This further improves the engine
efficiency. Besides the cost increase, gasoline direct injection results in higher
NOx emissions.
3.1.7.3
Multi- and Variable-Valve Timing
While many engines use only two valves, high-performance engines use three,
four, or five valves in order to increase the intake flow area. Multiple valves
provide a significant increase of torque at high engine speed, but sacrifice the
low-speed torque because the larger intake flow area results in slower flows
at low speed. Multiple valves imply multiple camshafts, which increase the
cost and complexity of the engine.
It has been previously shown that valve timing needs to be optimized for
each engine but also for specific operating conditions in order to avoid reverse
Internal Combustion Engines
87
flows. While conventional engines use a fixed, compromised valve timing,
some advanced engines use a variable valve timing. This allows a better
control of the amount of mixture and therefore helps in reducing fuel consumption and emissions. The major drawbacks are the cost and complexity
of variable valve systems.
3.1.7.4 Throttle-Less Torque Control
Because most of the losses at partial torque output result from the throttle, eliminating the throttle eliminates these losses and improves the partial
torque fuel consumption. Torque control is achieved by means of variable
valve timing or by varying the fuel/air ratio in lean-burn engines.
3.1.7.5 Variable Compression Ratio
A variable compression ratio allows operating a forced induction engine at
optimal compression ratio at any intake pressure. If the charging mechanism
does not provide maximum intake pressure, then the compression ratio of the
engine can be increased without risking auto-ignition or knock. The increased
compression ratio results in improved fuel economy at partial torque output.
3.1.7.6
Exhaust Gas Recirculation
Exhaust gas recirculation (EGR) consists in readmitting some of the exhaust
gases into the combustion chamber in order to reduce the effective displacement of the engine. This technique is used in conventional vehicles to decrease
the fuel consumption at partial torque output, while preserving the acceleration capabilities of the engine. The greatest benefit of EGR is in emission
reduction because it reduces the amount of fuel burned in the chamber and
therefore the temperature of the exhaust gases. The nitric oxide emissions are
greatly reduced.
3.1.7.7
Intelligent Ignition
Intelligent ignition systems can set the spark advances at their optimum valve
at any operating speed and load for optimum performance, efficiency, and
exhaust emissions. High-power ignition systems can prevent losing fire in
any cylinder, especially for engines with a lean mixture combination.
3.1.7.8
New Engine Materials
New materials developed for engine components will contribute to better
fuel economy in two ways. Firstly, ceramic materials can be expected to offer
better thermal insulation than metallic ones, with corresponding lower heat
transfer (and therefore lower heat loss) and hence higher thermal efficiency.
88
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Secondly, lightweight materials such as fiber-reinforced plastics with high
tensile strength can save a lot of weight.
3.2
4S, Compression-Ignition IC Engines
Compression-ignition (CI) engines normally use diesel as their fuel. The 4S,
CIIC engine has operation principles similar to the 4S, SI engine. It also has
four distinctive strokes—induction, compression, expansion, and exhaust.
However, in CI engines, air alone is induced into the cylinder. The fuel is
injected directly into the cylinder just before the piston moves up to the TDC.
High temperature in the compressed air ignites the fuel. Airflow at a given
engine speed is essentially unchanged and load control is achieved by varying
the amount of fuel injected at each cycle. Compared with SI engines, a CI
engine works differently as follows:
1. The compression ratio is higher.
2. During the initial part of compression, only air is present.
3. The fuel/air mixture is always stoichiometrically weak.
These operation characteristics result in a better fuel efficiency than a SI
engine. Furthermore, the CI engine is, in general, designed to operate at lower
speeds, and consequently the friction losses are smaller.
Since the fuel/air ratio in a CI engine is always lean, CO emission is much
lower than in SI engines, and can be negligible. Unburned HCs in a properly
regulated diesel engine come from two sources. Firstly, around the perimeters
of the reaction zone there is a mixture that is too lean to burn, and the longer
the delay period, the greater the amount of HC emissions from this source.
However, there is a delay period below which no further reductions in HC
emissions are obtained. Under these conditions, the HC emissions mostly
originate from a second source: the fuel retained in the nozzle sac (the space
between the nozzle seat and the spray holes) and the spray holes. Fuel from
these sources can enter the combustion chamber late in the cycle, thereby
producing HC emissions.
The formation of NOx is strongly dependent on temperature, the local concentration of oxygen, and the duration of combustion. Thus, in diesel engines,
NOx is formed during the diffusion combustion phase, on the weak side of
the reaction zone. Reducing the diffusion-controlled combustion duration
by increasing the rate of injection leads to a reduction in NOx emissions.
Retarding the injection timing also reduces the NOx emissions, since the later
injection leads to lower temperature. However, injection retarding will reduce
the fuel efficiency of the engine.
89
Internal Combustion Engines
The black smoke from diesel engine originates from the fuel-rich side of the
reaction zone in the diffusion-controlled combustion phase. After the rapid
combustion at the end of the delay period, the subsequent combustion of the
fuel is controlled by the rates of diffusion of air into the fuel vapor and vice
versa, and the diffusion of the combustion products away from the reaction
zone. Carbon particles are formed by the thermal decomposition (cracking)
of the large HC molecules, and soot particles form by agglomeration. The soot
particles can be oxidized when they enter the lean side of the reaction zone
and further oxidation occurs during the expansion stroke, after the end of the
diffusion combustion phase.
Smoke generation is increased by high temperature in the fuel-rich zone
during diffusion combustion. The smoke emission can be reduced by shortening the diffusion combustion phase, because this gives less time for soot
formation and more time for soot oxidation. The diffusion phase can be
shortened by increased swirl, more rapid rejection, and a finer fuel spray.
Advancing the injection timing can also reduce the smoke.
3.3 2S Engines
The 2S cycle combines the necessary processes of induction, exhaust, compression, and expansion in a single crankshaft rotation, unlike a 4S cycle that
requires two rotations.3 A basic 2S cycle is shown in Figure 3.11. 2S engines
eliminate the separate induction and exhaust strokes. The basic operation
principle is illustrated in Figure 3.11 and explained below.
1. In Figure 3.11a, the piston is approaching the top center. Above the
piston, the trapped air and fuel charge is ignited by the spark plug,
(a)
(b)
(c)
(d)
Exhaust
Exhaust
Exhaust
Exhaust
Inlet
Inlet
Inlet
Inlet
Compression and
induction
Blowdown exhaust
period
FIGURE 3.11 Basic 2S cycle.
Fresh charge
transfer
Approaching exhaust
closing
90
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
producing a rapid rise in temperature and pressure that will drive
the piston down on the power stroke. Below the piston, the opened
inlet port induces air from the atmosphere into the crankcase due to
the increasing volume of the crankcase lowering the pressure below
the atmospheric value. To induce fuel into the engine, various options
exist: placing a carburetor in the inlet tract, injecting fuel into the inlet
tract, injecting fuel into the crankcase transfer ducts, or injecting fuel
directly into the cylinder before or after the closure of the exhaust
port. If one operates the engine as a diesel power unit, the latter is the
only option.
2. In Figure 3.11b, the exhaust port above the piston has been opened.
It is often called the “release” point in the cycle and this allows the
transmission of a pulse of hot, high-pressure exhaust gas from the
combustion process into the exhaust duct. As the area of the port
increases with crankshaft angle and the cylinder pressure falls with
time, it is clear that the exhaust duct pressure profile with time is
one that increases to a maximum value and then decays. Such a flow
process is described as unsteady gas flow and such a pulse can be
reflected from all pipe area changes, or at the pipe end termination to
the atmosphere. These reflections have a dramatic influence on engine
performance. Below the piston, the compression of the fresh charge is
taking place. The pressure and temperature achieved will be a function of the proportion reduction of the crankcase volume, that is, the
crankcase compression ratio.
3. In Figure 3.11c, above the piston, the initial exhaust process referred to
as “blow-down” is nearing completion. The piston uncovers the transfer ports and connects the cylinder directly to the crankcase through
the transfer ducts. If the crankcase pressure exceeds the cylinder pressure, then the fresh charge enters the cylinder in what is known as the
“scavenge process.” Clearly, if the transfer ports are badly directed
then the fresh charge can exit directly out of the exhaust port and
be totally lost from the cylinder. Such a process, referred to as “shortcircuiting,” would result in the cylinder being filled only with exhaust
gas at the onset of the next combustion process, and no pressure rise or
power output would ensue. Worse, all of the fuel in a carbureted configuration would be lost to the exhaust—a consequentially monstrous
emission of unburned HC. This is the main reason why a conventional
2S engine has a lower efficiency and bad emissions.
4. In Figure 3.11d, in the cylinder, the piston is approaching what is
known as the “trapping” point. The scavenge process has been completed and the cylinder is now filled with a mix of air, fuel if a
carbureted design, and exhaust gas. As the piston rises, the cylinder
pressure also rises, but the exhaust port is still open and, barring the
intervention of some unsteady gas-dynamic effect generated in the
exhaust pipe, the piston will spill fresh charge into the exhaust duct
91
Internal Combustion Engines
to the detriment of resulting power output, fuel consumption, and
pollutant emissions.
The simplest method of allowing fresh charge access into, and exhaust gas
discharge from, the 2S engine is by the movement of the piston exposing port
in the cylinder wall as shown in Figure 3.11. This means that all port timing
events are symmetrical with respect to TDC and BDC. It is possible to produce
asymmetrical inlet and exhaust timing events by the use of disc valves, reed
valves, and poppet valves.
The use of poppet valves for both inlet and exhaust timing control is
sketched in Figure 3.12 as an example of uniflow scavenging. This design
is commonly used in 4S engines. Generally, the poppet valves’ design is considered to be difficult to design so as to adequately flow sufficient charge into
a 2S engine, compared with using the port in a cylinder wall with the same
access area.4
The disc valves and reed valves are another two valves for 2S engines, as
shown in Figure 3.13. The disc valve can provide asymmetric porting timing
by designing the disc. However, the porting timing can be varied with the
engine speed. Consequently, the porting timing is optimized only at a specified speed, and at other speeds the engine performance will be as good as at
(a)
Cams
Exhaust
(b)
Poppet
valves
Cams
Inlet
Inlet
Exhaust
FIGURE 3.12 Uniflow scavenged with crankcase compression: (a) top exhaust and (b) top inlet
design.
92
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
(a)
(b)
Exhaust
Disc
valve
Inlet
Reed
valve
FIGURE 3.13 (a) Disc valve and (b) reed valve with crankcase compression.
this speed. Reed valves have always been popular as they provide an effective automatic valve whose timings vary with both engine load and engine
speed. The high-performance outboard racing engine demonstrated that high
specific power output was possible with reed valves.4 Racing motorcycle
organizations developed this further. Today, most reed valves are designed
as V-blocks (refer to Figure 3.13b).
The essential element of the original 2S engines was the use of a crankcase as
the air pumping device. The conventional lubrication method has been to mix
the lubricant with the gasoline and supply it through the carburetor in ratios
of gasoline to lubricant from 25:1 to 100:1, depending on the application. Combined with some fuel charge being short-circuited to the exhaust duct along
with the air, the resulting exhaust is rich in unburned HCs and lubricant,
some partially burned and some totally unburned. It is consequently visible as
smoke. This is ecologically unacceptable. By definition, an external air pump
can be utilized to replace the crankcase air pumping. This can be either a positive displacement blower of a Roots type or a centrifugal blower driven from
the crankshaft. Clearly, it would be more efficient thermodynamically to use a
turbocharger, where the exhaust energy to the exhaust turbine is available to
drive the air compressor. This arrangement is shown in Figure 3.14, where the
engine has both a blower and a turbocharger. The blower is used as a starting aid and as an air supplementary device at low loads and speeds, with
the turbocharger being used as the main air supply unit at higher torque and
power levels at any speed. To prevent any short-circuiting fuel to the exhaust,
a fuel injector is used to supply fuel directly to the cylinder, after the exhaust
port is closed and not in the position as shown in Figure 3.14, at BDC. Such an
engine type has already demonstrated excellent fuel economy behavior, good
exhaust emission characteristics of unburned HCs and carbon monoxide, and
93
Internal Combustion Engines
Fuel injector
Turbocharger
Roots blower
Fuel
spray
Blower drive
Oil control ring
FIGURE 3.14 A supercharged and turbocharged fuel-injection 2S engine.
superior emission characteristics of oxides of nitrogen, in comparison with
an equivalent 4S engine.
3.4 Wankel Rotary Engines
The rotary-piston engine (or Wankel engine, named by its inventor) generates
power by the compression, ignition, and expansion of a gasoline/air mixture
in a 4S cycle in the same way as conventional IC engines. The completely
different mechanical design allows all moving parts to have a continuous
rotary motion instead of a reciprocating movement. The rotor (or piston) is
roughly triangular shaped and rotates on an eccentric shaft on the output
shaft within a housing of epitrochoid shape. The term is given to the path
described by a point within a circle rolling around another circle. The twolobe epitrochoid of the Wankel is generated when the interior circle is twice
the radius of the rolling circle.
The four strokes of the cycle occur in the spaces formed between the three
sides of the rotor and the two-lobed trochoidal chamber or housing; the spaces
are expanded and contracted as the rotor turns. Figure 3.15 illustrates the
operating cycle. The three separate working chambers are labeled A, B, and
C. As the intake port is uncovered, volume A expands and draws a fresh
fuel/air charge into the engine (Figure 3.15a). Next the trailing apex seal
94
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
(a)
A
(b)
B
(c)
A
(d)
C
B
A
C
C
B
B
A
(e)
C
B
A
C
FIGURE 3.15 The Wankle rotary engine cycle: (a) three separate working chambers and air/fuel
intake into chamber A, (b) starting compression in chamber A, (c) ignition and starting expansion
in chamber A, (d) starting exhaust in chamber A, and (e) starting intake in chamber A.
isolates volume Afrom the intake port and the charge is compressed as volume
A steadily decreases with rotation of the rotor, reaching the minimum shown
at the point of ignition (Figure 3.15b and c). While this sequence of events is
occurring, the other two faces of the rotor are defining volumes B and C, which
experience exactly the same sequence, each offset by one-third of a rotation.
Unlike the piston in a conventional engine, which requires a connecting
rod to transmit power to the crankshaft, the rotor (or “rotary piston”) runs
directly on an eccentric shaft, from which torque output is derived. The
Wankel also differs from the conventional automobile engine in having no
valves. The fuel/air inlet port and the exhaust port are opened and closed at
the appropriate time in the combustion cycle by the passage of the rotor itself.
The advantages of the Wankel engine are as follows:
Because the engine delivers one power stroke for each full crankshaft rotation, the Wankel uses its displacement volume twice as often as the SI engine
does. One immediate advantage is that for equivalent power output, a Wankel
engine is only about half the size and weight of a conventional engine.
Another advantage of the rotary engine is the reduced number of parts—
typically 40% of the number of parts moving in a V-8 engine, although
the advantage over a four-cylinder engine would be less. This has certain
implications for easier and faster manufacturing of the engine.
Since the Wankel engine possesses only rotary motion, all the unbalanced
forces are simple harmonic and can be fully canceled with rotating counterweights to produce very little vibration as compared with one-quarter of a
revolution in the 4S engine. This dispersed application of torque adds to the
basic smoothness of the Wankel with consequent lower noise (aided by the
absence of a valve train). Asingle-rotor Wankel is as smooth as a three-cylinder
piston engine. Most rotary engines being developed for automotive use have
two rotors.
The engine is capable of burning fuel of a lower octane rating than a conventional piston engine with a slightly better performance gained from unleaded
petrol. It can run at the stoichiometric fuel/air ratio allowing NOx control
(raw NOx emissions are generally lower than a conventional petrol engine
Internal Combustion Engines
95
due to the slower combustion) with a combination of EGR and a reduction
catalyst, and HC and CO control with an oxidation catalyst.
The Wankel engine breathes well because of a greater length (crank degrees)
of induction period and because of a relatively unrestricted flow path for
the incoming mixture and exhaust gases. Volumetric efficiency in excess of
100% has been claimed for some designs. In the one- and two-rotor versions,
the induction system is simple, which reduces problems associated with bad
distribution of the fuel/air mixture. Also, the speed range of the engine is
broad (i.e., higher than a conventional reciprocating engine) due to better
breathing characteristics, potentially lower mechanical friction, no limitations
imposed by valve train dynamics, and the absence of reciprocating parts.
Perhaps the most far-reaching economic significance of the Wankel lies outside the engine itself in the potential it offers for the redesign of the passenger
car. The comparatively small size and light weight of the engine would permit
vehicle aerodynamic improvements by way of lower hood lines and weight
reduction because of the lower engine mass and the requirement for a smaller
engine compartment.
The Wankel engine’s main disadvantages are reported to be its expensive construction materials, its requirement for high-precision manufacturing
techniques, poor fuel economy, and high emissions of unburned HCs. The
high HC emissions are because of poor sealing between the rotor and housing. The apex seal especially suffered from reliability and durability problems
and has been unable to match conventional piston rings. But newly developed
sealing materials are expected to provide a much higher level of service. The
apex seal in current rotary engines is composed of a single piece of carbon
impregnated with an aluminum alloy.
The high HC emissions have also contributed to the generally poor fuel
economy. The other main factor is that combustion in the Wankel engine tends
to be slower and hence less efficient than in the conventional piston engine.
In most new designs, the combustion is accelerated by using two spark plugs
in each combustion chamber. Continuing efforts to improve engine efficiency
via direct injection and stratified charge have been reported. Stratification
of the charge can readily be accomplished by injecting the fuel and creating
a rich mixture in the vicinity of the TDC position of the rotor, which will
substantially reduce the peak temperature and therefore decrease the NOx
produced. A lean mixture in the leading section of the chamber will enhance
the oxidation of CO and HCs.
3.5 Stirling Engines
The Stirling engine is a reciprocating, continuous external combustion (EC)
engine that has a piston and cylinder similar to a conventional IC engine as
shown in Figure 3.16.5 However, inside the engine, the working fluid (usually
96
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Exhaust products
of combustion
Air inlet
Exhaust gas/inlet
air preheater
Regenerator
Water
outlet
Compression
space
Heater
Fuel inlet
Expansion space
Cooler
Water
inlet
FIGURE 3.16 Diagram of a practical opposed-piston Stirling engine.
hydrogen or helium) is sealed. Alternately heating and cooling the working
fluid causes pressure fluctuation that acts on the piston to produce power.
A Stirling engine has a high-temperature heat source and a low-temperature
heat sink. The heat source sounds the compression space and the heat sink
sounds the expansion space. Both the source and the heat sink are separated
by a regenerator (thermodynamics sponge), as shown in Figure 3.17.
The Stirling engine works based on the Stirling thermodynamic cycle. The
ideal Stirling cycle is illustrated in Figure 3.17,5,6 which consists of a cylinder
containing two opposed pistons, with a regenerator between the pistons. The
regenerator, a thermodynamic sponge, is usually a matrix of finely divided
metal in the form of wires or strips. One of the two volumes between the
regenerator and the pistons is the expansion space, in which high temperature,
Tmax , is maintained by a heat source surrounding it. The other volume is the
compression space, in which low temperature, Tmin , is maintained by the heat
sink surrounding it. Therefore, there is a temperature gradient (Tmax − Tmin )
across the transverse faces of the regenerator. It is usually assumed that there
is no thermal conduction in the longitudinal direction.
To start the cycle, we assume that the compression-space piston is at the
outer dead point and the expansion-space piston is at the inner dead point,
close to the face of the regenerator. All the working fluid is then in the cold
compression space. The volume is maximum, so that the pressure and temperature are at their minimum values, represented as 1 on the P–V and T–S
diagrams as shown in Figure 3.17a. During compression (process 1–2), the
97
Internal Combustion Engines
(a)
3
3 Tmax 4
T
P
4
2
2 Tmin 1
1
(b)
V
Expansion space
S
Regenerator
Compression space
Tmin
(1)
(2)
(3)
Tmax
(4)
(c)
Displacement
Time
(1)
(2)
(3)
(4)
(1)
FIGURE 3.17 Ideal Stirling cycle: (a) P–V and T–S diagram, (b) piston arrangement at the
terminal points of the cycle, and (c) time–displacement diagram.
compression piston moves towards the inner dead point and the expansionspace piston remains stationary. The working fluid is compressed in the
compression space, and the pressure increases. The temperature is maintained
constant because heat Qc is abstracted from the compression-space cylinder
to the surroundings (heat sink).
In the transfer process 2–3, both pistons move simultaneously, the compression piston towards (and the expansion piston away from) the regenerator,
so that the volume between them remains constant. Therefore, the working
98
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
fluid is transferred, through the porous metallic matrix of the regenerator,
from the compression space to the expansion space. In its passage through
the regenerator, the working fluid is heated from Tmin to Tmax , by heat transfer
from the matrix, and emerges from the regenerator into the expansion space
at temperature Tmax . The gradual increase in temperature in passage through
the matrix, at constant volume, causes an increase in pressure.
In the expansion process 3–4, the expansion piston continues to move
away from the regenerator towards the outer dead center point; the compression piston remains stationary at the inner dead center point, adjacent to
the regenerator. As the expansion progresses, the pressure decreases as the
volume increases; the temperature remains constant because heat QE is added
to the system from an external heat source.
The final process in the cycle is the transfer process 4–1, during which
both pistons move simultaneously to transfer the working fluid (at constant
volume) back, through the regenerative matrix, so that the working fluid
decreases in temperature and emerges at Tmin into the compression space.
Heat transferred in the process is contained in the matrix, for transfer to the
gas in process 2–3 of the subsequent cycle.
The cycle is composed of heat transfer; therefore, there are four heat-transfer
processes:
1. Process 1–2: isothermal compression: heat transfer from the working
fluid at Tmin to the external heat sink.
2. Process 2–3: constant volume: heat transfer to the working fluid from
the regenerative matrix.
3. Process 3–4: isothermal expansion: heat transfer to the working fluid
at Tmax from an external heat source.
4. Process 4–1: constant volume: heat transfer from the working fluid to
the regenerative matrix.
If the heat transferred in process 2–3 has the same magnitude as in process 4–
1, then the only heat transfer between the engine and its surroundings are (a)
heat supply at Tmax and (b) heat rejection at Tmin . This heat supply and heat
rejection at constant temperature satisfy the requirement of the second law
of thermodynamics for maximum thermal efficiency, so that the efficiency
of the Stirling engine is the same as the Carnot cycle, that is, η = (Tmax −
Tmin )/Tmax . The principal advantage of the Stirling cycle over the Carnot cycle
lies in the replacement of two isentropic processes by two constant-volume
processes, which greatly increase the area of the P–V diagram. Therefore, to
obtain a reasonable amount of work from the Stirling cycle, it is not necessary
to resort to the impractically high pressures and swept volumes, as in the
Carnot cycle.
The torque–speed characteristic of a Stirling engine is relatively flat
as shown in Figure 3.18. This is particularly favorable for automotive
99
Internal Combustion Engines
150
Torque (Nm)
125
100
75
50
25
0
0
250 500 750 1000 1250 1500 1750 2000 2250 2500
Engine speed (rpm)
FIGURE 3.18 Torque/speed characteristics of typical Stirling engines.
applications where high torque at low speed is advantageous to achieve good
acceleration.
The performance and fuel consumption characteristic map for a Stirling
engine is shown in Figure 3.19.5 Compared with IC engines, the outstanding
feature of the Stirling engine is that the fuel consumption at part load is
2000
18.75 37.5 75
112.5 150 187.5 225 kW
1800
1600
Torque (Nm)
1400
1200
196
1000
203
800
207
600
213
227
240
267
293
400
200
0
0
200
400
600
800 1000 1200 1400 1600
Engine speed (rpm)
FIGURE 3.19 Performance and fuel consumption map for a four-cylinder Stirling engine for
traction applications.
100
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
higher compared to its optimum fuel consumption operating point, much
less than that of IC engines. This outstanding characteristic will lead to good
fuel economy for a vehicle that operates in partial load most of the time.
The exhaust emissions are low and easy to control because combustion
is isolated from the cycle changes in pressure and temperature experienced
by the working fluid. An excess of air of between 20% and 80%, continuous combustion, and no “quench” of flame by a “cold” metal surface also
help to ensure almost complete combustion. Because continuous combustion replaces the intermittent explosive nature of combustion in other piston
engines, the Stirling engine usually has low noise.
Other important advantages of the Stirling engine for its application of vehicles are the multifuel and self-starting characteristics. Because the combustion
is isolated from its working fluid, and only heat is transferred to the working
fluid from a heat source, a variety of fuels can be used.
However, Stirling engines also have some disadvantages. The main disadvantages have traditionally been high manufacture cost, a heavy and bulky
end result, and difficulty in controlling the power output accurately. Also it
has so far proven impossible to hold the leakage rate of hydrogen or helium
from the system down to a level that would be acceptable. The durability of
piston seals has been as much of a problem as the effectiveness of the seals to
prevent leakage.
In practical everyday automotive operations, there are some additional
shortcomings. First, EC prolongs the start-up time. Power cannot be drawn
from a cold engine until the heater head reaches a reasonable operating
temperature. Second, fuel energy is spent in raising the heater head to operating temperature, and when the engine is shut down, that stored energy is
dissipated without performing useful work.
3.6
Gas Turbine Engines
The gas turbine engine is a rotary, continuous IC engine where the fuel is
supplied to a burner and burned with an excess of compressed air (lean burn).
The hot combustion gases then expand and pass through a turbine, which
generates power and is transferred to the output shaft by gearing, as shown
in Figure 3.20.7
The gas turbine engine operates in accordance with the principles described
below:
1. Compression: Air enters the gas turbine and is compressed.
2. Heat exchange: Heat is drawn from the exhaust gases and communicated to the compressed air. There is recuperation of the exhaust gases’
energy that is otherwise lost.
101
Internal Combustion Engines
15°C, 1 bar
Compressor
Combustion
chamber
20°C, 4.54 bar
4.25 bar
1040°C
Compressor
turbine
Power
turbine
4.45 bar,
660°C
2.2 bar,
863°C
1 bar,
260°C
4.25 bar,
1040°C
Exhaust
Power output
Exhaust
FIGURE 3.20 Gas turbine engine with heat exchanger.
3. Combustion: Fuel is mixed with the hot air and ignited. The pressure
increases.
4. Expansion: The hot exhaust gases drive the turbine, thus releasing their
energy. The turbine turns the compressor and the output shaft.
Figures 3.21 and 3.22 show the bsfc map for a complete range of output power
and speed. The shape of the bsfc map of a turbine engine is quite different
from that of SI and CI engines. The speed of a turbine engine can start from
zero, and extend to its maximum. The drop in the fuel economy at partial load
from full load is much smaller than that of the SI and CI engines. The bsfc
values are more sensitive to speed than to power. Therefore, a turbine engine
favors running at constant speed, and load variations have less influence on
the fuel economy.
102
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
1.0
100%
bsfc, g/kWh
0.875
Power (per unit)
0.75
92%
245
0.625
0.50
253
83%
260
0.375
267
0.25
67%
320
0.125
0
75%
280
293
373 347
0
1000
58%
50%
2000 3000 4000 5000
Engine (rpm)
Gas
generator
6000
FIGURE 3.21 Fuel consumption characteristics of a Kronograd KTT gas turbine.
266
1.0
bsfc, gm/kW.h
Min.bsfc line
Limit curve
266
Power kW/(kW)max
0.8
272
285
0.6
290
296
0.4
302
333
0.2
0
339
423
544
0
0.2
0.4
0.6
0.8
1.0
Engine rpm/(rpm)max
FIGURE 3.22 Fuel consumption characteristics of a Chrysler upgraded turbine.
Internal Combustion Engines
103
The advantages of gas turbines include the following:
1. A very compact engine because of its high rotational speed.
2. Vibration-free operation due to the rotating movement.
3. Ability to operate on a wide variety of fuels.
4. Reduced HC and CO emissions compared to IC engines because the
combustion is continuous and therefore easier to control.
However, gas turbines suffer from some major disadvantages for automotive
applications:
1. High noise levels due to the quenching of gases by the turbine and
compressor.
2. High fuel consumption because the gas turbine does not scale down
efficiently: the efficiency of a dynamic compressor and turbine is low
for small sizes (below a few hundred kilowatts).
3. High rotating speeds are required for small (below a few hundred kilowatts) turbines to operate at their maximum efficiency. This requires
sophisticated materials to withstand the mechanical constraints.
4. The high temperatures continuously encountered by the turbine
require expensive materials that also have to withstand high rotational
speeds.
5. The efficiency is extremely speed dependent, thus requiring a speed
decoupling.
6. No torque output at low speed.
7. High cost of the heat exchanger.
3.7 Quasi-Isothermal Brayton Cycle Engines
The quasi-isothermal Brayton cycle engine (QIBCE) is an invention of Prof.
Mark Holtzapple at Texas A&M University. It is a variation of the gas turbine, with two major differences intended to improve on the most crippling
disadvantages of the gas turbine.
The most significant improvement is the compressor and turbine principle.
Instead of being of the dynamic type, and therefore bound to operate at very
high speed, these elements are of the positive displacement type that operate
effectively and efficiently at low speeds. The principle is that of a gear pump,
of the kind used in automobiles for oil.
The second improvement is the quasi-isothermal compression, which
requires less energy than the adiabatic compression in conventional Brayton
cycles. This is achieved by spraying water in the compressor. The liquid water
104
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
“captures” the heat resulting from the compression, while remaining as liquid
or slightly vaporizing. This keeps the overall flow temperature from rising,
therefore reducing the energy necessary for the compression.
The claimed advantages of the QIBCE engine are as follows:
1. Potential for very high efficiency
2. Compactness
3. Continuous combustion that results in low HC and CO emissions
4. Reduced noise emissions compared to a gas turbine because of the
absence of quenching of gases
5. Simplicity due to a reduced number of parts
The disadvantages are as follows:
1. Specialized and potentially expensive materials for the turbine
because it is permanently exposed to high temperatures. However,
these materials should be less stressed than in a gas turbine because
of the lower operating speeds
2. High cost of the heat exchanger
To this day, this engine is only a concept and a prototype must be implemented
to verify the claimed advantages.
References
1. J. B. Heywood, Internal Combustion Engine Fundamentals, McGraw-Hill Inc., New
York, 1988.
2. R. Stone, Introduction to Combustion Engines, Second Edition, Society of Automotive
Engineers (SAE), Warrendale, PA, 1992.
3. P. Duret, A New Generation of Two-Stroke Engines for The Future?, Edison’s Technip,
Paris, 1993.
4. G. P. Blair, Design and Simulation of Two Stroke Engines, Society of Automotive
Engineers (SAE), Warrendale, PA, 1996.
5. G. Walker, G. Reader, O. R. Fauvel, and E. R. Bingham, The Stirling Alternative—
Power System, Refrigerants and Heat Pumps, Gordon and Breach Science Publishers,
London, 1994.
6. M. J. Collie, Stirling Engine Design and Feasibility for Automotive Use, Noyes Data
Corporation, New Jersey, 1979.
7. J. P. O’Brien, Gas Turbines for Automotive Use, Noyes Data Corporation, New
Jersey, 1980.
4
Electric Vehicles
EVs use an electric motor for traction, and chemical batteries, fuel cells,
ultracapacitors, and/or flywheels for their corresponding energy sources. The
EV has many advantages over the conventional internal combustion engine
vehicle (ICEV), such as absence of emissions, high efficiency, independence
from petroleum, and quiet and smooth operation. The operational and fundamental principles in EVs and ICEVs are similar, as described in Chapter
2. There are, however, some differences between ICEVs and EVs, such as the
use of a gasoline tank versus batteries, ICE versus electric motor, and different transmission requirements. This chapter will focus on the methodology
of power train design and will investigate the key components, including
traction motor, energy storage, and so on.
4.1 Configurations of EVs
Previously, the EV was mainly converted from the existing ICEV by replacing
the IC engine and fuel tank with an electric motor drive and battery pack while
retaining all the other components, as shown in Figure 4.1. Drawbacks such as
its heavy weight, lower flexibility, and performance degradation have caused
the use of this type of EV to fade out. In its place, the modern EV is purposely
built, based on original body and frame designs. This satisfies the structure
requirements unique to EVs and makes use of the greater flexibility of electric
propulsion.1
A modern electric drive train is conceptually illustrated in Figure 4.2.1 The
drive train consists of three major subsystems: electric motor propulsion,
energy source, and auxiliary. The electric propulsion subsystem is comprised
of the vehicle controller, the power electronic converter, the electric motor,
mechanical transmission, and driving wheels. The energy source subsystem
involves the energy source, the energy management unit, and the energy
refueling unit. The auxiliary subsystem consists of the power steering unit,
the hotel climate control unit, and the auxiliary supply unit.
Based on the control inputs from the accelerator and brake pedals, the
vehicle controller provides proper control signals to the electronic power
105
106
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Mechanical
transmission
Electric
motor
drive
Electric
energy
storage
FIGURE 4.1 Primary EV power train.
converter, which functions to regulate the power flow between the electric
motor and energy source. The backward power flow is due to the regenerative braking of the EV and this regenerated energy can be restored into the
energy source, provided the energy source is receptive. Most EV batteries
as well as ultracapacitors and flywheels readily possess the ability to accept
regenerative energy. The energy management unit cooperates with the vehicle controller to control the regenerative braking and its energy recovery. It
also works with the energy refueling unit to control the refueling unit and
to monitor the usability of the energy source. The auxiliary power supply
Electric propulsion subsystem
Wheel
Brake
Vehicle
controller
Electronic
power
converter
Electric
motor
Mechanical
transmission
Accelerator
Wheel
Energy
management
unit
Energy
source
Energy
refueling
unit
Energy source
subsystem
Auxiliary
power
supply
Power
steering
unit
Steering
wheel
Hotel climate
control unit
Auxiliary subsytem
Mechanical link
Electric link
Control link
FIGURE 4.2 Conceptual illustration of a general EV configuration.1
107
Electric Vehicles
provides the necessary power with different voltage levels for all the EV
auxiliaries, especially the hotel climate control and power steering units.
There are a variety of possible EV configurations due to the variations in
electric propulsion characteristics and energy sources, as shown in Figure 4.3.1
a. Figure 4.3a shows the configuration of the first alternative, in which
an electric propulsion replaces the IC engine of a conventional vehicle
drive train. It consists of an electric motor, a clutch, a gearbox, and a
differential. The clutch and gearbox may be replaced by an automatic
transmission. The clutch is used to connect or disconnect the power of
the electric motor from the driven wheels. The gearbox provides a set
of gear ratios to modify the speed–power (torque) profile to match the
load requirement (refer to Chapter 2). The differential is a mechanical
device (usually a set of planetary gears), which enables the wheels of
(a)
(b)
M
GB
M
D
(c)
GB
D
(d)
FG
M
D
FG
M
(e)
M
FG
(f)
M
FG
M
M
FG
M
C: Clutch
D: Differential
FG: Fixed gearing
GB: Gearbox
M: Electric motor
FIGURE 4.3 Possible EV configuration: (a) conventional driveline with multigear transmission
and clutch, (b) single-gear transmission without need of a clutch, (c) integrated fixed gearing
and differential, (d) two separate motors and fixed gearing with their driveshaft, (e) direct drive
with two separate motors and fixed gearing, and (f) two separate in-wheel motor drives.1
108
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
both sides to be driven at different speeds when the vehicle runs along
a curved path.
b. With an electric motor that has a constant power in a long speed range
(refer to Chapter 2), a fixed gearing can replace the multispeed gearbox
and reduce the need for a clutch. This configuration not only reduces
the size and weight of the mechanical transmission, it also simplifies
the drive train control because gear shifting is not needed.
c. Similar to the drive train in (b), the electric motor, the fixed gearing,
and the differential can be further integrated into a single assembly
while both axles point at both driving wheels. The whole drive train
is further simplified and compacted.
d. In Figure 4.3d, the mechanical differential is replaced by using two
traction motors. Each of them drives one side wheel and operates at a
different speed when the vehicle is running along a curved path.
e. In order to further simplify the drive train, the traction motor can
be placed inside a wheel. This arrangement is the so-called in-wheel
drive. A thin planetary gear set may be employed to reduce the motor
speed and enhance the motor torque. The thin planetary gear set offers
the advantage of a high-speed reduction ratio as well as an inline
arrangement of the input and output shaft.
f. By fully abandoning any mechanical gearing between the electric
motor and the driving wheel, the out-rotor of a low-speed electric
motor in the in-wheel drive can be directly connected to the driving
wheel. The speed control of the electric motor is equivalent to the control of the wheel speed and hence the vehicle speed. However, this
arrangement requires the electric motor to have a higher torque to
start and accelerate the vehicle.
4.2
Performance of EVs
A vehicle’s driving performance is usually evaluated by its acceleration time,
maximum speed, and gradeability. In EV drive train design, proper motor
power rating and transmission parameters are the primary considerations
to meet the performance specification. The design of all these parameters
depends mostly on the speed–power (torque) characteristics of the traction
motor, as mention in Chapter 2, and will be discussed in this chapter.
4.2.1 Traction Motor Characteristics
Variable-speed electric motor drives usually have the characteristics shown
in Figure 4.4. At the low-speed region (less than the base speed as marked in
Figure 4.4), the motor has a constant torque. In the high-speed region (higher
109
Electric Vehicles
400
80
350
70
Power
Motor power (kW)
250
50
Torque
40
200
30
150
20
100
Base
speed
10
0
0
1000
2000
3000
Motor speed (rpm)
Motor torque (Nm)
300
60
50
4000
5000
FIGURE 4.4 Typical variable-speed electric motor characteristics.
than the base speed), the motor has a constant power. This characteristic is
usually represented by a speed ratio x, defined as the ratio of its maximum
speed to its base speed. In low-speed operation, voltage supply to the motor
increases with the increase of speed through the electronic converter while
the flux is kept constant. At the point of base speed, the voltage of the motor
reaches the source voltage. After the base speed, the motor voltage is kept
constant and the flux is weakened, dropping hyperbolically with increasing
speed. Hence, its torque also drops hyperbolically with increasing speed.2−4
Figure 4.5 shows the torque–speed profiles of a 60 kW motor with different
speed ratios x (x = 2, 4, and 6). It is clear that with a long constant power
region, the maximum torque of the motor can be significantly increased, and
hence vehicle acceleration and gradeability performance can be improved and
the transmission can be simplified. However, each type of motor inherently
has its limited maximum speed ratio. For example, a permanent magnet motor
has a small x (<2) because of the difficulty of field weakening due to the
presence of the permanent magnet. Switched reluctance motors may achieve
x > 6 and induction motors about x = 4.2,5
4.2.2 Tractive Effort and Transmission Requirement
The tractive effort developed by a traction motor on driven wheels and the
vehicle speed are expressed as
Ft =
Tm ig i0 ηt
rd
(4.1)
110
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
700
x=6
Motor torque (Nm)
600
500
x=4
400
300
x=2
200
100
0
0
500
1000 1500 2000 2500 3000 3500 4000 4500 5000
Motor speed (rpm)
FIGURE 4.5 Speed–torque profile of a 60-kW electric motor with x = 2, 4, and 6.
and
V=
πNm rd
(m/s),
30ig i0
(4.2)
where Tm and Nm are the motor torque output in N m and speed in rpm,
respectively, ig is the gear ratio of transmission, i0 is the gear ratio of final
drive, ηt is the efficiency of the whole driveline from the motor to the driven
wheels, and rd is the radius of the driven wheels.
The use of a multigear or single-gear transmission depends mostly on the
motor speed–torque characteristic. That is, at a given rated motor power,
if the motor has a long constant power region, a single-gear transmission
would be sufficient for a high tractive effort at low speeds. Otherwise, a
multigear (more than two gears) transmission has to be used. Figure 4.6
shows the tractive effort of an EV, along with the vehicle speed with a
traction motor of x = 2 and a three-gear transmission. The first gear covers the speed region of a–b–c, the second gear covers d–e–f, and the third
gear covers g–f–h. Figure 4.7 shows the tractive effort with a traction motor
of x = 4 and a two-gear transmission. The first gear covers the speed
region of a–b–c and the second gear d–e–f. Figure 4.8 shows the tractive
effort with a traction motor of x = 6 and a single-gear transmission. These
three designs have the same tractive effort versus vehicle speed profiles.
Therefore, the vehicles will have the same acceleration and gradeability
performance.
111
Electric Vehicles
8
7
a–b–c: 1st gear operation region
d–e–f: 2nd gear operation region
g–e–h: 3rd gear operation region
b
a
6
Tractive effort (kN)
Motor speed ratio: x = 2
5
e
d
Tractive
effort
4
c
Rolling resistance
+
Aerodynamic drag
3
g
f
1
0
Ft – Fr – Fw
2
Vb
0
20
40
60
h
80
100 120 140
Vehicle speed (km/h)
160 180
Maximum
speed
200
FIGURE 4.6 Tractive effort versus vehicle speed with a traction motor of x = 2 and three-gear
transmission.
8
7
a–b–c: 1st gear operation region
d–e–f: 2nd gear operation region
a
b
Motor speed ratio: x = 4
Tractive effort (kN)
6
5
d
Tractive
effort
e
4
Rolling resistance
+
Aerodynamic drag
3
1
0
Ft – Fr – Fw
2
Vb
0
20
40
60
c
80
100 120 140
Vehicle speed (km/h)
f
160 180
Maximum
speed
200
FIGURE 4.7 Tractive effort versus vehicle speed with a traction motor of x = 4 and two-gear
transmission.
112
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
8
7 a
Single gear transmission
Motor speed ratio: x = 6
b
Tractive effort (kN)
6
Tractive
effort
5
4
Rolling resistance
+
Aerodynamic drag
3
1
0
Ft – Fr – Fw
2
Vb
0
20
40
60
80
c
100
120
Vehicle speed (km/h)
140
160
180
200
Maximum
speed
FIGURE 4.8 Tractive effort versus vehicle speed with a traction motor of x = 6 and single-gear
transmission.
4.2.3 Vehicle Performance
Basic vehicle performance includes maximum cruising speed, gradeability,
and acceleration. The maximum speed of a vehicle can be easily found by the
intersection point of the tractive effort curve with the resistance curve (rolling
resistance plus aerodynamic drag), in the tractive effort versus vehicle speed
diagram shown in Figures 4.6 through 4.8. It should be noted that such an
intersection point does not exist in some designs, which usually use a larger
traction motor or a large gear ratio. In this case, the maximum vehicle speed
is determined by the maximum speed of the traction motor as
Vmax =
πNm max rd
(m/s),
30ig min i0
(4.3)
where Nm max is the allowed maximum rpm of the traction motor and ig min
is the minimum gear ratio of the transmission (highest gear).
Gradeability is determined by the net tractive effort of the vehicle, Ft−net
(Ft−net = Ft − Fr − Fw ), as shown in Figures 4.6 through 4.8. The gradeability
at mid- and high speeds is smaller than that at low speeds. The maximum
grade that the vehicle can overcome at the given speed can be calculated by
i=
Ft − (Fr + Fw )
Ft−net
=
,
Mg
Mg
(4.4)
113
Electric Vehicles
where Ft is the tractive effort on the driven wheels, Fr is the tire rolling
resistance, and Fw is the aerodynamic drag. However, at low speeds, the
gradeability is much larger. Calculations based on Equation 4.4 will cause
significant error; instead, Equation 4.5 should be used:
d − fr 1 − d2 + fr2
sin α =
.
1 + fr2
(4.5)
d = Ft − Fw /Mg is called the vehicle performance factor (refer to Chapter 2)
and fr is the tire rolling resistance coefficient.
Acceleration performance of a vehicle is evaluated by the time used to
accelerate the vehicle from a low speed V1 (usually zero) to a higher speed
(100 km/h for passenger cars). For passenger cars, acceleration performance
is more important than maximum cruising speed and gradeability, since it
is the acceleration requirement rather than the maximum cruising speed or
the gradeability that dictate the power rating of the motor drive. Referring to
Equation 2.58 and Figures 2.28 and 2.29, the acceleration time for an EV can
be expressed as
ta =
Vb
0
+
Pt /Vb − Mg
Vf
Vb
Pt /V − Mg
Mδ
dV
fr − 1/2ρa CD Af V 2
Mδ
dV,
fr − 1/2ρa CD Af V 2
(4.6)
where Vb and Vf are the vehicle base speed (as shown in Figures 4.6 through
4.8) and the final acceleration speed, respectively, and Pt is the tractive power
on the driven wheels transmitted from the traction motor corresponding to
the vehicle base speed. The first term on the right-hand side of Equation 4.6 is
in correspondence with the speed region less than the vehicle base speed; the
second term is in correspondence with the speed region beyond the vehicle
base speed.
It is difficult to obtain the analytical solution from Equation 4.6. For initial
evaluation of acceleration time versus tractive power, one can ignore rolling
resistance and aerodynamic drag and obtain
ta =
δM 2
Vf + Vb2 ,
2Pt
(4.7)
where the vehicle rotational inertial factor, δ, is a constant. The tractive power,
Pt , can then be expressed as
Pt =
δM 2
Vf + Vb2 .
2ta
(4.8)
114
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
It should be noted that the power rating obtained from Equation 4.8 is only the
power consumed for vehicle acceleration. To accurately determine the tractive power rating, the power consumed in overcoming the rolling resistance
and dynamic drag should be considered. The average drag power during
acceleration can be expressed as
P̄drag
1
=
ta
ta 1
Mg fr V + ρa CD Af V 3
2
dt.
(4.9)
0
Referring to Figures 2.28 and 2.29, the vehicle speed V may be expressed,
using time t, as
t
V = Vf
.
(4.10)
ta
Substituting Equation 4.10 into Equation 4.9 and integrating, one obtains
P̄drag =
2
1
Mg fr Vf + ρa CD Af Vf3 .
3
5
(4.11)
The total tractive power for accelerating the vehicle from zero to speed Vf in
ta seconds can be finally obtained as
Pt =
2
δM 2
1
Vf + Vb2 + Mg fr Vf + ρa CD Af Vf3 .
2ta
3
5
(4.12)
Equation 4.12 indicates that for a given acceleration performance, low vehicle base speed will result in a small motor power rating. However, the
power rating decline rate to the vehicle base speed reduction is not identical. Differentiating Equation 4.12 with respect to vehicle speed Vb , one
can obtain
δMv
dPt
=
Vb .
(4.13)
dVb
ta
Figure 4.9 shows an example of the tractive power rating and the power rating
decline rate to the vehicle speed reduction (dPt /dVb ) versus the speed factor
x. In this example, the acceleration time is 10 s, the vehicle mass is 1200 kg, the
rolling resistance coefficient is 0.01, the aerodynamic drag coefficient is 0.3,
and the front area is 2 m2 . This figure clearly indicates that a low x (high Vb )
reduction in Vb will result in significant decline in the power rating requirement. But with a high x (low Vb ), x > 5 for example, it is not so effective.
Figure 4.10 gives an example of the acceleration time and distance versus
vehicle speed, using Equation 4.6 and numerical methods.
115
Electric Vehicles
100
10
90
80
Tractive power (kN)
70
9
8
7
60
6
50
5
40
4
30
3
dPt/dVb
20
2
10
1
0
2
3
4
dPt/dVb (kW/(m/s2))
ta = 10 s
M = 1200 kg
fr = 0.01
CD = 0.3
Af = 2.0 m2
Tractive power, Pt
5
6
7
x, Max, speed/base speed
8
9
0
10
FIGURE 4.9 Power rating versus speed factor.
16
300
Pt = 63 kW, x = 4
M = 1200 kg, fr = 0.01,
14 C = 0.3, A = 2.0
m2
D
f
260
200
Time (s)
10
Time
8
160
6
120
Distance (m)
240
12
80
4
Distance
40
2
0
0
20
40
100
60
80
Final speed, Vf' (km/h)
120
0
140
FIGURE 4.10 Acceleration time and distance versus final speed.
4.3 Tractive Effort in Normal Driving
The vehicle performance described in the previous section dictates vehicle
capabilities with respect to speed, gradeability, and acceleration, thus
116
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
dictating the power capacity of the power train. However, in normal driving conditions these maximum capabilities are rarely used. During most of
the operation time, the power train operates with partial load. Actual tractive
effort (power) and vehicle speed vary widely with operating conditions such
as acceleration, deceleration, uphill and downhill motion, and so on. These
variations are associated with traffic environment as well as type of vehicle.
City and highway traffic conditions vary greatly, as do the different functionality of vehicles, such as passenger cars and vehicles with regular operation
routes and schedules.
It is difficult to describe the tractive effort and vehicle speed variations in
all actual traffic environments accurately and quantitatively. However, some
representative driving cycles (driving schedules) have been developed to
emulate typical traffic environments. These driving cycles are represented
by vehicle speeds versus operating time while driving on a flat road. Some
typical drive cycles are illustrated in Figure 4.12, which include (a) FTP75
urban cycle, (b) FTP75 highway cycle, (c) US06 cycle, which is a high-speed
and high-acceleration drive cycle, (d) J227a schedule B, (e) J227a schedule C,
and (f ) J227a schedule D. The J227a series are recommended by the Society of
Automotive Engineers in the United States6 and is applied in the evaluation
of EVs and batteries.
In a specific drive cycle, the tractive effort of a vehicle can be expressed as
1
dV
.
Ft = Mg fr cos α + ρa CD Af V 2 + Mδ
2
dt
(4.14)
In a short time period, the speed is assumed to be linear with time, and
acceleration is constant as shown in Figure 4.11. The acceleration, dV/dt in a
driving cycle, can be obtained by
Vk+1 − Vk
dV
=
dt
tk+1 − tk
Speed
(k = 1, 2, . . . , n, n − total number of points).
Vk+1
Vk
tk tk+1
Time
FIGURE 4.11 Acceleration being assumed constant with a short time period.
(4.15)
117
(a)
Vehicle speed (km/h)
Electric Vehicles
100
80
60
40
20
0
FTP 75 urban
0
200
400
600 800 1000 1200 1400
Time (s)
7
1st gear
6
FTP 75 urban
Tractive effort (kN)
5
4 2nd gear
3 3rd gear
2
1
0
–1
(b)
Vehicle speed (km/h)
–2
100
80
60
40
20
0
0
FTP 75 highway
0
7
Tractive effort (kN)
20 40 60 80 100 120 140 160 180 200
Vehicle speed (km/h)
100 200 300 400 500 600 700 800
Time (s)
1st gear
6
FTP 75 highway
5
4 2nd gear
3
2 3rd gear
1
0
–1
–2
0
20 40 60 80 100 120 140 160 180 200
Vehicle speed (km/h)
FIGURE 4.12 Speed profile and tractive effort in different representative drive cycles; operating
points are marked by ‘+’: (a) FTP75 urban, (b) FTP75 highway, (c) US06, (d) J227a schedule B,
(e) J227a schedule C, and (f ) J227a schedule D.
118
(c)
Vehicle speed (km/h)
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
150
US06
100
50
0
0
100
200
300
400
Time (s)
500
600
8
US06
Tractive effort (kN)
6 1st gear
2nd gear
4
3rd gear
2
0
–2
(d)
Vehicle speed (km/h)
–4
0
20 40 60 80 100 120 140 160 180 200
Vehicle speed (km/h)
40
J227a Schedule B
30
20
10
0
0
10
20
30
40
Time (s)
50
60
70
7
Tractive effort (kN)
6
1st gear
J227a Schedule B
5
4
2nd gear
3
3rd gear
2
1
0
–1
–2
FIGURE 4.12 Continued.
0
20 40 60 80 100 120 140 160 180 200
Vehicle speed (km/h)
119
(e)
Vehicle speed (km/h)
Electric Vehicles
50
40
30
20
10
0
J227a Schedule C
0
10
20
30
7
6 1st gear
60
70
80
J227a Schedule C
5 2nd gear
Tractive effort (kN)
40 50
Time (s)
4
3
3rd gear
2
1
0
–1
–2
0
20 40 60 80 100 120 140 160 180 200
Tractive effort (kN)
(f)
Vehicle speed (km/h)
Vehicle speed (km/h)
FIGURE 4.12 Continued.
80
J227a Schedule D
60
40
20
0
0
20
40
60
80
Time (s)
100
120
140
7
1st gear
J227a Schedule D
6
5
4 2nd gear
3 3rd gear
2
1
0
–1
–2
–3
0 20 40 60 80 100 120 140 160 180 200
Vehicle speed (km/h)
120
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
14.0
Time (%)
10.5
7.0
3.5
0
2204
1852
1500
1148
Trac
ti
796
444
ve eff
92
–260
ort (
–612
N)
–964
0
9
18
27
45
54
63
peed
cle s
Vehi
36
72
81
90
h)
(km/
FIGURE 4.13 Time distribution on vehicle speed and tractive effort in an FTP75 urban cycle.
By using Equation 4.14, the tractive efforts in any instant in a driving cycle can
be calculated, as shown in Figure 4.12. The operating points of tractive effort
versus vehicle speed scatter over the plane, and clearly show the operating
area in which the power train operates most of the time. Furthermore, the time
distribution of vehicle speed and tractive effort can be generated as shown
in Figure 4.13. This time distribution information is very helpful for power
train design, in which the most efficient region of the power train is designed
to overlap the greatest operation time area.
4.4 Energy Consumption
In transportation, the unit of energy is usually kilowatt-hour (kWh) rather
than joule or kilojoule ( J or kJ). The energy consumption per unit distance
in kWh/km is generally used to evaluate the vehicle energy consumption.
However, for ICEVs the commonly used unit is a physical unit of fuel volume
per unit distance, such as liters per 100 km (L/100 km). In the United States,
the distance per unit volume of fuel is usually used; this is expressed as miles
per gallon (mpg). On the other hand, for battery-powered EVs, the original
energy consumption unit in kWh, measured at the battery terminals, is more
suitable. The battery energy capacity is usually measured in kWh and the
driving range per battery charge can be easily calculated. Similar to ICEVs,
121
Electric Vehicles
L/100 km (for liquid fuels) or kg/100 km (for gas fuels such as hydrogen) or
mpg or miles per kilogram is a more suitable unit of measurement for vehicles
that use gaseous fuels.
Energy consumption is an integration of the power output at the battery
terminals. For propelling, the battery power output is equal to the resistance
power and power losses in the transmission and motor drive, including power
losses in the electronics. The power losses in transmission and motor drive
are represented by their efficiencies ηt and ηm , respectively. Thus, the battery
power output can be expressed as
Pb-out
V
1
dV
2
Mg( fr + i) + ρa CD Af V + Mδ
=
.
ηt η m
2
dt
(4.16)
Here, the nontraction load (auxiliary load) is not included. In some cases, the
auxiliary loads may be too significant to be ignored and should be added to
the traction load. When regenerative braking is effective on an EV, a part of
the braking energy—wasted in conventional vehicles—can be recovered by
operating the motor drive as a generator and restoring it into the batteries. The
regenerative braking power at the battery terminals can also be expressed as
Pb-in =
αV
1
dV
Mg( fr + i) + ρa CD Af V 2 + Mδ
,
ηt η m
2
dt
(4.17)
where road grade i or acceleration dV/dt or both are negative, and α (0 <
α < 1) is the percentage of the total braking energy that can be regenerated
400
350
Motor torque (Nm)
300
250
200
150
90%
100
88%
86%
83%
78%
70%
50
0
60%
0
500
1000 1500 2000 2500 3000 3500 4000 4500 5000
Motor speed (rpm)
FIGURE 4.14 Typical electric motor efficiency characteristics.
122
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
by the electric motor, called the regenerative braking factor. The regenerative
braking factor α is a function of the applied braking strength and the design
and control of the braking system, which will be discussed in detail in the
later chapters. The net energy consumption from the batteries is
Eout =
Pb-out dt +
traction
Pb-in dt.
(4.18)
braking
It should be noted that the braking power in Equation 4.17 has a negative
sign. When the net battery energy consumption reaches the total energy in
the batteries, measured at their terminal, the batteries are empty and need to
be charged. The traveling distance between two charges (usually called effective travel range) is determined by the total energy carried by the batteries,
the resistance power, and the effectiveness of the regenerative braking (α).
The efficiency of a traction motor varies with its operating points on the
speed–torque (speed–power) plane as shown in Figure 4.14, where the most
efficient operating area exists. In power train design, this area should overlap
or at least be as close as possible to the area of the greatest operation as
mentioned in the previous section.
References
1. C. C. Chan and K. T. Chau, Modern Electric Vehicle Technology, Oxford University
Press, New York, 2001.
2. Y. Gao, H. Maghbelli, M. Ehsani, G. Frazier, J. Kajs, and S. Bayne, “Investigation
of proper motor drive characteristics for military vehicle propulsion,” Society of
Automotive Engineers (SAE) Journal, Paper No. 2003-01-2296, Warrendale, PA, 2003.
3. Z. Rahman, M. Ehsani, and K. Butler, “An investigation of electric motor
drive characteristics for EV and HEV propulsion systems,” Society of Automotive
Engineers (SAE) Journal, Paper No. 2000-01-3062, Warrendale, PA, 2003.
4. Z. Rahman, M. Ehsani, and K. Butler, “Effect of extended-speed, constant-power
operation of electric drives on the design and performance of EV-HEV propulsion
system,” Society of Automotive Engineers (SAE) Journal, Paper No. 2000-01-1557,
Warrendale, PA, 2003.
5. K. M. Rahman and M. Ehsani, “Performance analysis of electric motor drives
for electric and hybrid electric vehicle application,” IEEE Power Electronic in
Transportation, 49–56, ISBN 0-7803-3292-X, 1996.
6. D. A. J. Rand, R. Woods, and R. M. Dell, Batteries for Electric Vehicles, Research
Studies Press, Ltd, Hertfordshire, UK, 1998.
5
Hybrid Electric Vehicles
Conventional vehicles with IC engines provide good performance and
long operating range by utilizing the high-energy-density advantages of
petroleum fuels. However, conventional IC engine vehicles have the disadvantages of poor fuel economy and environmental pollution. The main
reasons for their poor fuel economy are (1) mismatch of engine fuel efficiency
characteristics with the real operation requirement (refer to Figures 2.34 and
2.35); (2) dissipation of vehicle kinetic energy during braking, especially while
operating in urban areas; and (3) low efficiency of hydraulic transmission in
current automobiles in stop-and-go driving patterns (refer to Figure 2.21).
Battery-powered EVs, on the other hand, possess some advantages over
conventional IC engine vehicles, such as high-energy efficiency and zero
environmental pollution. However, the performance, especially the operation
range per battery charge, is far less competitive than IC engine vehicles, due
to the much lower energy density of the batteries than that of gasoline. HEVs,
which use two power sources(a primary power source and a secondary power
source), have the advantages of both IC engine vehicles and EVs and overcome their disadvantages.1,2 In this chapter, the basic concept and operation
principles of HEV power trains are discussed.
5.1 Concept of Hybrid Electric Drive Trains
Basically, any vehicle power train is required to (1) develop sufficient power
to meet the demands of vehicle performance, (2) carry sufficient energy onboard to support the vehicle driving a sufficient range, (3) demonstrate high
efficiency, and (4) emit few environmental pollutants. Broadly, a vehicle may
have more than one power train. Here, the power train is defined as the
combination of the energy source and the energy converter or power source,
such as the gasoline (or diesel)–heat engine system, the hydrogen–fuel cell–
electric motor system, the chemical battery–electric motor system, and so on.
A vehicle that has two or more power trains is called a hybrid vehicle. A
hybrid vehicle with an electrical power train is called an HEV. The drive train
of a vehicle is defined as the aggregation of all the power trains.
123
124
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
A hybrid vehicle drive train usually consists of no more than two power
trains. More than two power trains will make the drive train very complicated.
For the purpose of recapturing braking energy that is dissipated in the form
of heat in conventional IC engine vehicles, a hybrid drive train usually has a
power train that allows energy to flow bidirectionally. The other one is either
bidirectional or unidirectional. Figure 5.1 shows the concept of a hybrid drive
train and the possible different power flow routes.
A hybrid drive train can supply its power to the load by a selective power
train. There are many available patterns of operating two power trains to meet
the load requirement:
1. Power train 1 alone delivers its power to the load.
2. Power train 2 alone delivers its power to the load.
3. Both power train 1 and power train 2 deliver their power to the load
simultaneously.
4. Power train 2 obtains power from the load (regenerative braking).
5. Power train 2 obtains power from power train 1.
6. Power train 2 obtains power from power train 1 and the load
simultaneously.
7. Power train 1 delivers power to the load and to power train 2
simultaneously.
Power train (1)
(unidirectional)
Energy
source
(1)
Energy
converter
(1)
∑
Power train (2)
(bidirectional)
Energy
source
(2)
Energy
converter
(2)
Power flow while propelling
Power flow while charging power train (2)
FIGURE 5.1 Conceptual illustration of a hybrid electric drive train.
Load
Hybrid Electric Vehicles
125
8. Power train 1 delivers its power to power train 2, and power train 2
delivers its power to the load.
9. Power train 1 delivers its power to the load, and the load delivers the
power to power train 2.
In the case of hybridization with a gasoline (diesel)–IC engine (power train
1) and a battery–electric machine (power train 2), pattern (1) is the enginealone propelling mode. This may be used when the batteries are almost
completely depleted and the engine has no remaining power to charge the
batteries, or when the batteries have been fully charged and the engine is
able to supply sufficient power to meet the power demands of the vehicle.
Pattern (2) is the pure electric propelling mode, in which the engine is shut
off. This pattern may be used for situations where the engine cannot operate
effectively, such as very low speed, or in areas where emissions are strictly
prohibited. Pattern (3) is the hybrid traction mode and may be used when
large power is needed, such as during sharp accelerating or steep hill climbing. Pattern (4) is the regenerative braking mode, by which the kinetic or
potential energy of the vehicle is recovered through the electric motor functioning as a generator. The recovered energy is then stored in the batteries
and reused later on. Pattern (5) is the mode in which the engine charges the
batteries while the vehicle is at a standstill, coasting, or descending a slight
grade, in which no power goes into or comes from the load. Pattern (6) is
the mode in which both regenerating braking and the IC engine charge the
batteries simultaneously. Pattern (7) is the mode in which the engine propels
the vehicle and charges the batteries simultaneously. Pattern (8) is the mode
in which the engine charges the batteries, and the batteries supply power to
the load. Pattern (9) is the mode in which the power flows into the batteries from the heat engine through the vehicle mass. The typical configuration
of this mode is that the two power trains are separately mounted on the
front and rear axles of the vehicle, which will be discussed in the following
sections.
The abundant operation modes in a hybrid vehicle create much more flexibility over a single power train vehicle. With proper configuration and control,
applying a specific mode for a special operating condition can potentially
optimize the overall performance, efficiency, and emissions. However, in a
practical design, deciding which mode should be implemented depends on
many factors, such as the physical configuration of the drive train, power
train efficiency characteristics, load characteristics, and so on.
Operating each power train in its optimal efficiency region is essential for
the overall efficiency of the vehicle. An IC engine generally has the best efficiency operating region with a wide throttle opening. Operating away from
this region will cause low operating efficiency (refer to Figures 2.30, 2.32, 2.34,
2.35, and 3.6). On the other hand, efficiency suffering in an electric motor is
not as detrimental when compared to an IC engine that operates away from
its optimal region (refer to Figure 4.14).
126
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Load power
Dynamic power
Average power
=
0
Time
+
0
0
Time
Time
FIGURE 5.2 A load power is decomposed into steady and dynamic components.
The load power of a vehicle varies randomly in real operation due to
frequent acceleration, deceleration, and climbing up and down grades, as
shown in Figure 5.2. Actually, the load power is composed of two components: one is steady (average) power, which has a constant value, and the
other is dynamic power, which has a zero average. In designing the control
strategy of a hybrid vehicle, one power train that favors steady-state operation, such as an IC engine and fuel cell, may be used to supply the average
power. On the other hand, another power train, such as an electric motor,
may be used to supply the dynamic power. The total energy output from the
dynamic power train will be zero in a whole driving cycle. This implies that
the energy source of the dynamic power train does not lose energy capacity
at the end of the driving cycle. It functions only as a power damper.
In a hybrid vehicle, steady power may be provided by an IC engine, a
Stirling engine, a fuel cell, and so on. The IC engine or the fuel cell can be
much smaller than that in a single power train design because the dynamic
power is taken by the dynamic power source, and then can operate steadily in
its most efficient region. The dynamic power may be provided by an electric
motor powered by batteries, ultracapacitors, flywheels (mechanical batteries),
and their combinations.1,3
5.2
Architectures of Hybrid Electric Drive Trains
The architecture of a hybrid vehicle is loosely defined as the connection
between the components that define the energy flow routes and control ports.
Traditionally, HEVs were classified into two basic types: series and parallel.
It is interesting to note that, in 2000, some newly introduced HEVs could
not be classified into these kinds.4 Hence, HEVs are presently classified into
four kinds—series hybrid, parallel hybrid, series–parallel hybrid, and complex hybrid—that are functionally shown in Figure 5.3.5 Scientifically, the
classifications above are not very clear and may cause confusion. Actually,
in an HEV, there are two kinds of energy flowing in the drive train: one
is mechanical energy and the other is electrical energy. Adding two powers together or splitting one power into two at the power merging point
always occurs with the same power type, that is, electrical or mechanical,
127
Hybrid Electric Vehicles
(a)
(b)
Fuel
tank
IC
engine
Generator
Battery
IC
engine
Battery
Power
converter
Electric
motor
IC
engine
Mech.
coupler
Transmission
Power
converter
(elec.
coupler)
Mech.
coupler
Transmission
Electric
motor
(c)
Fuel
tank
Fuel
tank
(d)
IC
engine
Mech.
coupler
Fuel
tank
Generator
Battery
Power
converter
(elec.
coupler)
Transmission
Electric
motor
Power
converter
Battery
(elec.
coupler)
Motor/
generator
Power
converter
Transmission
Electric
motor
Electrical link
Hydraulic link
Mechanical link
FIGURE 5.3 Classifications of hybrid EVs. (a) Series (electrically coupling), (b) parallel
(mechanical coupling), (c) series–parallel (mechanical and electrical coupling), and (d) complex
(mechanical and electrical coupling).
not electrical and mechanical. So perhaps a more accurate definition for HEV
architecture may be to take the power coupling or decoupling features such
as an electrical coupling drive train, a mechanical coupling drive train, and a
mechanical–electrical coupling drive train.
Figure 5.3a functionally shows the architecture that is traditionally called
a series hybrid drive train. The key feature of this configuration is that two
electrical powers are added together in the power converter, which functions
as an electric power coupler to control the power flows from the batteries and
generator to the electric motor, or in the reverse direction, from the electric
motor to the batteries. The fuel tank, the IC engine, and the generator constitute the primary energy supply and the batteries function as the energy
bumper.
Figure 5.3b is the configuration that is traditionally called a parallel hybrid
drive train. The key of this configuration is that two mechanical powers are
added together in a mechanical coupler. The IC engine is the primary power
plant, and the batteries and electric motor drive constitute the energy bumper.
The power flows can be controlled only by the power plants—the engine and
electric motor.
Figure 5.3c shows the configuration that is traditionally called a series–
parallel hybrid drive train. The distinguished feature of this configuration
is the employment of two power couplers—mechanical and electrical. Actually, this configuration is the combination of series and parallel structures,
128
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
possessing the major features of both and more plentiful operation modes
than those of the series or parallel structure alone. On the other hand, it is
relatively more complicated and may be of higher cost.
Figure 5.3d shows a configuration of the so-called complex hybrid, which
has a similar structure to the series–parallel one. The only difference is that the
electric coupling function is moved from the power converter to the batteries
and one more power converter is added between the motor/generator and
the batteries.
We will concentrate more on the first three configurations—series, parallel,
and series–parallel.
5.2.1
Series Hybrid Electric Drive Trains (Electrical Coupling)
Fuel tank
Tractive
Effort
Torque
A series hybrid drive train is a drive train in which two electrical power
sources feed a single electrical power plant (electric motor) that propels the
vehicle. The configuration that is most often used is the one shown in Figure 5.4. The unidirectional energy source is a fuel tank and the unidirectional
energy converter (power plant) is an IC engine coupled to an electric generator. The output of the electric generator is connected to a power DC bus
through a controllable electronic converter (rectifier). The bidirectional energy
source is a battery pack connected to the power DC bus by means of a controllable, bidirectional power electronic converter (DC/DC converter). The
power bus is also connected to the controller of the electric motor. The traction motor can be controlled as either a motor or a generator, and in forward
or reverse motion. This drive train may need a battery charger to charge
the batteries by wall plug-in from a power grid. The series hybrid drive train
Speed
Engine
Generator
Power
Engine
operating
region
Speed
Electrical
coupler
Rectifier
Motor
controller
Traction
motor
Mech.
Trans.
Vehicle Speed
DC
DC
......
Battery
pack
Battery
charger
FIGURE 5.4 Configuration of a series hybrid electric drive train.
Traction
Battery charge
Hybrid Electric Vehicles
129
originally came from an EV on which an additional engine–generator is added
to extend the operating range that is limited by the poor energy density of
the batteries.
The drive train needs a vehicle controller to control the operation and
power flows based on the driver’s operating command through accelerator
and brake pedals and other feedback information from the components (not
shown in Figure 5.4, but for details see Figure 7.1). The vehicle controller will
control the IC engine through its throttle, electric coupler (controllable rectifier and DC/DC converter), and traction motor to produce the demanded
propelling torque or regenerative braking torque with one of the following
operation modes:
1. Pure electric traction mode: The engine is turned off and the vehicle is
propelled only from the batteries.
2. Pure engine traction mode: The vehicle traction power comes only from
the engine–generator, while the batteries neither supply nor accept
any power from the drive train. The electric machines serve as an
electric transmission from the engine to the driven wheels.
3. Hybrid traction mode: The traction powers are drawn from both the
engine–generator and the batteries, merging together in the electrical
coupler.
4. Engine traction with battery charging mode: The engine–generator
supplies power to charge the batteries and to propel the vehicle
simultaneously. The engine–generator power is split in the electric
coupler.
5. Regenerative braking mode: The engine–generator is turned off and the
traction motor is operated as a generator powered by the vehicle
kinetic or potential energy. The power generated is charged to the
batteries and reused in later propelling.
6. Battery charging mode: The traction motor receives no power and the
engine–generator is operated only to charge the batteries.
7. Hybrid battery charging mode: Both the engine–generator and the
traction motor operate as generators in braking to charge the batteries.
Series hybrid drive trains offer several advantages:
1. There is no mechanical connection between the engine and the driven
wheels. Consequently, the engine can be potentially operated at any
point on its speed–torque (power) map. This distinguished advantage,
with a sophisticated power flow control, provides the engine with
opportunities to be operated always within its maximum efficiency
region, as shown in Figure 5.4. The efficiency and emissions of the
engine in this narrow region may be further improved by some special
design and control technologies, which is much easier than in the
130
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
whole operating domain. Furthermore, the mechanical decoupling
of the engine from the driven wheels allows the use of high-speed
engines, where it is difficult to directly propel the wheels through a
mechanical link, such as gas turbines or power plants that have slow
dynamic responses (e.g., Stirling engine, etc.).
2. Because electric motors have a torque–speed profile that is very close
to the ideal for traction, as shown in Figures 2.12, 2.14, and 4.4, the drive
train may not need multigear transmission, as discussed in Chapter
3. Therefore, the structure of the drive train can be greatly simplified
and is of less cost. Furthermore, two motors may be used, each powering a single wheel, and the mechanical differential can be removed.
Such an arrangement also has the following advantages of decoupling the speeds of two wheels, a similar function of a mechanical
differential, and an additional function of antislip similar to the conventional traction control. Furthermore, four in-the-wheel motors may
be used, each one driving a wheel. In such a configuration, the speed
and torque of each wheel can be independently controlled. Consequently, the drivability of the vehicle can be significantly enhanced.
This is very important for off-road vehicles which usually operate on
difficult terrain, such as ice, snow, and soft ground.
3. The control strategy of the drive train may be simple, compared
to other configurations, because of its fully mechanical decoupling
between the engine and wheels.
However, series hybrid electric drive trains have some disadvantages,
such as the following:
1. The energy from the engine changes its form twice to reach its
destination—driven wheels (mechanical to electrical in the generator
and electrical to mechanical in the traction motor). The inefficiencies
of the generator and traction motor may cause significant losses.
2. The generator adds additional weight and cost.
3. Because the traction motor is the only power plant propelling the
vehicle, it must be sized to produce enough power for optimal vehicle
performance in terms of acceleration and gradeability.
The design and control principle of a series HEV will be discussed in
Chapter 7.
5.2.2
Parallel Hybrid Electric Drive Trains (Mechanical Coupling)
A parallel hybrid drive train is a drive train in which the engine supplies its
mechanical power directly to the driven wheels in a manner similar to a conventional IC engine vehicle. The engine is assisted by an electric motor that is
mechanically coupled to the driveline. The powers of the engine and electric
131
Hybrid Electric Vehicles
Fuel tank
Final drive
and differential
Mechanical
coupler
Engine
Mechanical
Transmission
Battery
Motor
Controller
......
Battery
charger
Traction
Battery charge
FIGURE 5.5 Configuration of a parallel hybrid electric drive train.
motor are coupled together by mechanical coupling, as shown in Figure 5.5.
The distinguished feature of this architecture is that two mechanical powers from the engine and electric motor are added together by a mechanical
coupler.
All the possible operating modes mentioned in the series hybrid drive train
are still effective. The major advantages of the parallel hybrid drive train over
the series one are the following: (1) both the engine and the electric motor
directly supply torques to the driven wheels and no energy form conversion occurs, thus the energy loss may be less; and (2) it is compact because
there is no need for an additional generator and the traction motor is smaller
than in series. Its major disadvantage is the mechanical coupling between the
engine and the driven wheels, since then the engine operating points cannot
be fixed in a narrow speed and torque region. Another disadvantage may be
the complex structure and control.
Generally, mechanical coupling consists of torque coupling and speed coupling. In torque coupling, the mechanical coupler adds the torques of the
engine and motor together and delivers the total torque to the driven wheels.
The engine and motor torque can be independently controlled. But the speeds
of the engine, motor, and vehicle are linked together with a fixed relationship
and cannot be independently controlled because of the power conservation
constraint. Similarly in speed coupling, the speeds of the engine and motor
can be added together and all the torques are linked together and cannot
be independently controlled. The details of these two kinds of mechanical
coupler are described hereafter.
132
5.2.2.1
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Parallel Hybrid Drive Train with Torque Coupling
5.2.2.1.1 Torque-Coupling Devices
Figure 5.6 conceptually shows a mechanical torque coupling, which is a threeport, two-degree-of-freedom mechanical device. Port 1 is a unidirectional
input and ports 2 and 3 are bidirectional input or output, but both are not
input at the same time. Here input means the energy flow into the device and
output means the energy flow out of the device. In a hybrid vehicle application, port 1 is connected to an IC engine directly or through a mechanical
transmission. Port 2 is connected to the shaft of an electric motor directly or
through a mechanical transmission. Port 3 is connected to the driven wheels
through a mechanical linkage.
If the losses are ignored and in steady state, the power input to the torque
coupler is always equal to the power output from it. Suppose here port 2
(electric motor) is in propelling, that is, input. The power output to the vehicle
wheels is:
T3 ω3 = T1 ω1 + T2 ω2 .
(5.1)
The torque coupler can be expressed as
T3 = k1 T1 + k2 T2 ,
(5.2)
where k1 and k2 are the structural parameters of the torque coupler, which are
described by the gear ratios and usually are constant when the device design
is fixed. For the torque coupler, T3 is load torque and T1 and T2 are propelling
torques that are independent of each other and can be independently controlled. However, due to the constraint of Equation 5.1, the angular velocities
ω1 , ω2 , and ω3 are linked together and cannot be independently controlled,
as expressed by
ω3 =
ω1
ω2
=
.
k1
k2
(5.3)
Figure 5.7 shows some common mechanical torque-coupling devices.
Port 1
T1, w 1
Mechanical
torque
coupling
Port 2
FIGURE 5.6 Torque-coupling device.
T2, w 2
T3, w 3
Port 3
133
Hybrid Electric Vehicles
Gear Box
w1 T1
w1 T1
Z1
Z1
w3 T3
Z3
w2 T2
w2 T2
w3 T3
Z2
Z2
k1 =
z
z3
, k2 = 3
z1
z2
z2
, k2 = 1
z1
k1 =
Z1, Z2, Z3---Tooth numbers of the gears
Z1, Z2---Tooth numbers of the gears
Pulley or chain assembly
w1
r1
T1
r3
r2
k1 =
T1
T3
w3
w2
r1
w1
w2
T2
r2
w3 T3
r4
T2
r
r2
, k2 = 3
r1
r4
k1 =
r1, r2, r3 and r4 ---Radii of the pulleys
r2
, k2 = 1
r1
r1, and r2 ---Radii of the pulleys
Shaft
w2, T2
w1
T1
w3
T3
k1 = 1
k2 = 1
Rotor
Stator
FIGURE 5.7 Commonly used mechanical torque-coupling devices.
5.2.2.1.2 Drive Train Configurations with Torque Coupling
Torque couplers can be used to constitute hybrid drive trains with many different configurations. Based on the torque coupler used, a two- or one-shaft configuration may be constituted. In each, transmission may be placed in different
positions with different gears, resulting in various tractive characteristics.
134
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Te
Clutch
Torque
coupling
Engine
Tm
Trans. 1
we
Motor
Trans. 2
wm
Batteries
Motor
controller
FIGURE 5.8 Two-shaft configuration.
A good design will depend mostly on the tractive requirements, engine size,
motor size, and motor speed–torque characteristics.
Figure 5.8 shows a two-shaft configuration, in which two transmissions are
used. One is placed between the engine and the torque coupler and the other
between the motor and the torque coupler. Both transmissions may be singleor multigear. Figure 5.9 shows the general tractive effort–speed profiles of
a vehicle with different transmission gears. It is evident that two multigear
transmissions produce many tractive effort profiles. The performance and
overall efficiency of the drive train may be superior to other designs, because
two multigear transmissions provide more opportunities for both the engine
and the electric traction system (electric machine and batteries) to operate in
their optimum region. This design also provides great flexibility in the design
of the engine and electric motor characteristics. However, two multigear transmissions will significantly complicate the drive train and increase the burden
of the control system for selecting the proper gear in each transmission.6,7
In Figure 5.8, a multigear transmission, 1, and a single-gear transmission, 2,
may be used. Referring to the relative positions of transmissions and the electric motor, this configuration is referred to as a pretransmission configuration
(the electric motor is in front of the transmission). The tractive effort–speed
profiles are shown in Figure 5.9b. In the design of a hybrid drive train, the
maximum tractive effort with this transmission arrangement may be sufficient
for hill climbing performance; greater tractive effort would not be needed
because of the limitation of tire–ground contact adhesion. Utilizing a singlegear transmission, 2, takes advantage of the high torque of an electric machine
at low speed. The multigear transmission, 1, is used to overcome the disadvantages of the IC engine speed–torque characteristics (flat torque output in
its entire speed range). The multigear transmission, 1, also tends to improve
the operating efficiency of the engine and reduces the speed range of the
135
Hybrid Electric Vehicles
1 st gear of trans. 1
Tractive effort (N)
7000
6000
5000
4000
3000
2000
0
0
50
(c) 8000
100
Speed (km/h)
1st gear end here
2nd gear
end here
3000
2000
1st gear of trans. 1
5000
2nd gear of trans. 1
3rd gear of trans. 1
4000
3000
2000
50
3rd gear of trans. 2
1000
100
Speed (km/h)
150
8000
1 gears of trans. 1
1 gears of trans. 2
Motor speed ratio: x = 4
7000
2nd gear of trans. 2
4000
6000
0
0
150
1st gear of trans. 2
5000
3 gears of trans. 1
1 gears of trans. 2
Motor speed ratio: x = 4
1000
(d)
6000
8000
7000
1 gears of trans. 1
3 gears of trans. 2
Motor speed ratio: x = 2
7000
0
(b)
3rd gear 2nd gear
of trans. 1 of trans. 1
1000
Tractive effort (N)
3 gears of trans. 1
3 gears of trans. 2
Motor speed ratio: x = 2
1st gear of trans. 2
2nd gear of trans. 2
3rd gear of trans. 2
Tractive effort (N)
8000
Tractive effort (N)
(a)
6000
5000
4000
3000
2000
1000
0
50
100
Speed (km/h)
150
0
0
50
100
Speed (km/h)
150
FIGURE 5.9 Tractive effort along with vehicle speed with different transmission schemes: (a)
two multigear transmissions, (b) multigear engine transmission and single-gear motor transmission, (c) single-gear engine transmission and multigear motor transmission, and (d) both
single-gear transmissions.
vehicle, in which the electric machine must propel the vehicle alone, thus
preventing the batteries from quickly discharging.
In contrast to the above design, Figure 5.9c shows the tractive effort–speed
profile of a drive train that has a single-gear transmission, 1, for the engine
and a multigear transmission, 2, for the electric motor. This configuration is
considered to be an unfavorable design, because it does not use the advantages
of the two power plants.
Figure 5.9d shows the tractive effort–speed profile of a drive train that has
two single-gear transmissions. This arrangement results in a simple configuration and control. The limitation to the application of this drive train is
the maximum tractive effort of the drive train. When the power ratings of
the engine, electric motor, batteries, and transmission parameters are properly designed, this drive train would serve the vehicle with satisfactory
performance and efficiency.
Another configuration of a two-shaft parallel hybrid drive train is shown
in Figure 5.10, in which the transmission is located between the torque coupler and the drive shaft and may be categorized as a pretransmission. The
transmission amplifies the torques of both the engine and the electric motor
with the same scale. The design of the gear ratios k1 and k2 in the torque
coupler (Equation 5.3) allows the electric motor and the engine to reach their
136
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Clutch
Te
Torque
coupling
Ft
Engine
we
Tm
V
Trans
Motor
wm
Batteries
Motor
controller
FIGURE 5.10 Two-shaft configuration.
maximum speeds at the same time. This configuration would be suitable
when a relatively small engine and electric motor are used, where a multigear
transmission is needed to enhance the tractive effort at low speeds.
The simplest and most compact architecture of the torque-coupling parallel hybrid is the single-shaft configuration, where the rotor of the electric
motor functions as the torque coupler (k1 = 1 and k2 = 1 in Equations 5.2
and 5.3). The electric motor may be located either between the engine and
transmission as shown in Figure 5.11, referred to as pretransmission, or
between the transmission and final drive as shown in Figure 5.12, referred
to as post-transmission.
Batteries
Ft
Motor
controller
Engine
Trans.
V
Tm
Te
we
wm
FIGURE 5.11 Pretransmission single-shaft torque combination parallel hybrid electric drive
train.
137
Hybrid Electric Vehicles
Batteries
Ft
Motor
controller
V
Engine
Trans.
Te
Tm
we
wm
FIGURE 5.12 Post-transmission single-shaft torque combination parallel hybrid electric drive
train.
In the pretransmission configuration as shown in Figure 5.11, torques of
both the engine and motor are modified by the transmission. However, the
engine and motor are required to have the same speed range. This configuration is usually used in the case of a small motor, referred to as a mild hybrid
drive train, in which the electric motor functions as an engine starter, an
electrical generator, an engine power assistant, and for regenerative braking.
In the post-transmission configuration as shown in Figure 5.12, the transmission only modifies the engine torque while the motor directly delivers its
torque to the final drive without modification. This configuration may be used
in a drive train where a large electric motor with a long constant power region
is employed. The transmission is only used to change the engine operating
points for improving the vehicle performance and engine operating efficiency.
It should be noted that the batteries cannot be charged from the engine by
running the electric motor as a generator when the vehicle is at a standstill,
since the motor is rigidly connected to the driven wheels.
Another torque-coupling parallel hybrid drive train is the separated axle
architecture, in which one axle is powered by the engine and the other by the
electric motor, as shown in Figure 5.13. The tractive efforts produced by the
two power trains are added together through the vehicle chassis and road. The
operating principle is similar to the two-shaft configuration shown in Figure
5.8. Both transmissions for the engine and electric motor may be single-gear
or multigear. This configuration has similar tractive effort characteristics, as
shown in Figure 5.9.
The separated axle architecture offers some of the advantages of a conventional vehicle. It keeps the original engine and transmission unaltered and
adds an electrical traction system on the other axle. It is also a four-wheel
138
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Engine
Trans.
Trans.
Motor
Motor
controller
Batteries
FIGURE 5.13 Separated axle torque combination parallel hybrid electric drive train.
drive, which improves the traction on slippery roads and reduces the tractive
effort on a single tire.
However, electric machines and the eventual differential gear system
occupy a lot of space and may reduce the available passenger space and
luggage space. This problem may be solved if the transmission behind the
electric motor is single-gear and the single electric motor is replaced by two
small-sized electric motors that can be placed within the two driven wheels.
It should be noted that the batteries cannot be charged from the engine when
the vehicle is at a standstill.
5.2.2.2
Parallel Hybrid Drive Train with Speed Coupling
5.2.2.2.1 Speed-Coupling Devices
The powers produced by the two power plants may be coupled together
by adding their speeds, as shown in Figure 5.14. Similar to the mechanical
torque coupler, the speed coupler is also a three-port, two-degree-of-freedom
mechanical device. Port 1 may be connected to an IC engine with unidirectional energy flow. Ports 2 and 3 may be connected to an electric motor and
to the load (final drive), both with bidirectional energy flow.
T1, w1
Port 1
Mechanical
speed
coupler
Port 2
FIGURE 5.14 Speed coupling.
T2, w2
T3, w3
Port 3
139
Hybrid Electric Vehicles
The mechanical speed coupler has the property
ω3 = k1 ω1 + k2 ω2 ,
(5.4)
where k1 and k2 are constants associated with the structural and geometric
design. Among the three speeds, ω1 , ω2 , and ω3 , two of them are independent
of each other and can be controlled independently. Due to the constraint of
power conservation, the torques are linked together by
T3 =
T1
T2
=
,
k1
k2
(5.5)
In which the minimum torque determines the other two.
A typical speed-coupling device is the planetary gear unit as shown in
Figure 5.15. The planetary gear unit is a three-port unit consisting of sun gear,
ring gear, and yoke labeled 1, 2, and 3, respectively. The speed relationship in
the sun gear, ring gear, and yoke can be obtained as follows.
First, let the yoke be attached to a stationary frame, that is, ω3 = 0; the gear
ratio from the sun gear to ring gear is
ω32
3
=
i1−2
ω31
=−
R2
Z2
=− ,
R1
Z1
(5.6)
where ω31 and ω32 are the angular velocities of the sun gear and ring gear with
respect to the yoke (when the yoke is at a standstill), R1 and R2 are the radii of
the sun gear and ring gear, respectively, and Z1 and Z2 are the tooth numbers
of the sun gear and ring gear, respectively, which are proportional to the radii
of the sun gear and ring gear. Here, rotating in the counterclockwise direction is defined as positive angular velocity, whereas rotating in the clockwise
2
w2, T2
3
R1
R3
R2
w3
1
FIGURE 5.15 Planetary gear unit used as a speed-coupling device.
w1, T1
T3
140
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
direction is defined as negative angular velocity, as shown in Figure 5.15.
Equation 5.6 indicates that ω31 and ω32 have different rotating directions and
3
is negative. When the yoke is free from the stationary
thus the gear ratio i1−2
frame, the absolute angular velocities of the sun gear, ring gear, and yoke can
be expressed by
ω1 − ω3
3
= i1−2
.
(5.7)
ω2 − ω 3
Then we obtain
3
3
ω2 − (1 − i1−2
)ω3 = 0.
ω1 − i1−2
(5.8)
Conventionally, we are not accustomed to negative gear ratio. If we define
the gear ratio as a positive number, as
3
ig = −i1−2
=
R2
Z2
=
,
R1
Z1
(5.9)
Equation 5.8 can be rewritten as
ω1 + ig ω2 − (1 + ig )ω3 = 0
(5.10)
ig
1
ω1 +
ω2 .
1 + ig
1 + ig
(5.11)
or
ω3 =
Comparing Equation 5.11 with Equation 5.4, k1 = 1/(1 + ig ) and k2 = ig /
(1 + ig ) are obtained.
Similar to the definition of speed, when the torque acting on each element
of the planetary gear unit is defined to be positive in the counterclockwise
direction and negative in the clockwise direction, the total power into the
unit should be zero (output power is negative) when the loss inside the unit
is ignored, that is,
T1 ω1 + T2 ω2 + T3 ω3 = 0.
(5.12)
Combining Equations 5.11 and 5.12 yields
T3 = −(1 + ig )T1 = −
1 + ig
T2 .
ig
(5.13)
Equation 5.13 indicates that the torques acting on the sun gear, T1 , and ring
gear, T2 , always have the same sign (both positive or negative), and the torque
acting on the yoke, T3 , always has the direction opposite to T1 and T2 , as shown
in Figure 5.15.
When one element of the sun gear, ring gear, or yoke is locked to the stationary frame, that is, one degree of freedom is constrained, the unit will become
141
Hybrid Electric Vehicles
Element fixed
Speed
ig
Sun gear
w3 =
Ring gear
w3 =
Yoke
w1 = –ig w2
1 + ig
1
1 + ig
Torque
1 + ig
w2
T3 = –
w1
T3 = –(1 + ig) T1
T1 =
ig
T2
1 T
2
ig
FIGURE 5.16 Speed and torque relationships while one element is fixed.
a single-gear transmission (one input and one output). The speed and torque
relationships, while different elements are fixed, are shown in Figure 5.16.
Another interesting device used as a speed coupler is an electric motor with
a floating stator (called transmotor in this book), in which the stator, generally
fixed to a stationary frame in a traditional motor, is released to form a doublerotor machine—outer and inner rotor. The outer rotor, inner rotor, and air gap
are the three ports. Electric power is converted into mechanical power through
the air gap, as shown in Figure 5.17. The motor speed, in conventional terms,
is the relative speed of the inner rotor with respect to the outer rotor. Because
of the action and reaction effect, the torques acting on both rotors are always
the same and result in constants k1 = 1 and k2 = 1. The speed relationship can
be expressed as
ωor = ωir + ωoi ,
(5.14)
where ωio is the inner rotor speed relative to the outer rotor stator. The torque
relationship can be expressed as
Tir = Tos = Te .
(5.15)
Port 2
Port 1
Port 3
woi
wir
Tir
Te
wor
Inner rotor
Outer rotor
FIGURE 5.17 Transmotor used as a speed coupler.
Tor
142
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Lock 2
Clutch
Lock 1
Engine
Trans.
Motor
Batteries
Motor
controller
FIGURE 5.18 Hybrid electric drive train with speed coupling of a planetary gear unit.
5.2.2.2.2 Drive Train Configurations with Speed Coupling
Similar to the torque-coupling device, the speed-coupling units can be used to
constitute various hybrid drive trains. Figures 5.18 and 5.19 show two examples of hybrid drive trains with speed coupling using a planetary gear unit
and an electric transmotor. In Figure 5.18, the engine supplies its power to
the sun gear through a clutch and transmission. The transmission is used
to modify the speed–torque profile of the engine so as to match the traction requirements. The transmission may be multigear or single-gear based
on the engine speed–torque profile. The electric motor supplies its power
to the ring gear through a pair of gears. Lock 1 and lock 2 are used to lock
the sun gear and ring gear to the stationary frame of the vehicle in order to
implement different operation modes. The following operation modes can be
carried out:
1. Hybrid traction: When lock 1 and lock 2 are released (the sun gear
and ring gear can rotate), both the engine and electric machine supply positive speed and torque (positive power) to the driven wheels.
The output speed and torque from the yoke of the planetary unit is
described by Equations 5.11 and 5.13. That is, the rotational speed of
the yoke is the summation of the sun gear speed (engine speed, or
proportional to engine speed) and the ring gear speed (electric motor
speed, or proportional to motor speed). However, the output torque
from the yoke is proportional to the engine torque and motor torque.
Torque control will be studied in Chapter 8.
2. Engine-alone traction: When lock 2 locks the ring gear to the vehicle
frame and lock 1 is released, the engine alone supplies power to the
driven wheels. From Equations 5.11 and 5.13, the speed of the yoke is
proportional to the speed of the sun gear as ω3 = ω1 /(1 + ig ), and the
143
Hybrid Electric Vehicles
Batteries
Controller
Engine
Trans.
Clutch 1
Lock 1
Clutch 2
FIGURE 5.19 Hybrid electric drive train with speed coupling of an electric transmotor.
torque output from the yoke is proportional to the torque applied on
the sun gear from the engine as T3 = (1 + ig )T1 .
3. Motor-alone traction: When lock 1 locks the sun gear to the vehicle
frame (engine is shut off and engine clutch is disengaged) and lock
2 is released, only the electric motor supplies its power to the driven
wheels. From Equations 5.11 and 5.13, the speed of the yoke is proportional to the speed of the ring gear as ω3 = (ω2 ig )/(1 + ig ), and the
torque output from the yoke is proportional to the torque applied on
the ring gear by the electric motor as T3 = (1 + ig )/(ig T1 ).
4. Regenerative braking: The states of lock 1 and lock 2 are the same as
in motor-alone traction, the engine is also shut off, the engine clutch
is disengaged, and the electric machine is controlled in regenerating
mode (negative torque). The kinetic or potential energy of the vehicle
can be absorbed by the electric system.
5. Battery charging from the engine: The engine clutch and lock 1 and lock
2 are in the same state as in the hybrid traction mode. However, the
electric motor is controlled to rotate in the opposite direction, that
is, negative speed. Thus, the electric machine operates with positive
torque and negative speed (negative power) and absorbs energy from
the engine and delivers it to the batteries. In this case, the engine power
is split into two parts by decomposing its speed.
The drive train, consisting of the transmotor as shown in Figure 5.19, has a
similar structure as that in Figure 5.18. Lock 1 and clutch 2 are used to lock
the outer rotor to the vehicle frame and the outer rotor to the inner rotor,
respectively. This drive train can fulfill all the operation modes mentioned
above. The operating modes analysis is left to the readers.
Figure 5.20 shows an implementation of speed coupling with a transmotor.
Clutch 1 is the substitution for clutch 1 as shown in Figure 5.19, clutch 2 has
144
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Batteries
Housing Clutch 1 Clutch 2
Outer Inner
rotor rotor Clutch 3
Motor
controller
Shaft connected
to engine
Output shaft
FIGURE 5.20 Implementation of speed coupling with a transmotor.
the same function as clutch 2 in Figure 5.19, and clutch 3 has the same function
as lock 1 in Figure 5.19.
The main advantage of the hybrid drive train with speed coupling is that
the speed of two power plants is decoupled from vehicle speed. Therefore, the
speed of both power plants can be chosen freely. This advantage is important
to power plants, such as the Stirling engine and the gas turbine engine, in
which, the operating efficiencies are sensitive to speed and less sensitive to
torque.
5.2.2.3
5.2.2.3.1
Hybrid Drive Trains with Both Torque and Speed Coupling
With Optional Coupling Mode
By combining torque and speed coupling, one may establish a hybrid drive
train in which torque- and speed-coupling states can be alternatively chosen.
Figure 5.21 shows such a drive train.8 When the torque-coupling operation
mode is chosen, lock 2 locks the ring gear of the planetary unit to the vehicle
frame while clutches 1 and 3 are engaged and clutch 2 is disengaged. The
power of the engine and the electric motor are added together by adding
their torques together through gear Za , Zb and clutch 3 to the sun gear shaft.
In this case, the planetary gear unit functions only as a speed reducer. The
gear ratio from the sun gear to yoke, defined as ω1 /ω3 , equals (1 + ig ). This
is a typical parallel hybrid drive train with torque coupling.
When the speed-coupling mode is chosen as the current operating mode,
clutches 1 and 2 are engaged, whereas clutch 3 is disengaged, and locks 1 and
2 release the sun gear and ring gear. The speed of the yoke, connected to the
145
Hybrid Electric Vehicles
Clutch 1
Lock 2
Lock 1
Clutch 3
Zr
Zd
Zb
Zp
Zs
Engine
Trans.
Motor
Batteries
Motor
controller
Za
Zc
Clutch 2
FIGURE 5.21 Alternative torque and speed hybrid electric drive train with a planetary gear
unit.
drive wheels, is a combination of engine speed and motor speed (refer to Equation 5.11). But the engine torque, the electric motor torque, and torque on the
driven wheels are kept in a fixed relationship as described by Equation 5.13.
With the option to choose the power-coupling mode (torque or speed coupling), the power plant has more opportunities to choose its operation manner
and operation region so as to optimize their performance. For instance, at low
vehicle speeds, the torque combination operation mode may be suitable for
high acceleration or hill climbing. On the other hand, at high vehicle speeds,
the speed combination mode would be used to keep the engine speed in its
optimal region.
The planetary gear unit and the traction motor in Figure 5.21 can be replaced
by a transmotor to constitute a similar drive train as shown in Figure 5.22.
When clutch 1 is engaged to couple the output shaft of the transmission to
the inner rotor shaft of the transmotor, clutch 2 is disengaged to release the
engine shaft from the inner rotor of the transmotor and the lock is activated
to fix the outer rotor of the transmotor to the vehicle frame. The drive train
then works in the torque-coupling mode. On the other hand, when clutch 1
is disengaged and clutch 2 is engaged and the lock is released, the drive train
works in the speed-coupling mode.
The distinguishable characteristic of the above hybrid drive trains is that
the drive train can optionally choose the best coupling mode in different
driving situations so as to achieve the best vehicle performance and efficiency.
However, they cannot run on both coupling modes at the same time, since
only two power plants are available.
146
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Batteries
Motor
controller
Engine
Trans.
Clutch 2
Clutch 1
Lock
FIGURE 5.22 Alternative torque- and speed-coupling hybrid electric drive train with a
transmotor.
5.2.2.3.2 With Both Coupling Modes
By adding another power plant, a hybrid drive train with both speed- and
torque-coupling modes at the same time can be realized. A good example is
the one developed and implemented in the Toyota Prius by Toyota Motor
Company.9 This drive train is schematically illustrated in Figure 5.23. The
drive train uses a planetary gear unit as the speed-coupling device and a set
of fixed axle gears as the torque-coupling device. An IC engine is connected
to the yoke of the planetary gear unit, and a small motor/generator (few
kilowatts) is connected to the sun gear of a planetary gear unit to constitute the
speed-coupling configuration. The ring gear is connected to the driven wheels
through the axle-fixed gear unit (torque coupler). Meanwhile, a traction motor
is also connected to the fixed axle gear unit to constitute the torque-coupling
configuration.
From Equation 5.11, the rotational speed of the ring gear or gear Za , which
is proportional to vehicle speed, is related to the rotational speed of the engine
(yoke) and of the motor/generator (sun gear) and is expressed as
ωr =
1 + ig
ωice − ig ωm/g ,
ig
(5.16)
where ig is the gear ratio defined by Equation 5.9 and ωice and ωm/g are the
rotational speeds of the engine and motor/generator, respectively. The load
torque, acting on the ring gear of the planetary gear unit by gear Z4 , is related
to the engine torque and the motor/generator torque by
−Tr =
ig
Tice = −ig Tm/g .
1 + ig
(5.17)
Equation 5.17 indicates that the torque, acting on the sun gear, supplied by the
motor/generator has opposite direction to engine torque and same direction
147
Hybrid Electric Vehicles
as load torque on the ring gear. With low vehicle speed (small ωr ) and a
not very low engine speed (larger than its idle speed), the motor/generator
has to rotate in the positive direction (same direction as engine speed). In
this condition, the motor/generator operates with a negative power, that is,
generating. The power of the engine is split into two parts: one part goes to the
motor/generator and the other to vehicle load through the ring gear.This is
how the drive train gets its name of power-split hybrid drive train. However,
at high vehicle speed, while trying to maintain the engine speed below a
given speed, for high engine operating efficiency, the motor/generator may
be operated in negative speed, that is, rotating in the opposite direction to
engine speed. In this case, the motor/generator delivers positive power to
the planetary gear unit, that is, motoring. It becomes clear through the above
analysis that the major function of a motor/generator is to control engine
speed, that is, decouple engine speed from wheel speed.
The traction motor adds additional torque to the torque output from the
ring gear of the planetary gear unit with a torque-coupling mode through
gears Zc , Zb , Zd , and Ze , by which the engine torque is decoupled from the
vehicle load.
The small motor and the planetary gear unit in Figure 5.23 can be replaced
by an individual transmotor, as shown in Figure 5.24.10 This drive train
has very similar characteristics as the drive train shown in Figure 5.23.
Planetary
gear unit
Clutch
Motor/generator
Za
Lock
Engine
Zs
Motor
controller
Yoke
Zr
Zb
Traction
motor
Zd
Zc
Ze
Motor
controller
Batteries
FIGURE 5.23 Integrated speed- and torque-coupling hybrid electric drive train.
148
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Batteries
Motor
controller
Engine
Clutch
Transmission
Traction
motor
Motor
controller
FIGURE 5.24 Hybrid electric drive train with speed and torque coupling of a transmotor and
double shaft.
Another variation of the drive train in Figure 5.24 is the single-shaft design
as shown in Figure 5.25. A more compact design of the drive train in
Figure 5.25 is to integrate the transmotor and the traction motor together
as shown in Figure 5.26. The design and control may be more complicated
than the separated structure due to correlated magnetic field in the double
air gaps.
Batteries
Motor
controller
Engine
Motor
Clutch controller
Transmission
Transmotor
Traction motor
FIGURE 5.25 Hybrid electric drive train with speed and torque coupling of a transmotor and
single shaft.
149
Hybrid Electric Vehicles
Stator
Outer rotor
Inner rotor
Output shaft
Input shaft
Clutch
Motor
controller
Trans-motor
controller
+
–
Batteries
FIGURE 5.26 Integrated transmotor and traction motor.
In the literature, the integrated or separated transmotor and traction
motor in Figures 5.25 and 5.26 is called electrical variable transmission
(EVS).11,12 This name is derived from the fact that engine speed is electrically
decoupled from vehicle speed by the speed-coupling device: the transmotor. The operating characteristics and drive train control will be discussed
later.
References
1. M. Ehsani, Y. Gao, and J. M. Miller, “Hybrid electric vehicles: Architecture and
motor drives,” Proceedings of the IEEE, Special issue on Electric, Hybrid Electric and
Fuel Cells Vehicle, Vol. 95, No. 4, April 2007.
2. M. Ehsani, K. L. Butler, Y. Gao, and K. M. Rahman, “Next generation passenger cars
with better range, performance, and emissions: The ELPH car concept,” Horizon
in Engineering Symposium, Texas A&M University Engineering Program Office,
College Station, Texas, September 1998.
3. M. Ehsani, The Electrically Peaking Hybrid System and Method, US Patent No.
5,586,613, December 1996.
150
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
4. K. Yamaguchi, S. Moroto, K. Kobayashi, M. Kawamoto, and Y. Miyaishi, “Development of a new hybrid system-dual system,” Society of Automotive Engineers
(SAE) Journal, Paper No. 960231, Warrendale, PA, 1997.
5. C. C. Chan and K. T. Chau, Modern Electric Vehicle Technology, Oxford University
Press, New York, 2001.
6. Y. Gao, K. M. Rahman, and M. Ehsani, “The energy flow management and battery
energy capacity determination for the drive train of electrically peaking hybrid,”
Society of Automotive Engineers (SAE) Journal, Paper No. 972647, Warrendale, PA,
1997.
7. Y. Gao, K. M. Rahman, and M. Ehsani, “Parametric design of the drive train of an
electrically peaking hybrid (ELPH) vehicle,” Society of Automotive Engineers (SAE)
Journal, Paper No. 970294, Warrendale, PA, 1997.
8. Y. Gao and M. Ehsani, New Type of Transmission for Hybrid Vehicle with Speed and
Torque Summation, US Patent pending.
9. Available at http://www.toyota.com, Toyota Motor Company, visited in September 2003.
10. Y. Gao and M. Ehsani, Series–Parallel Hybrid Drive Train with an Electric Motor of
Floating Stator and Rotor, US Patent pending.
11. M. J. Hoeijimakes and J. A. Ferreira, “The electrical variable transmission,” IEEE
on Industry Application, 42 (4), 1092–1100, July–August 2006.
12. S. Cui, Y. Cheng, and C. C. Chan, “A basic study of electrical variable transmission
and its application in hybrid electric vehicle,” IEEE on Vehicle Power and Propulsion
Conference, (VPPC), 2006.
6
Electric Propulsion Systems
Electric propulsion systems are at the heart of EVs and HEVs. They consist
of electric motors, power converters, and electronic controllers. The electric
motor converts the electric energy into mechanical energy to propel the vehicle or vice versa, to enable regenerative braking and/or to generate electricity
for the purpose of charging the on-board energy storage. The power converter
is used to supply the electric motor with proper voltage and current. The electronic controller commands the power converter by providing control signals
to it, and then controls the operation of the electric motor to produce proper
torque and speed, according to the command from the driver. The electronic
controller can be further divided into three functional units—sensor, interface circuitry, and processor. The sensor is used to translate the measurable
quantities, such as current, voltage, temperature, speed, torque, and flux, into
electric signals through the interface circuitry. These signals are conditioned
to the appropriate level before being fed into the processor. The processor
output signals are usually amplified via the interface circuitry to drive power
semiconductor devices of the power converter. The functional block diagram
of an electric propulsion system is illustrated in Figure 6.1.
The choice of electric propulsion systems for EVs and HEVs mainly depends
on a number of factors, including the driver’s expectation, vehicle constraints,
and energy source. The driver’s expectation is defined by a driving profile,
which includes the acceleration, maximum speed, climbing capability, braking, and range. The vehicle constraints, including volume and weight, depend
on the vehicle type, vehicle weight, and payload. The energy source relates
to batteries, fuel cells, ultracapacitors, flywheels, and various hybrid sources.
Thus, the process of identifying the preferred feature and package options for
electric propulsion has to be carried out at the system level. The interaction
of subsystems and the likely impacts of system trade-offs must be examined.
Differing from the industrial applications of motors, the motors used in EVs
and HEVs usually require frequent starts and stops; high rates of acceleration/deceleration; high torque and low-speed hill climbing; low torque and
high-speed cruising, and a very wide speed range of operation. The motor
drives for EVs and HEVs can be classified into two main groups, namely the
commutator motors and commutatorless motors, as illustrated in Figure 6.2.
Commutator motors mainly are the traditional DC motors, which include
151
152
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Energy storage
Electric controller
Power converter
Electric motor
Software
Hardware
Devises
Topology
CAD
Type
VVVF
FOC
MARC
STC
VSC
NNC
Fuzzy
m processor
m controller
DSP
Transputer
IGBT
MOSFET
GTO
MCT
BJT
Chopper
Inverter
PWM
Resonant
FEM
EM
Force
Thermal
Graphics
DC
IM
SRM
PMSM
PMBM
PMHM
Transmission
& differential
FIGURE 6.1 Functional block diagram of a typical electric propulsion system.1
series excited, shunt excited, compound excited, separately excited, and permanent magnets (PMs) excited motors. DC motors need commutators and
brushes to feed current into the armature, thus making them less reliable
and unsuitable for maintenance-free operation and high speed. In addition,
winding-excited DC motors have low specific power density. Nevertheless,
because of their mature technology and simple control, DC motor drives have
been prominent in electric propulsion systems.
Technological developments have recently pushed commutatorless electric motors into a new era. Advantages include higher efficiency, higher
power density, and lower operating cost. They are also more reliable and
maintenance-free compared to commutator DC motors; thus, commutatorless
electric motors have now become more attractive.
Induction motors are widely accepted as a commutatorless motor type
for EV and HEV propulsion. This is because of their low cost, high reliability, and maintenance-free operation. However, conventional control of
induction motors such as variable-voltage variable-frequency cannot provide
Motor drives
Commutator
Selfexcited
Series
Shunt
Commutatorless
Separatelyexcited
Field
excited
PM
excited
Induction
Woundrotor
Squirrel
cage
Synchronous
Woundrotor
PM
rotor
PM
Brushless
Switched
reluctance
Reluctance
FIGURE 6.2 Classification of electric motor drives for EV and HEV applications.
PM
Hybrid
Electric Propulsion Systems
153
the desired performance. With the advent of the power electronics and
microcomputer era, the principle of field-oriented control (FOC) or vector
control of induction motors has been accepted to overcome their control complexity due to their nonlinearity.2 However, these EV and HEV motors using
FOC still suffer from low efficiency at light loads and limited constant-power
operating range.
By replacing the field winding of conventional synchronous motors with
PMs, PM synchronous motors can eliminate conventional brushes, slip rings,
and field copper losses.3 Actually, these PM synchronous motors are also
called PM brushless AC motors, or sinusoidal-fed PM brushless motors,
because of their sinusoidal AC current and brushless configuration. Since
these motors are essentially synchronous motors, they can run from a
sinusoidal or pulsed waveform modulation supply (PWM supply) without
electronic commutation. When PMs are mounted on the rotor surface, they
behave as nonsalient synchronous motors because the permeability of PMs is
similar to that of air. By burying those PMs inside the magnetic circuit of the
rotor, the saliency causes an additional reluctance torque, which leads to facilitating a wider speed range at constant power operation. On the other hand,
by abandoning the field winding or PMs while purposely making use of the
rotor saliency, synchronous reluctance motors are generated. These motors
are generally simple and inexpensive, but with relatively low output power.
Similar to induction motors, these PM synchronous motors usually use FOC
for high-performance applications.3 Because of their inherently high power
density and high efficiency, they have been accepted as having great potential
to compete with induction motors for EV and HEV applications.
By virtually inverting the stator and rotor of PM DC motors (commutator),
PM brushless DC (BLDC) motors are generated. It should be noted that the
term “DC” may be misleading, since it does not refer to a DC current motor.
Actually, these motors are fed by rectangular AC current and hence are also
rectangular-fed PM brushless motors.4 The most obvious advantage of these
motors is the removal of brushes. Another advantage is the ability to produce
a large torque because of the rectangular interaction between current and
flux. Moreover, the brushless configuration allows more cross-sectional area
for the armature windings. Since the conduction of heat through the frame is
improved, an increase in electric loading causes higher power density. Different from PM synchronous motors, these PM BLDC motors generally operate
with shaft position sensors. Recently, sensorless control technologies have
been developed in the Power Electronics and Motor Drive Laboratory at Texas
A&M University.
Switched reluctance motors (SRMs) have been recognized to have considerable potential for EV and HEV applications. Basically, they are direct
derivatives of single-stack variable-reluctance stepping motors. SRMs have
the definite advantages of simple construction, low manufacturing cost,
and outstanding torque–speed characteristics for EV and HEV applications.
Although they possess simplicity in construction, this does not imply any
154
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
simplicity of their design and control. Because of the heavy saturation of
pole tips and the fringe effect of pole and slots, their design and control
are difficult and subtle. Traditionally, SRMs operate with shaft sensors to
detect the relative position of the rotor to the stator. These sensors are usually vulnerable to mechanical shock and sensitive to temperature and dust.
Therefore, the presence of the position sensor reduces the reliability of SRMs
and constrains some applications. Recently, sensorless technologies have been
developed in the Power Electronics and Motor Drive Laboratory—again,
at Texas A&M University. These technologies can ensure smooth operation
from zero speed to maximum speed.5 This will be discussed in detail in the
following sections.
6.1
DC Motor Drives
DC motor drives have been widely used in applications requiring adjustable
speed; good speed regulation; and frequent starting, braking, and reversing. Various DC motor drives have been widely applied to different electric
traction applications because of their technological maturity and control
simplicity.
6.1.1
Principle of Operation and Performance
The operation principle of a DC motor is straightforward. When a wire carrying electric current is placed into a magnetic field, a magnetic force acting
on the wire is produced. The force is perpendicular to the wire and the magnetic field as shown in Figure 6.3. The magnetic force is proportional to the
wire length, magnitude of the electric current, and the density of the magnetic
field; that is,
F = BIL.
(6.1)
When the wire is shaped into a coil, as shown in Figure 6.3, the magnetic
forces acting on both sides produce a torque, which is expressed as
T = BIL cos α,
(6.2)
where α is the angle between the coil plane and the magnetic field as shown in
Figure 6.3. The magnetic field may be produced by a set of windings or PMs.
The former is called wound-field DC motor and the latter is called the PM DC
motor. The coil carrying electric current is called the armature. In practice,
the armature consists of a number of coils. In order to obtain continuous and
155
Electric Propulsion Systems
S
F
L
a
I
D
B
N
F
+
Slip Brush
–
rings
Coil
FIGURE 6.3 Operation principle of a DC motor.
maximum torque, slip rings and brushes are used to conduct each coil at the
position of α = 0.
Practically, the performance of DC motors can be described by the armature
voltage, back electromotive force (EMF), and field flux.
Typically, there are four types of wound-field DC motors, depending on
the mutual interconnection between the field and armature windings. They
are separately excited, shunt excited, series excited, and compound excited
as shown in Figure 6.4. In the case of a separately excited motor, the field
and armature voltage can be controlled independently of one another. In a
shunt motor, the field and armature are connected in parallel to a common
source. Therefore, an independent control of field or armature currents can
only be achieved by inserting a resistance in the appropriate circuit. This is an
inefficient method of control. The efficient method is to use power electronicsbased DC–DC converters in the appropriate circuit to replace the resistance.
The DC–DC converters can be actively controlled to produce proper armature
and field voltage. In the case of a series motor, the field current is the same as
the armature current; therefore, field flux is a function of armature current.
In a cumulative compound motor, the magnetomotive force (mmf) of a series
field is a function of the armature current and is in the same direction as the
mmf of the shunt field.6
The steady-state equivalent circuit of the armature of a DC motor is shown
in Figure 6.5. The resistor Ra is the resistance of the armature circuit. For
separately excited and shunt DC motors, it is equal to the resistance of
the armature windings; for the series and compound motors, it is the sum
of armature and series field winding resistances. The basic equations of
156
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Ia
A1
F1
A1
If
F1
Ia
+
+
Va
–
+
Va
–
Vf
–
–
S2
F2
F1
A1
Ia S
1
Ia
+
+
Va
–
–
A2
Shunt
A2
F2
Separately
S1
If
+
+
S2
+
Va
–
–
+
–
A2
Series
Cumulative compound
F2
FIGURE 6.4 Wound-field DC motors.
a DC motor are
Va = E + Ra Ia ,
E = Ke φωm
T = Ke φIa ,
(6.3)
(6.4)
where φ is the flux per pole in webers, Ia is the armature current in A, Va is
the armature voltage in volts, Ra is the resistance of the armature circuit in
ohms, ωm is the speed of the armature in rad/s, T is the torque developed by
the motor in N m, and Ke is a constant.
Ia
Ra
+
+
Va
–
E = Ke f wm
–
FIGURE 6.5 Steady-state equivalent circuit of the armature circuit of a DC motor.
157
Electric Propulsion Systems
Torque, p.u.
Series
Separately
excited or
Compound shunt
1.0
0.5
0
0.5
1.0
Speed, p.u.
FIGURE 6.6 Speed characteristics of DC motors.
From Equations 6.3 and 6.4, one can obtain
T=
Ke φ
(Ke φ)2
V−
ωm .
Ra
Ra
(6.5)
Equations 6.3 through 6.5 are applicable to all the DC motors, namely separately (or shunt) excited, series, and compound motors. In the case of
separately excited motors, if the field voltage is maintained as constant, one
can assume the flux to be practically constant as the torque changes. In this
case, the speed–torque characteristic of a separately excited motor is a straight
line, as shown in Figure 6.6. The nonload speed ωm is determined by the armature voltage and the field excitation. Speed decreases as torque increases,
and speed regulation depends on the armature circuit resistance. Separately
excited motors are used in applications requiring good speed regulation and
proper adjustable speed.
In the case of series motors, the flux is a function of armature current. In an
unsaturated region of the magnetization characteristic, φ can be assumed to
be proportional to Ia . Thus
φ = Kf Ia .
(6.6)
By Equations 6.4 through 6.6, the torque for series excited DC motors can
obtained as
T=
Ke Kf Va2
(Ra + Ke Kf ωm )2
,
(6.7)
158
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
where the armature circuit resistance Ra is now the sum of armature and field
winding resistance.
A speed–torque characteristic of a series DC motor is shown in Figure 6.6.
In the case of a series, any increase in torque is accompanied by an increase in
the armature current and, therefore, an increase in magnetic flux. Because flux
increases with the torque, the speed drops to maintain a balance between the
induced voltage and the supply voltage. The characteristic, therefore, shows
a dramatic drop. A motor of standard design works at the knee point of the
magnetization curve at the rated torque. At heavy torque (large current) overload, the magnetic circuit saturates and the speed–torque curve approaches
a straight line.
Series DC motors are suitable for applications requiring high starting torque
and heavy torque overload, such as traction. This was just the case for electric traction before the power electronics and micro-control era. However,
series DC motors for traction application bear some disadvantages. They
are not allowed to operate without the load torque with full supply voltage.
Otherwise, their speed will quickly increase up to a very high value (refer to
Equation 6.7) Another disadvantage is the difficulty of regenerative braking.
Performance equations for cumulative compound DC motors can be
derived from Equations 6.3 and 6.4. The speed–torque characteristics are
between series and separately excited (shunt) motors, as shown in Figure 6.6.
6.1.2
Combined Armature Voltage and Field Control
The independence of armature voltage and field provides more flexible control of the speed and torque than other types of DC motors. In EV and HEV
applications, the most desirable speed–torque characteristic is to have a constant torque below a certain speed (base speed), and a constant power in the
speed range of above the base speed, as shown in Figure 6.7. In the speed
range of lower than the base speed, the armature current and field are set at
their rated values, producing the rated torque. From Equations 6.3 through
6.4, it is clear that the armature voltage must be increased proportionally with
the increase of the speed. At the base speed, the armature voltage reaches its
rated value (equal to the source voltage) and cannot be increased further.
In order to further increase the speed, the field must be weakened with the
increase of the speed, then maintaining the back EMF E and armature current
constant. The torque produced drops parabolically with the increase of the
speed and the output power remains constant, as shown in Figure 6.7.
6.1.3
Chopper Control of DC Motors
Choppers are used for the control of DC motors because of a number of
advantages such as high efficiency, flexibility in control, light weight, small
size, quick response, and regeneration down to very low speeds. At present,
159
Electric Propulsion Systems
Torque
Power
Armature current
0
Maximum speed wm
Base speed
Armature
voltage control
Field control
FIGURE 6.7 Torque and power limitations in combined armature voltage and field control.
the separately excited DC motors are usually used in traction, due to the
control flexibility of armature voltage and field.
For a DC motor control in open-loop and closed-loop configurations, the
chopper offers a number of advantages due to the high operation frequency.
High operation frequency results in high-frequency output voltage ripple
and therefore less ripples in the motor armature current and a smaller region
of discontinuous conduction in the speed–torque plane. A reduction in the
armature current ripple reduces the armature losses. A reduction or elimination of the discontinuous conduction region improves speed regulation and
transient response of the drive.
The power electronic circuit and the steady-state waveform of a DC chopper
drive are shown in Figure 6.8. A DC voltage source, V, supplies an inductive
load through a self-commutated semiconductor switch S. The symbol of a selfcommutated semiconductor switch has been used because a chopper can be
built using any devices among thyristors with a forced commutation circuit:
GTO, power transistor, MOSFET, and IGBT. The diode shows the direction
in which the device can carry current. A diode DF is connected in parallel
with the load. The semiconductor switch S is operated periodically over a
period T and remains closed for a time ton = δT with 0 < δ < 1. The variable
δ = ton /T is called the duty ratio or duty cycle of a chopper. Figure 6.8 also
shows the waveform of control signal ic . Control signal ic will be the base
current for a transistor chopper, and a gate current for the GTO of a GTO
chopper or the main thyristor of a thyristor chopper. If a power MOSFET is
used, it will be a gate to the source voltage. When the control signal is present,
the semiconductor switch S will conduct, if forward biased. It is assumed that
the circuit operation has been arranged such that the removal of ic will turn
off the switch.
160
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Self commutated
semiconductor switch
is
Va
(b) V
ia
(a)
Va
+
V
–
ic
DF
0
+
Va
–
Load
dT
T
t
dT
T
t
dT
T
t
dT
T
t
(c) ic
0
(d) ia
ia2
ia1
0
(e) is
ia2
ia1
0
FIGURE 6.8 Principle of operation of a step down (or class A) chopper: (a) basic chopper circuit;
(b)–(e) waveforms.
During the on interval of the switch (0 ≤ t ≤ δT), the load is subjected to a
voltage V and the load current increases from ia1 to ia2 . The switch is opened
at t = δT. During the off period of the switch (δT ≤ t ≤ 1), the load inductance
maintains the flow of current through diode DF . The load terminal voltage
stays zero (if the voltage drop on the diode is ignored in comparison to V) and
the current decreases from ia2 to ia1 . The internal 0 ≤ t ≤ δT is called the duty
interval and the interval δT ≤ t ≤ T is known as the freewheeling interval.
Diode DF provides a path for the load current to flow when switch S is off,
and thus improves the load current waveform. Furthermore, by maintaining
the continuity of the load current at turn-off, it prevents transient voltage from
appearing across switch S, due to the sudden change of the load current. The
source current waveform is also shown in Figure 6.8e. The source current
flows only during the duty interval and is equal to the load current.
The direct component or average value of the load voltage Va is given by
Va =
1
T
0
T
va dt =
1
T
0
δT
V dt = δV.
(6.8)
By controlling δ between 0 and 1, the load voltage can be varied from 0 to
V; thus a chopper allows a variable DC voltage to be obtained from a fixed
voltage DC source.
Electric Propulsion Systems
161
The switch S can be controlled in various ways for varying the duty ratio δ.
The control technologies can be divided into two categories:
1. Time Ratio Control (TRC).
2. Current Limit Control (CLC).
In TRC, also known as pulse width control, the ratio of on time to chopper
period is controlled. The TRC can be further divided as follows:
1. Constant frequency TRC: The chopper period T is kept fixed and the on
period of the switch is varied to control the duty ratio δ.
2. Varied frequency TRC: Here δ is varied either by keeping ton constant
and varying T or by varying both ton and T.
In variable frequency control with constant on time, low-output voltage
is obtained at very low chopper frequencies. The operation of a chopper
at low frequencies adversely affects the motor performance. Furthermore,
the operation of a chopper with variable frequencies makes the design
of an input filter very difficult. Thus, variable frequency control is rarely
used.
In CLC, also known as point-by-point control, δ is controlled indirectly
by controlling the load current between certain specified maximum and
minimum values. When the load current reaches a specified maximum
value, the switch disconnects the load from the source and reconnects it
when the current reaches a specified minimum value. For a DC motor
load, this type of control is, in effect, a variable frequency variable on time
control.
The following important points can be noted from the waveform of
Figure 6.8.
1. The source current is not continuous but flows in pulses. The pulsed
current makes the peak input power demand high and may cause
fluctuation in the source voltage. The source current waveform can be
resolved into DC and AC harmonics. The fundamental AC harmonic
frequency is the same as the chopper frequency. The AC harmonics are
undesirable because they interfere with other loads connected to the
DC source and cause radio frequency interference through conduction and electromagnetic radiation. Therefore, an L-C filter is usually
incorporated between the chopper and the DC source. At higher chopper frequencies, harmonics can be reduced to a tolerable level by a
cheaper filter. From this point, a chopper should be operated at the
highest possible frequency.
2. The load terminal voltage is not a perfect direct voltage. In addition to
a direct component, it has the harmonics of the chopping frequency
and its multiples. The load current also has an AC ripple.
162
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
The chopper of Figure 6.8 is called a class A chopper. It is one of the number
of chopper circuits that are used for the control of DC motors. This chopper is
capable of providing only a positive voltage and a positive current. It is called a
single-quadrant chopper, only providing separately excited DC motor control
in the first quadrant, that is, positive speed and positive torque. Since it can
vary the output voltage from V to 0, it is also a step-down chopper or a DCto-DC buck converter. The basic principle involved can also be used to realize
a step-up chopper or DC-to-DC boost converter.
The circuit diagram and steady-state waveforms of a step-up chopper are
shown in Figure 6.9. This chopper is known as a class B chopper. The presence
of control signal ic indicates the duration for which the switch can conduct if
forward-biased. During a chopping period T, it remains closed for an interval
0 ≤ t ≤ δT and remains open for an interval δT ≤ t ≤ T. During the on period,
iS increases from iS1 to iS2 , thus increasing the magnitude of energy stored in
inductance L. When the switch is opened, current flows through the parallel
combination of the load and capacitor C. Since the current is forced against
the higher voltage, the rate of change of the current is negative. It decreases
from iS2 to iS1 in the switch’s off period. The energy stored in the inductance
L and the energy supplied by the low-voltage source are given to the load.
The capacitor C serves two purposes. At the instant of opening of switch S,
the source current, iS , and load current, ia , are not the same. In the absence of
C, the turn-off of S will force the two currents to have the same values. This
will cause high induced voltage in the inductance L and the load inductance.
Another reason for using capacitor C is to reduce the load voltage ripple. The
purpose of the diode D is to prevent any flow of current from the load into
switch S or source V.
L
+
V
–
ic
S
(b) ic
D
a
(a)
is
C
+
Va
–
0
Load
dT
T
t
dT
T
t
dT
T
t
(c)
Va
b
0
(d) iS
iS2
iS1
0
FIGURE 6.9 Principle of operation of a step-up (or class B) chopper: (a) basic chopper circuit;
(b)–(d) waveforms.
163
Electric Propulsion Systems
For understanding of the step-up operation, capacitor C is assumed large
enough to maintain a constant voltage Va across the load. The average voltage
across the terminals a and b is given as
Vab =
1
T
T
0
vab dt = Va (1 − δ).
(6.9)
The average voltage across the inductance L is
VL =
1
T
T
L
0
di
dt
dt =
1
T
iS2
iS1
L di = 0.
(6.10)
The source voltage
V = VL + Vab .
(6.11)
Substituting from Equations 6.9 and 6.10 into Equation 6.11 gives
V = Va (1 − δ)
or
Va =
V
.
1−δ
(6.12)
According to Equations 6.12, theoretically the output voltage Va can be
changed from V to ∞ by controlling δ from 0 to 1. In practice, Va can be
controlled from V to a higher voltage, which depends on the capacitor C, and
the parameters of the load and chopper.
The main advantage of a step-up chopper is the low ripple in the source
current. While most applications require a step-down chopper, the step-up
chopper finds application in low-power battery-driven vehicles. The principle
of the step-up chopper is also used in the regenerative braking of DC motor
drives.
6.1.4
Multi-Quadrant Control of Chopper-Fed DC Motor Drives
The application of DC motors on EVs and HEVs requires the motors to operate
in multiquadrants, including forward motoring, forward braking, backward
motoring, and backward braking, as shown in Figure 6.10. For vehicles with
reverse mechanical gears, two-quadrant operation (forward motoring and
forward braking, or quadrant I and quadrant IV) is required. However, for
the vehicles without reverse mechanical gears, four-quadrant operation is
needed. Multiquadrant operation of a separately excited DC motor is implemented by controlling the voltage poles and magnitude through power
electronics-based choppers.
164
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
T
I
II
wm
0
III
IV
FIGURE 6.10 Speed–torque profiles of a multiquadrant operation.
6.1.4.1 Two-Quadrant Control of Forward Motoring
and Regenerative Braking
Atwo-quadrant operation consisting of forward motoring and forward regenerative braking requires a chopper capable of giving a positive voltage and
current in either direction. This two-quadrant operation can be realized in the
following two schemes.6
6.1.4.1.1 Single Chopper with a Reverse Switch
The chopper circuit used for forward motoring and forward regenerative
braking is shown in Figure 6.11 where S is a self-commutated semiconductor
switch, operated periodically such that it remains closed for a duration of δT
and remains open for the duration of (1 − δ)T. C is the manual switch. When
C is closed and S is in operation, the circuit is similar to that of Figure 6.6,
permitting the forward motoring operation. Under this condition, terminal a
is positive and terminal b is negative.
The regenerative braking in the forward direction is obtained when C is
opened and the armature connection is reversed with the help of the reversing
switch RS, making terminal b positive and terminal a negative. During the on
C
+
V a
–
D1
D2
Ia R
S
b
+
Va
–
ic
S
FIGURE 6.11 Forward motoring and regenerative braking control with a single chopper.
Electric Propulsion Systems
165
period of the switch S, the motor current flows through a path consisting of the
motor armature, switch S, and diode D1 , and increases the energy stored in
the armature circuit inductance. When S is opened, the current flows through
the armature diode D2 , source V, diode D1 and back to the armature, thus
feeding energy into the source.
During motoring, the change over to regeneration is done in the following
steps. Switch S is deactivated and switch C is opened. This forces the armature
current to flow through diode D2 , source V and diode D1 . The energy stored
in the armature circuit is fed back to the source and the armature current falls
to zero. After an adequate delay to ensure that the current has indeed become
zero, the armature connection is reversed and switch S is reactivated with a
suitable value of δ to start regeneration.
6.1.4.1.2 Class C Two-Quadrant Chopper
In some applications, a smooth transition from motoring to braking and vice
versa is required. For such applications, the class C chopper is used as shown
in Figure 6.12. The self-commutated semiconductor switch S1 and diode D1
constitute one chopper and the self-commutator switch S2 , and diode D2
form another chopper. Both the choppers are controlled simultaneously, both
for motoring and regenerative braking. The switches S1 and S2 are closed
alternately. In the chopping period T, S1 is kept on for a duration δT and
S2 is kept on from δT to T. To avoid a direct short-circuit across the source,
care is taken to ensure that S1 and S2 do not conduct at the same time. This
is generally achieved by providing some delay between the turn-off of one
switch and the turn-on of another switch.
The waveforms of the control signals, va , ia , and is and the devices under
conducting during different intervals of a chopping period are shown in
Figure 6.11b. In drawing these waveforms, the delay between the turn-off
of one switch and turn-on of another switch has been ignored because it is
usually very small. The control signals for the switches S1 and S2 are denoted
by ic1 and ic2 , respectively. It is assumed that a switch conducts only when
the control signal is present and the switch is forward biased.
The following points are helpful in understanding the operation of this
two-quadrant circuit.
1. In this circuit, discontinuous conduction does not occur, irrespective of
its frequency of operation. The discontinuous conduction occurs when
the armature current falls to zero and remains zero for a finite interval
of time. The current may become zero either during the freewheeling
interval or in the energy transfer interval. In this circuit, freewheeling will occur when S1 is off and the current is flowing through D1 .
This will happen in interval δ T ≤ t ≤ T, which is also the interval for
which S2 receives the control signal. If ia falls to zero in the freewheeling interval, the back EMF will immediately drive a current through
S2 in the reverse direction, thus preventing the armature current from
166
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
is
(a)
S1
ic1
D2
+
ia
V
S2
ic2
–
+
Va
D1
–
(b)
ic1
0
ic2
dT
T
T +dT
2T
t
0
Va
dT
T
T +dT
2T
t
dT
T
T +dT
2T
t
dT
T
T +dT
2T
t
V
0
ia
0
is
0
dT
D2
S1
T +dT
T
D1
S2
D2
S1
D1
t
2T
S2
D2
FIGURE 6.12 Forward motoring and regenerative braking control using class C two-quadrant
chopper: (a) chopper circuit; (b) waveforms.
remaining zero for a finite interval of time. Similarly, the energy transfer will be present when S2 is off and D2 is conducting—that is, during
the interval 0 ≤ t ≤ δT. If the current falls to zero during this interval,
S1 will conduct immediately because ic is present and V > E. The
armature current will flow, preventing discontinuous conduction.
2. Since discontinuous conditions are absent, the motor current will be
flowing all the time. Thus, during the interval 0 ≤ t ≤ δT, the motor
167
Electric Propulsion Systems
armature will be connected either through S1 or D2 . Consequently, the
motor terminal voltage will be Vand the rated of change of ia will be
positive because V > E. Similarly, during the interval δT ≤ t ≤ T, the
motor armature will be shorted either through D1 or S2 . Consequently,
the motor voltage will be zero and the rate of change of ia will be
negative.
3. During the interval 0 ≤ t ≤ δT, the positive armature current is carried by S1 and the negative armature current is carried by D2 . The
source current flows only during this interval and it is equal to ia .
During the interval δT ≤ t ≤ T, the positive current is carried by D1
and the negative current is carried by S2 .
4. From the motor terminal voltage waveform of Figure 6.12b, Va = δV.
Hence,
δV − E
.
(6.13)
Ia =
Ra
Equation 6.13 suggests that the motoring operation takes place when
δ > (E/V), and that regenerative braking occurs when δ < (E/V). The
no-load operation is obtained when δ = (E/V).
6.1.4.2
Four-Quadrant Operation
The four-quadrant operation can be obtained by combining two class C choppers (Figure 6.12a) as shown in Figure 6.13, which is referred to as a class E
chopper. In this chopper, if S2 is kept closed continuously, and S1 and S4 are
controlled, a two-quadrant chopper is obtained, which provides positive terminal voltage (positive speed) and the armature current in either direction
(positive or negative torque), giving a motor control in quadrants I and IV.
Now if S3 is kept closed continuously and S1 and S4 are controlled, one gets
a two-quadrant chopper that can supply a variable negative terminal voltage
(negative speed), and the armature current can be in either direction (positive
or negative torque), giving a motor control in quadrants II and III.
D3
ic 1
+
ic3
S1
S3
D1
S2
D4
Ia
V
–
D2
ic 4
+
S4
FIGURE 6.13 Class E four-quadrant chopper.
Va
–
ic2
168
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
This control method has the following features: the utilization factor of
the switches is low due to the asymmetry in the circuit operation. Switches
S3 and S2 should remain on for a long period. This may create commutation problems when the switches are using thyristors. The minimum output
voltage depends directly on the minimum time for which the switch can be
closed, since there is always a restriction on the minimum time for which
the switch can be closed, particularly in thyristor choppers.7 The minimum
available output voltage, and therefore the minimum available motor speed,
is restricted.
To ensure that switches S1 and S4 , or S2 and S3 are not on at the same time,
some fixed time interval must elapse between the turn-off for one switch
and the turn-on of another switch. This restricts the maximum permissible
frequency of operation. It also requires two switching operations during a
cycle of the output voltage.
Dubey6 provided other control methods to solve the problems mentioned
above.
6.2
Induction Motor Drives
Commutatorless motor drives offer a number of advantages over conventional DC commutator motor drives for the electric propulsions of EVs and
HEVs. At present, induction motor drives are the mature technology among
commutatorless motor drives. Compared with DC motor drives, the AC
induction motor drive has additional advantages such as lightweight nature,
small volume, low cost, and high efficiency. These advantages are particularly
important for EV and HEV applications.
There are two types of induction motors, namely, wound-rotor and squirrelcage motors. Because of the high cost, need for maintenance, and lack
of sturdiness, wound-rotor induction motors are less attractive than their
squirrel-cage counterparts, especially for electric propulsion in EVs and
HEVs. Hence, squirrel-cage induction motors are loosely named as induction
motors.
A cross section of a two-pole induction motor is shown in Figure 6.14.
Slots in the inner periphery of the stator are inserted with three phase windings, a–a , b–b , and c–c . The turns of each winding are distributed such that
the current in the winding produces an approximate sinusoidally distributed
flux density around the periphery of the air gap. The three windings are
spatially arranged by 120◦ as shown in Figure 6.14.
The most common types of induction motor rotors are the squirrel-cage
motors in which aluminum bars are cast into slots in the outer periphery of the rotor. The aluminum bars are short-circuited together at both
ends of the rotor by cast aluminum end rings, which also can be shaped
as fans.
169
Electric Propulsion Systems
a′
Stator
b
c
Stator slot and
windings
120º
Rotor bar
120º
Rotor
120º
b′
c′
a
FIGURE 6.14 Cross section of an induction motor.
6.2.1
Basic Operation Principles of Induction Motors
Figure 6.15 shows, schematically, a cross section of the stator of a threephase, two-pole induction motor. Each phase is fed with a sinusoidal AC
current, which has a frequency of ω and a 120◦ phase difference between each
other. Current ias , ibs , and ics in the three stator coils a–a , b–b , and c–c produce alternative magnetic motive forces (mmfs), F as , F bs , and F cs , which are
space vectors. The resultant stator mmf vector F ss constitutes a vector sum of
the phase mmf vectors.
q
(a)
(b) i
a′
c
ias
ibs
ic
s
b
d
0
wt
c′
b′
a
FIGURE 6.15 Induction motor stator and stator winding current: (a) spatially symmetric threephase stator windings; (b) phase currents.
170
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
The mmfs produced by the phase currents can be written as
F as = F as sin ωt,
F bs = F bs sin ωt − 120◦ ,
F bs = F cs sin ωt − 240◦ .
(6.14)
(6.15)
(6.16)
The resultant stator mmf vector, F ss , is expressed as
◦
◦
◦
F ss = F sas ei0 + F sbs ej120 + F scs ej240 .
(6.17)
Assuming that the magnitudes of the three-phase mmfs are identical, equal
to F s , Equation 6.17 can be further expressed as
F ss =
3
◦
F s e(ωt−90 ) .
2
(6.18)
Equation 6.18 indicates that the resultant stator mmf vector is rotating with the
frequency of the angle velocity of ω, and its magnitude is 3/2Fs . Figure 6.16
graphically shows the stator mmf vectors at ωt = 0 and ωt = 90◦ ; here, ωt is
the angle in Equations 6.12 through 6.18, rather than the resultant stator mmf
vector relative to the d-axis. Actually, if the ωt Equations 6.14 through 6.16 is
taken as the reference, the resultant stator mmf vector is a 90◦ delay to the
phase a–a mmf.
The reaction between the rotating stator mmf and the rotor conductors
induces a voltage in the rotor and hence electric current in the rotor. In
turn, the rotating mmf produces a torque on the rotor, which is carrying the
induced current. It is clear that the induced current in the rotor is essential for
producing torque, and in turn the induced current depends on the relative
movements between the stator mmf and the rotor. That is why there must
exist a difference between the angular velocity of the rotating stator mmf and
the angular velocity of the rotor.
The frequency ω, or angular velocity of the rotating stator mmf in equation,
depends only on the frequency of the alternative current of the stator; thus it
is referred to as electrical angular velocity. For the machine with two poles,
the electrical angular velocity is identical to the mechanical angular velocity
of the rotating stator mmf. However, for the machine with more than two
poles, the mechanical angular velocity differs from the electrical one, which
can be expressed as
ωms =
2
4πf
ω=
rad/s,
p
p
(6.19)
where f is the frequency of the alternative current or angular velocity of
the rotating stator mmf in cycle/s. When the angular velocity of the rotor
is equal to the mechanical angular velocity of rotating stator mmf, there will
171
Electric Propulsion Systems
(a)
Fbs
q
Fas d
Fcs
q
(b)
q
(c)
a'
a′
c
c
b
b
1
Fcs = – 2
d
Fas = 0
3
b′
Fcs = 2 Fs
Fbs = –
s
Fs =
3
2
3
2
Fs
s
Fs
c¢
Fs
d
Fas = Fs
b′
Fs =
1
Fbs = – 2 Fs
3
2
Fs
c′
a
a
FIGURE 6.16 Stator mmf vectors: (a) positive direction of each phase mmf; (b) stator mmf
vectors at ωt = 0; (c) stator mmf vectors at ωt = 90◦ .
be no induced current in the rotor, and then no torque is produced. Thus,
the mechanical angular velocity of the rotating stator mmf is also called
synchronous speed.
If the rotor speed is ωm in rad/s, then the relative speed between the stator
rotating field and the rotor is given by
ωsl = ωms − ωm = sωms
(6.20)
where ωsl is called slip speed. The parameter s, known as slip, is given by
s=
ωsl
ωms − ωm
=
.
ωms
ωms
(6.21)
Because of the relative speed between the stator field and the rotor, balanced
three-phase voltages are induced in the rotor mentioned before. The frequency
of these voltages is proportional to the slip speed. Hence
ωr =
ωsl
ω = sω,
ωms
(6.22)
where ωr is the frequency of the rotor voltage induced.
For ωm < ωms , the relative speed is positive; consequently, the rotorinduced voltages have the same phase sequence as the stator voltages. The
three-phase current flowing through the rotor produces a magnetic field,
172
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
which moves with respect to the rotor at the slip speed in the same direction as
the rotor speed. Consequently, the rotor field moves in the space at the same
speed as the stator, and a steady torque is produced. For ωm = ωms , the relative speed between the rotor and stator field becomes zero. Consequently, no
voltages are induced and no torque is produced by the motor. For ωm > ωms ,
the relative speed between the stator field and the rotor speed reverses. Consequently, the rotor-induced voltages and currents also reverse and have a
phase sequence opposite to that of the stator. Moreover, the developed torque
has a negative sign, suggesting generator operation. (The generator is used
to produce regenerative braking.)
6.2.2
Steady-State Performance
A per-phase equivalent circuit of an induction motor is shown in Figure 6.17a.
The fields produced by the stator and rotor are linked together by an ideal
transformer. aT1 is the transformer factor, which is equal to ns /nr , where ns
and nr are the numbers of turns of stator and rotor windings, respectively.
For a squirrel-cage rotor, nr = 1. The equivalent circuit can be simplified by
referring the rotor quantities to the stator frequency and number of turns.
The resultant equivalent circuit is shown in Figure 6.17b, where Rr and Xr are
the rotor resistance and reactance referred to the stator, and is given by the
following equations:
Rr = a2T1 Rr
and Xr = a2T1 Xr .
(6.23)
The stator reactance, mutual reactance, and rotor reactance referred to the
stator can be expressed by the stator frequency and their inductances, Ls , Lm ,
and Lr , as shown in Figure 6.17. The impedances of stator, field, and rotor can
be expressed as
Zs = Rs + jLs ω
Zm = jLm ω
Zr =
Rr
+ jLr ω
s
(6.24)
(6.25)
(6.26)
The driving-point impedance of the circuit is
Z = Zs +
Z m Zr
.
Zm + Z r
(6.27)
Hence, the current Is and Ir can be calculated as
Is =
V
.
Z
(6.28)
173
Electric Propulsion Systems
Xs
Rs
(a)
Im
Is
V
Xs
n
Is
V
sE
aT1
Ideal
transformer
A
Lsw
Rs
nr
E
Stator
(b)
Rotor
Lr w
R'r /s
I'r
Im
Lmw
Rr
sXr
I'r
E
B
FIGURE 6.17 Per phase equivalent circuit of an induction motor: (a) with a transformer between
rotor and stator and (b) refer rotor quantities to stator.
and
Ir =
Zm
Is .
Zm + Z r
(6.29)
The total electrical power supplied to the motor for three phases is
Pelec = 3Ir2
Rr
.
s
(6.30)
The mechanical power of the rotor can be obtained by subtracting the total
power loss in the stator as
Pmech = Pelec − 3Ir2 Rr .
(6.31)
The angular velocity of the rotor, ωm , is
ωm =
2
ω (1 − s) .
P
(6.32)
The torque developed by the motor can be determined by
T=
Pmech
.
ωm
(6.33)
Figure 6.18 shows the torque–slip characteristics of an induction motor, which
has fixed voltage and frequency. In the region of 0 < s < sm , where sm is
174
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Torque
Forward braking Forward motoring
Reverse plugging
Tm,max
Ts
2wms
–1.0
wms
–0.5 –Sm
0S
m
0.5
1.0
1.5
–wms Speed, wm
2.0
Slip, s
Tr,max
FIGURE 6.18 Torque–slip characteristics of an induction motor with fixed stator frequency and
voltage.
the rated slip of the motor, the torque increases approximately linearly with
the increase of slip until reaching its maximum at s = sm , then it decreases
with the further increase of the slip. At s = 1, the rotor speed is zero and the
corresponding torque is the starting torque, which is less than its torque at
s = sm . The region of 0 < s < 1 is the forward motoring region. In the region
of s > 1, the rotor torque is positive and decreases further with the increase of
slip, and the rotor speed is negative, according to Equation 6.21. Thus in this
region, the operation of the motor is reverse braking. In the region of s < 0,
that is, when the rotor speed is greater than the synchronous speed, the motor
produces a negative torque.
It is clear that the speed–torque characteristic of a fixed voltage and fixed
frequency induction motor is not appropriate to vehicle traction applications.
This is due to the low starting torque, limited speed range, and unstable
operation in the range of s > sm , in which any additional disturbing torque in
the load will lead the machine to stop as the torque decreases with the speed
decreasing characteristically. The high slip also results in high current, which
may cause damage in the stator windings. Actually, the operation of the fixed
voltage and frequency induction motor are usually operated in the narrow slip
range of 0 < s < sm . Thus, for traction application, an induction motor must
be controlled to provide proper speed–torque characteristic as mentioned in
Chapters 2 and 4.
6.2.3
Constant Volt/Hertz Control
For traction application, the torque–speed characteristic of an induction motor
can be varied by simultaneously controlling the voltage and frequency, which
175
Electric Propulsion Systems
is known as the constant volt/hertz control. By emulating a DC motor at low
speed, the flux may be kept constant. According to Figure 6.17b, the field
current Im should be kept constant and equal to its rated value. That is,
Imr =
E
Erated
=
,
Xm
ωr L m
(6.34)
where Imr is the rated field current, Erated and ωr are the rated mmf and
frequency of the stator, respectively. To maintain the flux at constant, the
E/ω should be kept constant and equal to Erated /ωr . Ignoring the voltage
drop in the stator impedance Zs results in a constant V/ω until the frequency
and voltage reach their rated values. This approach is known as constant
volt/hertz control.6
From Figure 6.17b, the rotor current can be calculated as
Ir =
(ω/ωr )Erated
.
jLr ω + Rr /s
(6.35)
The torque produced can be obtained as
3
3
T = Ir2 Rr s =
ω
ω
(ω/ω)2 E2rated Rr /s
.
2
Rr s + (Lr ω)2
(6.36)
The slip sm corresponding to the maximum torque is
Rr
.
Lr ω
(6.37)
3 E2rated
.
2 Lr ω2r
(6.38)
sm = ±
And then, the maximum torque is
Tmax =
Equation 6.38 indicates that with the constant E/ω, the maximum torque is
constant with varying frequency. Equation 6.37 indicates that sm ω is constant,
resulting in constant slip speed, ωsl . In practice, due to the presence of stator
impedance and the voltage drop, the voltage should be somewhat higher than
that determined by constant E/ω, as shown in Figure 6.19.
When motor speed is beyond its rated speed, the voltage reaches its rated
value and cannot be increased with the frequency. In this case, the voltage
is fixed to its rated value and the frequency increases continuously with the
motor speed. The motor goes into the field weakening operation. The slip
s is fixed to its rated value corresponding to the rated frequency, and the
slip speed ωsl increases linearly with the motor speed. This control approach
results in constant power operation as shown in Figure 6.19.
176
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
V
T
Is
V
T
T
Pm
Pm
V
wsl
Is
Pm
wsl
T
wsl
0 (s = 1)
Constant
torque
region
wr – wsl (s = sr)
Constant
power
region
Region with
fixed voltage
and frequency
wm
FIGURE 6.19 Operating variables varying with motor speed.
In traction applications, speed control in a wide range is usually required
and the torque demand in the high-speed range is low. Control beyond constant power range is required. To prevent the torque from exceeding the
breakdown torque, the machine is operated at a constant slip speed and the
machine current and power are allowed to decrease as shown in Figure 6.19.
Figure 6.20 shows a general block diagram where constant V/f control is
implemented.
6.2.4
Power Electronic Control
As EV and HEV propulsion, an induction motor drive is usually fed with a
DC source (battery, fuel cell, etc.), which has approximately constant terminal
voltage. Thus a variable frequency and variable voltage DC/AC inverter is
needed to feed the induction motor. The general DC/AC inverter is constituted by power electronic switches and power diodes. The commonly used
topology of a DC/AC inverter is shown in Figure 6.21a, which has three legs
(S1 and S4 , S3 and S6 , and S5 and S2 ), feeding the phase a, phase b, and phase c
of the induction motor. When switches S1 , S3 , and S5 are closed, S4 , S6 , and S2
are opened, and phases a, b, and c are supplied with a positive voltage (Vd /2).
Similarly, when S1 , S3 , and S5 are opened and S4 , S6 , and S2 are closed, the
phases a, b, and c is supplied with a negative voltage. All the diodes provide
a path for the reverse current of each phase.
177
Electric Propulsion Systems
DC Supply
voltage
V *s
Slip speed
controller
Inverter
w*sl +
wm
+
+
w*syn
w*
P
2
wm
Motor
wm
Speed sensor
FIGURE 6.20 General configuration of constant V/f control.
For constant volt/hertz control of an induction motor, sinusoidal pulse
width modulation (PWM) is used exclusively. Three-phase reference voltage
Va , Vb and Vc of variable amplitude Aa , Ab , and Ac are compared with a
common isosceles triangular carrier wave Vtr of a fixed amplitude Am as
shown in Figure 6.21c. The outputs of comparators 1, 2, and 3 form the control
signals for the three legs of the inverter. When the sinusoidal reference voltage
Va , Vb , and Vc , at a time t is greater than the triangular waved voltage, turn-on
signals are sent to the switches S1 , S3 , and S5 and turn-off signals to S4 , S6 ,
and S6 . Thus the three phases of the induction motor have positive voltage.
On the other hand, when the reference sinusoidal voltage is smaller than the
triangular wave voltage, turn-on signals are sent to the switches S1 , S3 , and S5
and turn-off signals to S4 , S6 , and S2 . The three phases of the induction motor
then have a negative voltage. The voltages of the three phases are shown in
Figure 6.21d through 6.21f.
The frequency of the fundamental component of the motor terminal voltage
is the same as that of the reference sinusoidal voltage. Hence, the frequency
of the motor voltage can be changed by changing the frequency of the reference voltage. The ratio of the amplitude of the reference wave to that of the
triangular carrier wave, m, is called the modulation index; therefore,
m=
A
,
Am
(6.39)
where A is the multitude of the reference sinusoidal voltage, Va , Vb , or Vc ,
and Am is the multitude of angular carrier voltage. The fundamental (rms)
178
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
(a)
Is1
S1
D1
Vd /2
Is5
Is3
S3
D3
S5
D5
Zb
a
Vd /2
S4
Ia
D4
b
Ib
S6
D6
c
S2
1
+
Reference
sine-wave
generator
2
To (S3, S6)
–
Vc
3
+
To (S1, S4)
–
Vb
+
Za = R + j w L
Zc
D2
Va
(b)
n
Ic
To (S5, S2)
–
Comparator
Triangular
wave
generator
(c)
Am
Vtr
Va
Aa
Vb
Vc
Ab
Ac
wt
p
2p
(d)
Vd/2
wt
0
p
2p
–Vd/2
(e)
Vd/2
0
wt
p
2p
–Vd/2
(f )
Vd/2
0
wt
p
2p
–Vd/2
FIGURE 6.21 DC/AC inverter with sinusoidal pulse-width modulation: (a) inverter topology;
(b) control signals; (c) three phase reference voltage and triangular carrier waveforms; (d) voltage
of phase a; (e) voltage of phase b; and (f) voltage of phase c.
179
Electric Propulsion Systems
component in the phase waveform Vao , Vbo , or Vco is given by
mVd
Vf = √ .
2 2
(6.40)
Thus, the fundamental voltage increases linearly with m until m = 1 (i.e.,
when the amplitude of the reference wave becomes equal to that of the carrier
wave). For m > 1, the number of pulses in Vao , Vbo , or Vco , becomes less and
the modulation ceases to be sinusoidal.6
6.2.5
Field Orientation Control
The constant volt/hertz control of the induction motor is more suitably
applied to the motors that operate with relative slow speed regulation. However, this approach shows poor response to frequent and fast speed varying,
and also results in poor operation efficiency due to the poor power factor.
In the last two decades, FOC or vector control technology has been successfully developed. This technology mostly overcomes the disadvantages of the
constant volt/hertz control in AC motor drives.
6.2.5.1
Field Orientation Principles
The aim of FOC is to maintain the stator field perpendicular to the rotor field
so as to always produce the maximum torque as in DC motors. However,
for induction motors, phase voltages are the only accesses for the purpose of
control.
As mentioned in Section 6.2.1, when balanced three-phase sinusoidal currents flow through the three phases of the stator of an induction motor, the
rotating field is developed; current is induced in the rotor. In turn, the current
induced in the rotor is also three-phase and produces a field, which rotates
with the same angular velocity of the stator rotating field. The rotating fields
of both stator and rotor can be described by two retorting vectors, referring
to a common, stationary reference frame, d–q, as shown in Figure 6.16. The
mmf of the stator field is expressed by Equation 6.17. For convenience, it is
repeated as follows:
◦
◦
◦
F ss = F as ei0 + F bs e j120 + F cs e j240 .
(6.41)
Similarly, the stator voltage, stator current, and stator flux can be expressed
as vectors in the same way. That is,
◦
◦
◦
s j0
s j120
s j240
e + vbs
e
+ vcs
e
,
vss = vas
s
e
iss = ias
j0◦
λss = λsas e
s
+ ibs
e
j0◦
j120◦
+ λsbs e
s
+ ics
e
j120◦
j240◦
+ λscs e
(6.42)
,
j240◦
(6.43)
.
(6.44)
180
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
The subscript s refers to the stator and as, bs, and cs refer to phases a, b, and
c of the stator. The superscript s refers to the variable that is referred to the
stator fixed frame. Bold symbols stand for vector variables. The vectors of
stator voltage, current, and flux can be also described by its components in d
and q axes as follows:
s vds
s
vqs
s ids
s
iqs
⎡
⎢1
=⎢
⎣
0
⎡
⎢1
=⎢
⎣
0
⎡
s ⎢1
λds
=⎢
⎣
λsqs
0
1
−
√2
3
2
1
−
√2
3
2
1
−
2
√
3
2
⎤
⎡ ⎤
1
vas
⎥
2 ⎥ ⎣v ⎦
√
bs ,
3⎦ v
cs
−
2
⎤
⎡ ⎤
1
−
ias
⎥
2 ⎥ ⎣i ⎦
√
bs ,
3⎦ i
cs
−
2
⎤
⎡ ⎤
1
−
⎥ λas
2
√ ⎥ ⎣λbs ⎦ .
3⎦ λ
cs
−
2
−
(6.45)
(6.46)
(6.47)
In a real induction motor, the rotor winding differs from the stator winding,
that is, the effective number of turns per phase of the rotor winding, Nr , is not
equal to that of the stator winding, Ns . Therefore the turns ratio, υ = Ns /Nr ,
must be taken into account. The vectors of rotor current, voltage, and magnetic
flux can be described by irr , vrr , and λrr in the rotor frame. However, it is
necessary to transform the vectors from the rotor frame to the stator frame
for the purpose of easy analysis. The transformations of these vectors (refer
to Figure 6.22) are described by
isr =
ejθ0 r
i,
v r
(6.48)
vsr = vejθ0 vrr ,
(6.49)
λsr = vejθ0 λrr .
(6.50)
Using vector notation, either the stator or rotor windings can be represented by a simple resistive-plus-inductive circuit, using current, voltage,
and magnetic flux space vector as illustrated in Figure 6.23.
Using the vector version of Kirchhoff’s voltage law, the equation of the
stator winding can be written as
vss = Rs iss +
dλss
dt
(6.51)
181
Electric Propulsion Systems
q
r
ir
s
q
i qr
r
d
r
i qr
s
i qr
q0
d
s
i qr
FIGURE 6.22 Transformation of rotor current vector from rotor frame to stator frame.
and that of rotor winding as
vrr = Rrr irr +
dλrr
,
dt
(6.52)
where Rs and Rrr are the actual stator and rotor resistances per phase, respectively. As is known from the steady-state theory of induction machines, the
relation between Rrr and the rotor resistance referred to the stator is
Rrr =
1
Rr .
v2
(6.53)
Hence from Equations 6.48 and 6.53, the first term of Equation 6.52 is
Rrr irr =
e−jθ0
Rr isr .
v
(6.54)
R
i
Ri
V
e=
dl
dt
FIGURE 6.23 Resistive-plus-inductive equivalent circuit of either the stator or rotor windings.
182
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
The second term, from Equation 6.50, gives
ejθ0
dλrr
=
dt
v
dλsr
− jω0 .
dt
(6.55)
Finally, substituting Equations 6.54 and (6.55) into Equation 6.52 gives
vsr = Rr isr +
dλsr
− jω0 λsr .
dt
(6.56)
Introducing a differentiation operator p ≡ d/dt, the voltage equation of an
induction motor can be written as
vss = Rs iss + pλss ,
(6.57)
vsr
(6.58)
=
Rr isr
+ (p − jω0 )λsr .
The flux vector λss and λsr can then be expressed in terms of current vector
iss and isr , and the motor inductances as
s L
λs
= s
λsr
Lm
Lm iss
,
Lr isr
(6.59)
where Lm is the mutual inductance; Ls is the stator inductance calculated
as the sum of the stator leakage inductance Lls and the mutual inductance
Lm ; and Lr is the rotor inductance calculated as the sum of the rotor leakage
inductance Llr and the mutual inductance Lm .
Finally, the voltage equation can be written in matrix format as
⎡ s ⎤ ⎡
vds
Rs
⎢v s ⎥ ⎢
⎢ qs ⎥ ⎢ 0
⎢ s ⎥=⎣
0
⎣vdr ⎦
−ω
s
0 Lm
vqr
⎡
Ls
⎢0
+⎢
⎣ Lm
0
0
Rs
ω0 Lm
0
0
Ls
0
Lm
Lm
0
Lr
0
0
0
Rr
−ω0 Lr
⎤ ⎡is ⎤
ds
0
s ⎥
⎥⎢
iqs
0
⎢
⎥⎢ ⎥
⎥
ω0 Lr ⎦ ⎣is ⎦
dr
Rr
is
⎡s ⎤
ids
0
⎢
s ⎥
iqs
Lm ⎥
⎥
⎥ d ⎢
⎥.
⎢
0 ⎦ dt ⎣is ⎦
dr
Lr
is
dr
⎤
(6.60)
dr
s and vs are zero.
Because the rotor circuit of the induction motor is shorted, vdr
qr
At a given rotor speed, ω0 , the stator and rotor currents can be obtained by
solving Equation 6.60. The torque developed by the motor can be expressed as
T=
P
P
s s
s s
Lm iqs
idr − ids
iqr = Lm Im iss is∗
r ,
3
3
(6.61)
183
Electric Propulsion Systems
where Im stands for the imaginary part of the production of vector iss and
conjugate vector of is∗
r .
Transferring three-phase variables (voltage, current, and flux) into a stationary stator-based dq frame does not change the alternate characteristics of
the variable with time. AC quantities are somewhat inconvenient for control
purposes. For instance, control systems are usually represented by block diagrams in which the variables are time-varying DC signals. Therefore, another
transformation is necessary, which allows the conversion of the AC dq components of the motor vectors into DC variables. To do this, a transformation is
conducted from a stationary stator reference frame, dq, to the so-called excitation reference frame DQ, which rotates with the angular speed ω, in the same
direction as does mmf, F ss . As a result, in the steady state, coordinates of motor
vectors in the new reference frame do not vary in time. This is illustrated in
Figure 6.24, which shows the stator mmf vector in both reference frames.
The voltage vector of the stator in excitation reference frame can be
expressed as
veS = vss e−jωt
(6.62)
Considering e−jωt = cos(ωt) − sin(ωt), the components of stator voltage on
DQ frame is
e s vDS
cos(ωt)
sin(ωt) vds
=
.
(6.63)
e
s
vQS
− sin(ωt) cos(ωt) vqs
q
F ss
Fe
QS
a′
c
Fe
DS
Q
D
F sqs
wt
b
d
F sds
c′
b′
a
FIGURE 6.24 Stator mmf vector in the stator and excitation reference frames.
184
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Thus, the motor equation in excitation reference frame can be expressed as
veS = Rs ies + p + jω λes ,
veR = Rr ieR + p + jω − jω0 λeR = Rr ieR + p + jωr λeR ,
(6.64)
(6.65)
where ωr = ω − ω0 as the slip speed, and
λeS
L
= s
λeR
Lm
Lm
Lr
e
iS
.
ieR
(6.66)
Again, the rotor voltage vector is normally assumed to be zero because of
the shorted rotor winding.
The torque equation in the excitation reference frame is similar to that in
the stator frame as
T=
P
P
e e
e e
iDR − iDS
iQR = Lm Im ieS ie∗
Lm iQS
e .
3
3
(6.67)
In general, accurate control of instantaneous torque produced by a motor
is required in high-performance drive systems, such as EV and HEV propulsions. The torque developed in the motor is a result of the interaction between
current in the armature winding and the magnetic field produced in the
stator field of the motor. The field should be maintained at a certain optimal level, sufficiently high to yield a high torque per unit of ampere, but
not too high to result in excessive saturation of the magnetic circuit of
the motor. With a fixed field, the torque is proportional to the armature
current.
Independent control of the field and armature currents is desirable. In a
similar manner to that of a DC motor, the armature winding in induction
motors is also on the rotor, while the field is generated by currents in the
stator winding. However, the rotor current is not directly derived from an
external source, but results from the EMF induced in the winding as a result
of the relative motion of the rotor conductors with respect to the stator field. In
the most commonly used squirrel-cage motors, only the stator current can be
directly controlled, since the rotor winding is not accessible. Optimal torque
production conditions are not inherent due to the absence of a fixed physical
disposition between the stator and rotor fields, and the torque equation is
nonlinear. FOC or vector control can realize the optimal control for transient
operation of an induction drive. The FOC can decouple the field control from
the torque control. A field-oriented induction motor emulates a separately
excited DC motor in two aspects:
1. Both the magnetic field and the torque developed in the motor can be
controlled independently.
185
Electric Propulsion Systems
2. Optimal conditions for torque production, resulting in the maximum
torque per unit ampere, occur in the motor both in the steady state
and in transient conditions of operations.
As mentioned in Section 6.1.1, the optimal torque production conditions are
inherently satisfied in a DC motor (refer to Figure 6.3). The armature current
ia , supplied through brushes, is always orthogonal to the flux vector (field
flux), λf , produced in the stator and linking the rotor winding. In effect, the
developed torque, T, is proportional both to the armature current and the
field flux, that is,
T = KT ia λf ,
(6.68)
where KT is a constant depending on the physical parameters of the motor.
Thus, the torque of separately excited DC motors can be controlled by
independently controlling the armature current and flux as mentioned in
Section 6.1.2.
To emulate this independent armature and field control characteristic of a
DC motor, the torque Equation 6.67 can be rearranged so that the torque is
expressed in terms of the stator current and rotor flux. From Equation 6.66,
the flowing equation can be obtained as
ieR =
1 e
λ − Lm ieS .
Lr R
(6.69)
Torque Equation 6.67 can be rewritten as
T=
P Lm e e
e
iQS λDR − iDS
λeQR ,
3Rr τr
(6.70)
where τr = Lr /Rr is the rotor time constant.
In Equation 6.70, if
λeQR = 0,
(6.71)
then
T=
P Lm e e
λ i .
3Rr τr DR QS
(6.72)
Clearly, Equation 6.72 is analogous to Equation 6.68, describing a separately
excited DC motor.
186
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Substituting veR = 0 (shorted rotor winding) into Equation 6.65 yields
Rr ieR + p + jωr λeR = 0,
(6.73)
and substituting Equation 6.69 into Equation 6.73 yields
pλeR =
1 Lm ieS − 1 + jωr τr λeR .
τr
(6.74)
Lm e
1
i − λe .
τr DS τr DR
(6.75)
Thus,
pλeDR =
e .
Equation 6.75 indicates that the flux λeDR is produced by the current, iDS
Thus, the torque produced can be represented by the block diagram as shown
in Figure 6.25.
Furthermore, Equation 6.75 can be expressed into a transfer function as
G(p) =
λeDR
Lm
.
=
e
iDS
τr p + 1
(6.76)
Thus, the block diagram in Figure 6.25 can be further reduced as shown in
Figure 6.26.
If conditions (6.71) and λeDR = constant t are satisfied, that is, λeQR = 0 and
e = 0, that is, ie = jie . At the same
pλeDR = 0, then Equation 6.64 yields iDR
QR
R
e
e
time, λR = λDR . Consequently, vectors ieR and λeR are orthogonal, which represents the optimal conditions for the torque production, analogous to a DC
motor. In an induction motor, the optimal torque-production conditions are
always satisfied in the steady state. However, in transient operation, the motor
needs delicate control to achieve this optimal torque production.
1 e
tr lDS
ieDS
Lm
tr
Lm e
tr iDS
pleDR
1
tr
1
p
leDRieQS
leDR
PLm
3t r
ieQS
FIGURE 6.25 Block diagram of an induction motor with λeQR = 0.
T
187
Electric Propulsion Systems
leDR i eQS
i eQS
KT =
P Lm
3Rr tr
T
leDR
G(p)
i eDS
FIGURE 6.26 Block diagram of a field-orientated induction motor.
6.2.5.2
Control
As demonstrated in the above section, the field orientation principle defines
the conditions of optimal torque production. Orthogonality of the rotor current and stator flux vectors must be maintained at all times. This is inherently
satisfied in the steady state when the rotor settles down to such a speed that
the developed torque matches the load torque. Under transient conditions,
however, in order to meet the field orientation principle conditions, special
techniques are required to provide an algorithmic equivalent of the actual
physical disposition between the stator and rotor fields of the emulated DC
motor.
The general block diagram of a vector control system for an induction motor
drive is shown in Figure 6.27. A field orientation system produces reference
∗ , i∗ , and i∗ , of the stator currents, based on the input reference
signals, ias
cs
bs
∗
values, λr and T ∗ , of the rotor flux and motor torque, respectively, and the signals corresponding to selected variables of the motor. An inverter supplies the
motor currents, ias , ibs , and ics , such that their waveforms follow the reference
∗ , i∗ , and i∗ .
waveform, ias
cs
bs
e and
As shown in Figure 6.26, in a field orientated induction motor, the iDS
e
e
iQS components of the stator current vector, iS , in the excitation frame can
be used for independent control of the motor field and torque, respectively.
Hence, the field orientation system as shown in Figure 6.27 first converts λ∗r
e∗ and ie∗ , of the vector of
and T ∗ into the corresponding reference signals, iDS
QS
∗ , i∗ ,
the stator current and then transfers these into the reference signals, ias
bs
∗ , of the stator phase current, which are to be produced by the inverter.
and ics
∗ , i∗ , and i∗ , can be calculated using dq to abc
The stator phase currents, ias
cs
bs
s∗
transformation (see Equation 6.46) if the corresponding reference signals, ids
s∗ , in the stator reference frame are known. This is a simple scalar, or
and iqs
static, transformation, since the elements of the transformation matrix used
to perform this operation are constant.
However, it can be seen from Equation 6.63 that a dynamic transformation,
s∗ and is∗ from ie∗ and ie∗ .
that is, one involving time, is required to determine ids
qs
DS
QS
188
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
DC supply
i*as
l*r
Field
orientation
system
T*
i*bs
Inverter
i*cs
ias ibs ics
Motor
signals
Motor
FIGURE 6.27 General block diagram of a vector control system for an induction motor.
D
In Figure 6.24, it does not indicate which vector the excitation reference frame
DQ aligns with. Clearly, any one of the vectors can be used as a reference with
which the excitation frame is to be aligned. Usually, it is the rotor flux vector,
λsr , along which the excitation frame is orientated. This method is usually
referred to as the rotor flux orientation scheme,8 as shown in Figure 6.28.
If the angular position of the rotor flux vector in the stator reference frame
is denoted by θr , the DQ to dq transformation in the described scheme is
r
ls
q
Q
le
DR
lsqr
w
l
r
e R
lQ
=
0
qr
d
lsdr
FIGURE 6.28 Orientation of the excitation reference frame along the rotor flux vector.
189
Electric Propulsion Systems
expressed as
s∗ ids
cos(θr )
s∗ =
iqs
sin(θr )
e∗ − sin(θr ) iDS
e∗ .
cos(θr ) iQS
(6.77)
It can be observed that this orientation of the orientation frame inherently
satisfies the field orientation principle condition in Equation 6.71. The rotor
e component of the stator current vector—
flux is controlled by adjusting the iDS
e
independently of the torque control, which is realized by means of the iQS
component. The only requirement for this scheme is an accurate identification
of angle θr , that is, the position of λsr . This can be done in either a direct or
indirect way.
6.2.5.3
Direction Rotor Flux Orientation Scheme
In direct field orientation systems, the magnitude and angular position
(phase) of the reference flux vector, λer , are either measured or estimated from
the stator voltage and current using flux observers. For example, Hall sensors
can be used to measure magnetic fields. Placing the sensors in the air gap of
the motor, on the d and q axes, allows determination of the corresponding
components of vector λsm of the mutual flux (air gap flux). However, this air
gap flux differs from the rotor flux, which is taken as the reference flux vector
and needs derivation from the air gap flux λsm . Referring to the dynamic T
equivalent circuit shown in Figure 6.29, the flux appearing across the mutual
inductance Lm is
λsm = Lm ism = Lm iss + irs
(6.78)
or
isr =
i ss
Rs
1 s
λ − iss .
Lm m
Lls
Llr
(6.79)
Rr
i sm
V ss
plss
plsm
Lm
i sr
plsr
FIGURE 6.29 Dynamic T equivalent circuit of an induction motor.
+
–
jw0isr
190
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Esqm
q
a'
Hall sensors
b
d
b'
Rotor flux calculator
Lr
Lm
c'
i as
i bs
i cs
i sqs
abc
dq
i sds
Llr
Llr
–
+
lsqr
lsdr
–
+
Rect.
Polar
lr
qr
a
Esqm
Lr
Lm
FIGURE 6.30 Determination of the magnitude and position of the rotor flux vector using a Hall
sensor and a rotor flux calculator.3
Since λsr differs from the λsm by only the leakage flux in the rotor, then
Lr s
1 s
s
s
s
s
s
λr = λm + Llr ir = λm + Llr
λ − λs =
λ − Llr iss .
(6.80)
Lm m
Lm m
A microprocessor-based rotor flux calculator is shown in Figure 6.30. It
performs the algebraic operations as the follows:
s and is are calculated from the actual stator phase cur1. Signals ids
qs
rents, ias , ibs , and ics , using the abc to dq transformation expressed in
Equation 6.46.
2. Using Equation 6.80, signals λsdr and λsqr are calculated.
3. Magnitude λr and the phase θr of the rotor flux vector are determined
using the rectangular to polar coordinate transformation.
It must be pointed out that the orthogonal spacing of the flux sensors in Figure 6.30 applies only to two-pole machines. In a P-pole machine, the sensors
must be placed 180/P from each other.
Since λeDR = λr (see Figure 6.28), then the output variable, λr , of the rotor
flux calculator can be used as a feedback signal in the field control loop. The
same variables can also be used for calculation of the developed torque as
shown in Figure 6.31. The torque calculator computes torque in the following
steps:
1. The static abc to dq transformation is performed on the stator currents
s and is .
ias , ibs , and ics to obtain ids
qs
2. Angle θr supplied by the rotor flux calculator is substituted into Equas and is into ie and ie
tion 6.63 for ωt in order to transfer signals ids
qs
DS
QS
components of the stator current vector in the excitation frame.
191
Electric Propulsion Systems
lsdm
lsqm
ias
Rotor flux
calculator
ics
ics
Torque calculator
abc
dq
qr
isds
ieqs
dq
DQ
i eDS
i eQS
lr = leDR
KT =
P Lm
3Rr tr
T
FIGURE 6.31 Torque calculator.
3. Magnitude, λr , of the rotor flux, also supplied by the rotor flux cale and by the
culator and presumed equal to λeDS , is multiplied by iQS
torque constant KT to calculate the developed torque.
Figure 6.31 shows the torque calculation process block diagram.
Figure 6.32 illustrates an independent flux and torque control block diagram, based on the vector control of an induction motor with direct rotor
flux orientation. In the system, proportional-plus-integral (PI)-based field and
e∗ and ie∗ in excitorque controllers are used to generate the control signals iDS
QS
∗
tation frame by comparing the target rotor flux, λr , and target torque T ∗ ,
e∗ and ie∗ in excitation
with the actual rotor flux, λr , and torque, T. Then, iDS
QS
s∗
s∗
frame are transferred into ids and iqs in stator reference frame using rotor
s∗ and is∗ in stator reference
flux angle (see Equation 6.63). Furthermore, ids
qs
DC Supply
l*r
T*
lr
–
+
Dl*r
DT *
+
–
Field
controller e * qr
iDS
DQ
e*
iQS
Torque
controller
dq
s*
i ds
isqs*
dq
abc
i *as
i *bs
i *cs
Rotor
flux
calculator
lsdm
lsqm
i as
i bs
i cs
Inverter
Torque
calculator
lr
qr
Motor
FIGURE 6.32 Vector control system for an induction motor with direct rotor flux orientation.
192
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
∗ , i∗ , and i∗ through static
frame are transferred into phase current signals ias
cs
bs
transformation (see Equation 6.46). The phase current signals, as the reference
signals, are used to control the power electronics of the inverter to generate
corresponding phase current ias , ibs , and ics .
In practice, the ratio of Lr to Lm , and the rotor leakage inductance, Lls , which
are used in the rotor flux calculator (see Figure 6.30), are not significantly
affected by changes in the operation conditions of the motor, such as the
winding temperature or saturation of the magnetic circuit. Therefore, the field
orientation techniques described are considered to be the most robust and
accurate. However, it requires the placement of vulnerable Hall sensors in
the air gap of the motor, to the detriment of the cost and reliability of the
drive system.
6.2.5.4
Indirect Rotor Flux Orientation Scheme
The presence of vulnerable Hall sensors in vector control with direct rotor flux
orientation would weaken the reliability and enhance the cost of the motor
drive. The indirect approach is to obtain the rotor flux position by the calculation of the slip speed, ωr , required for correct field orientation, and the
imposition of this speed on the motor.
If the synchronous speed necessary to maintain the orthogonal orientation
of vectors λeR and ieR in the given operating conditions of the motor is denoted
by ω∗ , the θr angle can be expressed as
θr =
0
t
∗
t
ω dt =
ω∗r dt +
0
t
0
ω0 dt =
0
t
ω∗r dt + θ0 ,
(6.81)
where ω∗ , ωr , and ω0 are the synchronous speed, slip speed, and rotor speed,
respectively, and θ0 is the angular displacement of the rotor, which is easy to
measure using a shaft position sensor.
The required value of the slip speed ω∗r can be computed from Equation
6.69. Since λeR = λeDR , Equation 6.69 becomes
iRe =
1 e
λDR − Lm ieS .
Lr
(6.82)
Substituting Equation 6.82 into Equation 6.73 gives the real and imaginary
parts as
e
λeDR 1 + τr p = Lm iDS
,
(6.83)
and
e
ωr τr λeDR = Lm iQS
.
(6.84)
193
Electric Propulsion Systems
DC supply
l*r
tr
Lm
T*
1
Lml*r
p+
e*
i DS
1
tr
e*
i QS
∏
w*r
1
p
q*
dq
ias
i*as
i*bs
*
ab i cs
s*
i ds
DQ
dq
s*
i qs
ibs
ics
c
Inverter
+
+
Motor
q0
Position
sensor
FIGURE 6.33 Vector control system for an induction motor with indirect rotor flux orientation.
e with ω∗ , λ∗ , and ie∗ , respectively, in Equation 6.84
Replacing ωr , λeDR , and iQS
r
r
QS
yields
ω∗r
e∗
Lm iQS
=
.
τr λ∗r
(6.85)
e in Equation 6.83 with λ∗ and ie∗ yields
Replacing λeDr and iDS
r
DR
e∗
=
iDS
1 + τr p ∗
λR .
Lm
(6.86)
e∗ can be obtained as
From the torque Equation 6.68, the signal iQS
e∗
=
iQS
T∗
.
KT λ∗r
(6.87)
A vector control system for an induction motor based on the indirect rotor
flux orientation scheme is shown in Figure 6.33. The rotor flux and developed
torque are controlled in a feedforward manner. As a consequence of this, performance of the system strongly depends on an accurate knowledge of motor
parameters, a requirement that is difficult to satisfy in practical applications.
On the other hand, a major advantage of such a system is that a standard
motor can be used, whose rotor position is easily measurable by an external
sensor. Since the control scheme presented here constitutes an extension of
the scalar torque control methods, the reference flux and torque values must
satisfy the safe operation area condition described.9
6.2.6 Voltage Source Inverter for FOC
The power electronic inverter for FOC of induction motor drives has the
same topology as shown in Figure 6.21a, which is repeatedly illustrated in
Figure 6.34. The power switches in a given leg (a, b, or c) must never both
be in ON-state, since this would cause a short-circuit. On the other hand,
if both the switches on the same leg are in OFF-state, then the potential of
194
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
SB
SA
SC
DA'
DC'
DB'
Zb
+
a
ia
Vdc
b
–
ib
SB'
SA'
DA
c
n
ic
SC'
DB
Za
Zc
DC
FIGURE 6.34 Circuit diagram of a three-phase voltage source inverter.
the corresponding output terminal is unknown to the control system of the
inverter. The circuit can be completed through either the upper or lower diode
and, consequently, the potential can be equal to that of either positive bus (+)
or negative bus (−). Therefore the inverter is controlled in such a way that,
in a given leg, either the upper switch (SA, SB, or SC) is ON and the lower
switch (SA , SB , or SC ) is OFF or vice versa, the upper switch is OFF and the
lower switch is ON.
Since only two combinations of states of the switches in each leg are allowed,
a switching variable can be assigned to each phase of the inverter. In effect,
only eight logic states are permitted for the whole power circuit. Defining the
switching variables as
0
1
0
b=
1
0
c=
1
a=
if SA is OFF and SA is ON,
if SA is ON and SB is OFF,
(6.88)
if SB is OFF and SB’ is ON,
if SB is ON and SB’ is OFF,
(6.89)
if SC is OFF and SC’ is ON,
if SC is ON and SC’ is OFF.
(6.90)
The instantaneous values of the line-to-line output voltage of the inverter are
given by
vab = Vdc (a − b),
(6.91)
vbc = Vdc (b − c),
(6.92)
vca = Vdc (c − a),
(6.93)
where Vdc is the voltage of the DC supply of the inverter.
195
Electric Propulsion Systems
In a balanced three-phase system, the line-to-neutral voltage can be
calculated from the line-to-line voltages as9
1
(vab − vca ),
3
1
vb = (vbc − vab ),
3
1
vc = (vca − vbc ).
3
va =
(6.94)
(6.95)
(6.96)
Hence, after substituting Equations 6.88 through 6.90 into Equations 6.94
through 6.96, the line-to-neutral voltages are given by
Vdc
(2a − b − c),
3
Vdc
vb =
(2b − c − a),
3
Vdc
vc =
(2c − a − b).
3
va =
(6.97)
(6.98)
(6.99)
From Equations 6.91 through 6.93, line-to-line voltages can assume only three
values: −Vdc , 0, and Vdc . However, Equations 6.97 through 6.99 give out
five line-to-neutral voltage values: (−2/3)Vdc , (−1/3)Vdc , 0, (1/3)Vdc , and
(2/3)Vdc . The eight logic states of the inverter can be numbered from 0 to 7
using the decimal equivalent of binary number abc2 . For example, if a = 1,
b = 0, and c = 1, then abc2 = 1012 = 510 , and the inverter is said to be in state
5. Taking Vdc as the base voltage, and at state 5, the per-unit output voltages
are: vab = 1, vbc = −1, vca = 0, va = 1/3, vb = −2/3, and vc = 1/3.
Performing the abc to dq transformation, the output voltage can be represented as space vectors, the stator reference frame, each vector corresponding
to a given state of the inverter. The space vector diagrams of line-to-line
voltages, identified by the superscript of LTL, and line-to-neutral voltages,
identified by the superscript LTN, of voltage source inverter are shown in
Figure 6.35. The vectors are presented in the per-unit format.
6.2.6.1 Voltage Control in Voltage Source Inverter
A large number of different (PWM) technologies have been developed and
implemented in practical inverters. Currently, one of the most popular methods is based on the concept of space vectors of the inverter voltages, as shown
in Figure 6.35. This method is more suitable for application with the field
orientation control of induction motor drives.
For a wye-connected induction motor, the load currents are generated
by the line-to-neutral voltages of the inverter. Thus, the motor operation is
controlled by the line-to-neutral voltage inverter voltages.
196
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
q
j 3
V6LTL
V2LTN
V2LTL
V6LTN
–j 3
2
V4LTN
V3LTN
–1.5
V3LTL
–1
V4LTL
–0.5
0.5
1
d
1.5
V5LTL
–j 3 V5LTN
2
V1LTN
V1LTL
–j 3
FIGURE 6.35 Space vectors of output voltage of a voltage source inverter.
Space vectors of the line-to-neutral voltages are shown in Figure 6.36,
together with an arbitrary vector v∗ , to be generated by the inverter. In addition to showing six nonzero vectors (state 1 to state 6), another two zero
vectors, corresponding to state 0 and state 7, are also shown. Clearly, only
vectors v0 to v7 , further referred to as base vectors, can be produced at a
q
–j
3
2
V6 = Vy
V*
dy v
y
V2
V3
–1
V0 = V7 = 0
a
dxvx 0.5
–0.5
V1
–j
3
2
V5
FIGURE 6.36 Illustration of the space vector PWM strategy.
V4 = Vx
1
d
197
Electric Propulsion Systems
given instant of time. Therefore, vector v∗ represents an average rather than
an instantaneous value, the average being taken over a period of switching,
or sampling, interval, which, in practice, constitutes a small fraction of the
cycle of the output frequency. The switching interval, at the center of which
the reference vector is located, is shown in Figure 6.36 as the shaded segment.
The nonzero base vectors divided the cycle into six, 60◦ -wide sectors. The
desired voltage vector v∗ , located in a given sector, can be synthesized as a
linear combination of the two adjacent base vectors, vx and vy , which are
framing the sector, and either one of the two zero vectors. That is,
v∗ = dx vx + dr vy + dz vz ,
(6.100)
where vz is the zero vector, dx , dy , and dz denote the duty ratios of the states
of x, y, and z within the switching interval, respectively. For instance, the
reference voltage vector v∗, in Figure 6.36, is located within the first sector
in which vx = v4 and vy = v6 ; hence, it can be produced by an appropriately
timed sequence of states 4, 6, and 0 or 7 of the inverter.
The state duty ratio is defined as the ratio of the duration of the state to the
duration of the switching interval. Therefore,
dx + dy + dz = 1.
(6.101)
It can be seen that under this condition the locus of the maximum available
vectors v∗ constitutes the hexagonal envelope of the base vectors, as shown
in Figure 6.36. To avoid low-order voltage harmonics, resulting from the noncircular shape of the envelope, the locus of the synthesized voltage vectors
is, in practice, limited to the circle as shown in Figure 6.36. Consequently,
the
√
maximum available magnitude, Vmax , of the resulting voltage is ( 3/2)Vdc .
With respect to vector v∗ , in Figure 6.36, Equation 6.100 can be written as
v∗ = MVmax ejα = dx v4 + dy v6 + dz vz ,
(6.102)
where M is the modulation index, adjustable within the 0 to 1 range, and
α denotes the angular position of the vector v∗ inside the sector, that is, the
∗
angular distance between vectors
√ v and vx . As seen in Figure 6.36, vx = v4 =
1 + j0 p.u., vy = v6 = 1/2 + j 3/2 p.u., and vz (either v0 or v7 ) is zero, and
√
Vmax = ( 3/2)Vdc , Equation 6.100 can be rewritten as
√
3
1
M cos(α) = dx + dy ,
2
2
and
(6.103)
√
√
3
3
M sin(α) =
dy .
2
2
(6.104)
198
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Thus, dx and dy can be expressed as
dx = M sin(60◦ − α),
(6.105)
dy = M sin(α),
(6.106)
and
dz = 1 − dx − dy .
(6.107)
The same equations can be applied to the other sectors.
The simple algebraic formulas 6.105 through 6.107 allow duty ratios of the
consecutive logic states of an inverter to be computed in real time. Due to the
freedom of choice of the zero vectors, various state sequences can be enforced
in a given sector. Particularly efficient operation of the inverter is obtained
when the state sequence in consecutive switching intervals is
x − y − z z − y − x . . . ,
(6.108)
where z = 0 in sectors v6 − v2 , v3 − v1 and v5 − v4 , and z = 7 in the remaining
sectors. Figure 6.37 shows an example of switching signals and output voltages for a voltage source inverter in the previously described PWM mode
with M = 0.7 and 20◦ width of the switching interval.9
6.2.6.2
Current Control in Voltage Source Inverter
Since the output currents of an inverter depend on load, feedforward current control is not feasible and a feedback from current sensors is required.
There exist a number of different control technologies. The simplest one is the
controller, based on the so-called “hysteretic” or “bang-bang.”
The block diagram of a current control voltage source inverter is shown in
Figure 6.38. The output currents ia , ib , and ic of the inverter are sensed and
compared with the reference current signals ia∗ , ib∗ , and ic∗ . Current error signals
Δia , Δib , and Δic are then applied to the hysteresis current controller, which
generates switching signals, a, b, and c, for the inverter switches.
The input–output characteristic of the phase-a hysteretic current controller
is shown in Figure 6.39. The width of the hysteretic loop, denoted by h, represented the tolerance bandwidth for the controlled current. If the current error,
Δia , is greater than h/2, that is, ia is unacceptably lower than the reference current, ia∗ , the corresponding line-to-neutral voltage, va , must be increased. From
Equation 6.97, this voltage is most strongly affected by the switch variable a;
hence, it is this variable that is regulated by the controller, and is set 1 to a in
the described situation. Conversely, an error of less than −h/s results in a = 0;
in order to decrease the current ia that stays within the tolerance band, the
other two controllers are operated in a similar manner.
The width, h, of the tolerance band affects the switching frequency of the
inverter. The narrower the band, the more frequently the switching takes place
and the higher the quality of the current will be. This is illustrated in Figures
199
Electric Propulsion Systems
a
b
c
Va
Vb
Vc
Vab
Vbc
Vca
Time
FIGURE 6.37 Example switching signals and output voltage for voltage source inverter in the
PWM operation mode.
ia
Zb
ib
Inverter
Vdc
ic
n
Zc
c
b
a
–
Dia
Dib
Dic
Hysteretic
current
controllers
+
–
i*a
+
–
i*b
+
i*c
FIGURE 6.38 Block diagram of a current-controlled voltage source inverter.
Za
200
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
a
1–
Dia
0
h/2
–h/2
FIGURE 6.39 Input–output characteristics of a hysteresis current controller.
a
b
c
Va
Vb
Vc
ia, i*a
0
1/2p
ib, i*b
p
ic, ic*
3/2p
2p
FIGURE 6.40 Current-controlled voltage source inverter (10% tolerance bandwidth).
6.40 and 6.41, in which the switching variables, line-to-neutral voltages, and
currents for an inverter supplying a resistive-plus-inductive load are at values of h = 10 and 5% of the amplitude of the reference current, respectively.
In practice, the tolerance bandwidth should be set to a value that represents
an optimal trade-off between the quality of the currents and the efficiency of
the inverter.
6.3
Permanent Magnetic BLDC Motor Drives
By using high-energy PMs as the field excitation mechanism, a PM motor
drive can be potentially designed with high power density, high speed, and
201
Electric Propulsion Systems
a
b
c
Va
Vb
Vc
ia, i*a
0
1/2p
ib, i*b
p
ic, ic*
3/2p
2p
FIGURE 6.41 Current-controlled voltage source inverter (5% tolerance bandwidth).
high operation efficiency. These prominent advantages are quite attractive for
their application in EVs and HEVs. Of the family of PM motors, the BLDC
motor drive is the most promising candidate for EV and HEV applications.4
The major advantages of BLDC motor includes the following:
•
High efficiency: BLDC motors are the most efficient of all electric
motors. This is due to the use of PMs for the excitation, which consume no power. The absence of mechanical commutator and brushes
means low mechanical friction losses and therefore higher efficiency.
•
Compactness: The recent introduction of high-energy density magnets
(rare-earth magnets) has allowed achieving very high flux densities
in the BLDC motor. This allows achieving accordingly high torques,
which in turns allows making the motor small and light.
•
Ease of control: The BLDC motor can be controlled as easily as a DC
motor because the control variables are easily accessible and constant
throughout the operation of the motor.
•
Ease of cooling: There is no current circulation in the rotor. Therefore,
the rotor of a BLDC motor does not heat up. The only heat production
is on the stator, which is easier to cool than the rotor because it is static
and on the periphery of the motor.
•
Low maintenance, great longevity, and reliability: The absence of brushes
and mechanical commutators suppresses the need for associated
regular maintenance and suppresses the risk of failure associated
with these elements. The longevity is therefore only a function of the
winding insulation, bearings, and magnet life-length.
202
•
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Low noise emissions: There is no noise associated with the commutation because it is electronic and not mechanical. The driving converter
switching frequency is high enough so that the harmonics are not
audible.
However, BLDC motor drives also suffer from some disadvantages as
follows:
•
Cost: Rare-earth magnets are much more expensive than other
magnets and result in an increased motor cost.
•
Limited constant power range: A large constant power range is crucial to
achieving high vehicle efficiencies. The PM BLDC motor is incapable
of achieving a maximum speed greater than twice the base speed.
•
Safety: Large rare-earth PMs are dangerous during the construction
of the motor because flying metallic objects are attracted towards
them. There is also a danger in the case of vehicle wreck if the wheel
is spinning freely: the motor is still excited by its magnets and high
voltage is present at the motor terminals that can possibly endanger
the passengers or rescuers.
•
Magnet demagnetization: Magnets can be demagnetized by large
opposing magnetomotive forces and high temperatures. The critical
demagnetization force is different for each magnet material. Great
care must be brought to cooling the motor, especially if it is built
compact.
•
High-speed capability: The surface-mounted PM motors cannot reach
high speeds because of the limited mechanical strength of the
assembly between the rotor yoke and the PMs.
•
Inverter failures in BLDC motor drives: Because of the PMs on the rotor,
BLDC motors present major risks in the case of short-circuit failures
of the inverter. Indeed, the rotating rotor is always energized and
constantly induces an EMF in the short-circuited windings. A very
large current circulates in those windings and an accordingly large
torque tends to block the rotor. The dangers of blocking one or several
wheels of a vehicle are non-negligible. If the rear wheels are blocked
while the front wheels are spinning, the vehicle will spin uncontrollably. If the front wheels are blocked, the driver has no directional
control over the vehicle. If only one wheel is blocked, it will induce a
yaw torque that will tend to spin the vehicle, which will be difficult to
control. In addition to the dangers to the vehicle, it should be noted
that the large current resulting from an inverter short circuit poses a
risk to demagnetize and destroy the PMs.
Open circuit faults in BLDC motor drives are no direct threat to the vehicle
stability. The impossibility of controlling a motor due to an open circuit may,
203
Electric Propulsion Systems
however, pose problems in terms of controlling the vehicle. Because the magnets are always energized and cannot be controlled, it is difficult to control a
BLDC motor in order to minimize the fault. This is a particularly important
issue when the BLDC motor is operated in its constant power region. Indeed,
in this region, a flux is generated by the stator to oppose the magnet flux and
allow the motor to rotate at higher speeds. If the stator flux disappears, the
magnet flux will induce a large EMF in the windings, which can be harmful
to the electronics or passengers.
6.3.1
Basic Principles of BLDC Motor Drives
A BLDC motor drive consists mainly of the BLDC machine, the digital signal processor (DSP) based controller, and the power electronics-based power
converter, as shown in Figure 6.42. Position sensors H1, H2, and H3 sense
the position of the machine rotor. The rotor position information is fed to the
DSP-based controller, which, in turn, supplies gating signals to the power
converter by turning on and turning off the proper stator pole windings of
the machine. In this way, the torque and speed of the machines are controlled.
6.3.2
BLDC Machine Construction and Classification
BLDC machines can be categorized by the position of rotor PM, geometrically,
according to the way in which the magnets are mounted on the rotor. The
magnets can either be surface mounted or interior mounted.
Figure 6.43a shows the surface-mounted PM rotor. Each PM is mounted
on the surface of the rotor. It is easy to build, and specially skewed poles are
easily magnetized on this surface-mounted type to minimize cogging torque.
But there is a possibility that it will fly apart during high-speed operation.
DSP
controller
Coil 1
Rotor
N
H1
DC
power
supply
H2
N
S
S
Coil 3
FIGURE 6.42 BLDC motor.
H3 Coil 2
Power
converter
204
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
(a)
(b)
N
S
S
S
N
N
N
N
S
S
S
S
N
N
N
S
FIGURE 6.43 Cross sectional view of the PM rotor: (a) surface-mounted PM rotor; and (b)
interior-mounted PM rotor.
Figure 6.43b shows the interior-mounted PM rotor. Each PM is mounted
inside the rotor. It is not as common as the surface-mounted type but it is
a good candidate for high-speed operation. Note that there is inductance
variation for this type of rotor because the PM part is equivalent to air in the
magnetic circuit calculation.
In the case of the stator windings, there are two major classes of BLDC motor
drives, both of which can be characterized by the shapes of their respective
back EMF waveforms, namely trapezoidal and sinusoidal.
The trapezoidal-shaped back EMF BLDC motor is designed to develop
trapezoidal back EMF waveforms. It has the following ideal characteristics:
•
Rectangular distribution of magnet flux in the air gap.
•
Rectangular current waveform.
•
Concentrated stator windings.
Excitation waveforms take the form of quasisquare current waveforms with
two 60◦ electrical intervals of zero current excitation per cycle. The nature
of the excitation waveforms for trapezoidal back EMF permits some important system simplifications compared to sinusoidal back EMF machines. In
particular, the resolution requirements for the rotor position sensor are much
lower since only six commutation instants are necessary per electrical cycle.
Figure 6.44 shows the winding configuration of the trapezoidal-shaped back
EMF BLDC machine.
Figure 6.45a shows equivalent circuit and (b) shows trapezoidal back
EMF, current profiles, and Hall sensor signals of the three-phase BLDC motor
drive. The voltages seen in this figure, ea , eb , and ec , are the line-to-neutral
back EMF voltages, the result of the PM flux crossing the air gap in a radial
direction, and cutting the coils of the stator at a rate proportional to the
rotor speed. The coils of the stator are positioned in the standard three-phase
full-pitch, concentrated arrangement, and thus the phase trapezoidal back
EMF waveforms are displaced by 120◦ electrical degrees. The current pulse
205
Electric Propulsion Systems
ia
Stator
ic
Rotor
ib
ib
Air gap
ic
ja
FIGURE 6.44 Winding configuration of the trapezoidal-shaped back EMF BLDC.
generation is a “120◦ -on and 60◦ -off” type, meaning each phase current is
flowing for 2/3 of an electrical 360◦ period, 120◦ positively and 120◦ negatively. To drive the motor with maximum and constant torque per ampere, it
is desired that the line current pulses be synchronized with the line-neutral
back EMF voltages of the particular phase.
A sinusoidal-shaped back EMF BLDC motor is designed to develop
sinusoidal back EMF waveforms. It has the following ideal characteristics:
1. Sinusoidal distribution of magnet flux in the air gap.
2. Sinusoidal current waveforms.
3. Sinusoidal distribution of stator conductors.
The most fundamental aspect of the sinusoidal-shaped back EMF motor is that
the back EMF generated in each phase winding by the rotation of the magnet should be a sinusoidal wave function of rotor angle. The drive operation
of the sinusoidal-shaped back EMF BLDC machine is similar to the AC synchronous motor. It has a rotating stator MMF wave like a synchronous motor
and therefore can be analyzed with a phasor diagram. Figure 6.46 shows the
winding configuration of the sinusoidal-shaped back EMF BLDC machine.
6.3.3
Properties of PM Materials
There are three classes of PMs currently used for electric motors:
1. Alnicos (Al, Ni, Co, Fe).
2. Ceramics (ferrites), for example, barium ferrite (BaO × 6Fe2 O3 ) and
strontium ferrite (SrO × 6Fe2 O3 ).
206
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
a
(a)
R
L
ea
eb
n
ec
L
L
R
R
b
c
(b)
ea, ia
7p/6
p/6
p/2
5p/6 p
3p/2
11p/6
2p
eb, ib
ec, ic
Hall sensor A
Hall sensor B
Hall sensor C
Phase current : ia, ib, ic
Back EMF : ea, eb, ec
Hall sensor signals
FIGURE 6.45 (a) Three-phase equivalent circuit and (b) back EMFs, currents, and Hall sensor
signals of a BLDC motor.
3. Rare-earth materials, that is, samarium–cobalt (SmCo), and
neodymium–iron–boron (NdFeB).
Demagnetization curves of the above PM materials are shown in
Figure 6.47.10
6.3.3.1 Alnico
The main advantages of Alnico are its high magnetic remanent flux density
and low-temperature coefficients. The temperature coefficient of its remanent
207
Electric Propulsion Systems
Coil
Rotor
Gap, g
Stator
FIGURE 6.46 Winding configuration of sinusoidal-shaped back EMF BLDC.
B (T)
T = 20∞C
1.4
Neodymium-iron
boron
1.2
1.0
Samarium
cobalt
0.8
0.6
Alnico
0.4
Ferrite
900
800
700
600
500 400
–H (KA/m)
300
0.2
200
100
FIGURE 6.47 Demagnetization curves for different PM materials.
magnetic flux density Br , or remanence, is 0.02%/◦ C and the maximum service
temperature is 520◦ C. These advantages allow quite a high air gap flux density
and high operating temperature. Unfortunately, coercive force is very low and
the demagnetization curve is extremely nonlinear. Therefore, it is very easy
not only to magnetize but also to demagnetize Alnico. Alnico magnets have
been used in motors having ratings in the range from a few watts to 150 kW.
Alnicos dominated the PM industry from the mid-1940s to about 1970, when
ferrites became the most widely used materials.10
208
6.3.3.2
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Ferrites
Barium and strontium ferrites were invented in the 1950s. A ferrite has a
higher coercive force than that of Alnico, but at the same time has a lower
remanent magnetic flux density. Temperature coefficients are relatively high,
that is, the coefficient of Br is 0.20%/◦ C and the coefficient of coercive field
strength, Hc , or coercivity is 0.27%/◦ C. The maximum service temperature
is 400◦ C. The main advantages of ferrites are their low cost and very high
electric resistance, which means no eddy-current losses in the PM volume.
6.3.3.3
Rare-Earth PMs
During the last three decades, greater progress regarding available energy
density (BH)max has been achieved with the development of rare-earth PMs.
The first generation of the rear-earth PMs based on the composition of
samarium–cobalt (SmCo5 ) was inverted in the 1960s and has been commercially produced since the early 1970s. Today, it is a well-established hard
magnetic material. SmCo5 has the advantages of high remanent flux density, high coercive force, high-energy product, linear demagnetization curve,
and low-temperature coefficient. The temperature coefficient of Br is 0.03–
0.045%/◦ C and the temperature coefficient of Hc is 0.14 − 0.40%/◦ C. The
maximum service temperature is 250–300◦ C. It is well suited to build motors
with low volume and consequently high specific power and low moment of
inertia. The cost is the only drawback. Both Sm and Co are relatively expensive
due to their supply restriction.
With the discovery in recent years of a second generation of rare-earth
magnets on the basis of inexpensive neodymium (Nd) and iron, remarkable progress with regard to lowering raw material cost has been achieved.
NdFeB magnets, which are now produced in increasing quantities, have better
magnetic properties than those of SmCo, but only at room temperature. The
demagnetization curves, especially the coercive force, are strongly temperature dependent. The temperature coefficient of Br is 0.095 − 0.15%/◦ C and
the temperature coefficient of Hc is 0.40 − 0.7%/◦ C. The maximum service
temperature is 150◦ C and the Curie temperature is 310◦ C.
The latest grades of NdFeB have better thermal stability, enabling an
increase in working temperature by 50◦ C, and offer greatly improved
resistance to corrosion.10
6.3.4
Performance Analysis and Control of BLDC Machines
Speed–torque performance is most important for traction and other applications. As any other electric machine, the torque is produced by the interaction
of magnetic field and current. The magnetic field is produced in BLDC by
the PM, and the current depends on the source voltage, control, and the back
EMF, which is determined by the magnetic field and speed of the machine.
209
Electric Propulsion Systems
To obtain the desired torque and speed at a given load, the current needs to
be controlled.
6.3.4.1
Performance Analysis
The performance analysis of the BLDC machines is based on the following
assumption for simplification:
1. The motor is not saturated.
2. Stator resistances of all the windings are equal and self- and mutual
inductances are constant.
3. Power semiconductor devices in the inverter are ideal.
4. Iron losses are negligible.
A simplified equivalent circuit of one phase is shown in Figure 6.48, where
Vt is the voltage of the power supply, Rs is the resistance of the winding, Ls
is the leakage inductance (Ls = L − M, where L is the self-inductance of the
winding and M is the mutual inductance), and Es is the back EMF induced in
the winding by the rotating rotor.
Based on the equivalent circuit of Figure 6.48, the performance of the BLDC
motor can be described by
Vt = Rs Is + Ls
dIs
+ Es ,
dt
(6.109)
Es = kE ωr ,
(6.110)
Te = kT Is ,
(6.111)
Te = TL + J
dωr
+ Bωr ,
dt
(6.112)
where kE is the back EMF constant, which is associated with the PMs and rotor
structure, ωr is the angular velocity of the rotor, kT is the torque constant, TL
Rs
Ls
+
ls
Vt
+
Es
–
–
FIGURE 6.48 Simplified equivalent circuit of the BLDC motor.
210
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Te
wr0
wr
= Vt/kE
FIGURE 6.49 Speed–torque curve in the steady state with constant voltage.
is the load torque, and B is the viscous resistance coefficient. For steady-state
operation, Equations 6.109 through 6.111 can be simply reduced to
Te =
(Vt − kE ωr )kT
.
Rs
(6.113)
The speed–torque performance with constant voltage supply is shown in
Figure 6.49. It can be seen from Equation 6.113 and Figure 6.49 that at low
speed, especially while starting, very high torque is produced, which results
in very high current due to the low back EMF. This very high current would
damage the stator windings.
With variable voltage supply, the winding current can be restricted to its
maximum by actively controlling the voltage; thus a maximum constant
torque can be produced as shown in Figure 6.50.
For dynamic or transient operation, the performance of the BLDC machine
is described by Equations 6.109 through 6.112. However, Laplace transform
is helpful to simplify the analysis. Equations 6.109 through 6.112 can be
expressed by their Laplace forms as
Vt (s) = Es (s) + (Rs + sLs ) Is (s),
(6.114)
Es (s) = kE ωr (s),
(6.115)
Te (s) = kT Is (s),
(6.116)
Te (s) = TL (s) + (B + sJ) ωr (s).
(6.117)
Thus, the transfer function of the BLDC motor drive system is
ωr (s) =
kT
Rs + sLs
Vt (s) −
TL (s).
(Rs + sLs ) (sJ + B) + kT kE
(Rs + sLs ) (sJ + B) + kT kE
(6.118)
211
Electric Propulsion Systems
Te
ls
Vt
Vt-rated
RsIs
wr0
wr
= Vt/kE
FIGURE 6.50 Speed–torque curve in the steady state with variable voltage supply.
Ls and J in Equation 6.118 represent the electrical and mechanical delay in
transient operation. Ls determines how quickly the armature current builds
up in response to a step change in the terminal voltage, where the rotor speed
is assumed to be constant. J determines how quickly the speed builds up in
response to a step change in the terminal voltage.
6.3.4.2
Control of BLDC Motor Drives
In vehicle traction application, the torque produced is required to follow the
torque desired by the driver and commanded through the accelerator and
brake pedals. Thus, torque control is the basic requirement.
Figure 6.51 shows a block diagram of a torque control scheme for a BLDC
motor drive. The desired current I ∗ is derived from the commanded torque
T ∗ through a torque controller. The current controller and commutation
sequencer receive the desired current I ∗ , position information from the position sensors, and perhaps the current feedback through current transducers,
and then produce gating signals. These gating signals are sent to the threephase inverter (power converter) to produce the desired phase current to the
BLDC machine.
In traction application, speed control may be required, cruising control
operation for example (see Figure 6.52). Many high-performance applications include current feedback for torque control. At the minimum, a DC
bus current feedback is required to protect the drive and machine from overcurrents. The controller blocks, “speed controller” may be any type of classical
controller such as a proportional-integral (PI) controller or a more advanced
controller such as an artificial intelligence control. The “current controller and
212
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
DC supply
Commanded
torque T *
I *s
3-phase
inverter
Gating
signals
Current controller and
commutation sequencer
BLDC
motor
Load
Position
sensor
FIGURE 6.51 Block diagram of the torque control of the BLDC motor.
commutation sequencer” provides the properly sequenced gating signals to
the “three-phase inverter” while comparing sensed currents to a reference
to maintain a constant peak current control by hysteresis (current chopping) or with a voltage source (PWM)-type current control. Using position
information, the commutation sequencer causes the inverter to “electronically commutate,” acting as the mechanical commutator of a conventional
DC machine. The commutation angle associated with a brushless motor is
normally set so that the motor will commutate around the peak of the torque
angle curve. Considering a three-phase motor, connected in Delta or wye,
commutation occurs at electrical angles, which are plus or minus 30◦ (electrical) from the peaks of the torque–angle curves. When the motor position
moves beyond the peaks by an amount equal to 30◦ (electrical), then the
commutation sensors cause the stator phase excitation to switch to move
the motor suddenly to −30◦ relative to the peak of the next torque–angle
curve.11
DC supply
w *r
–
+
wr
+
T*
Speed
controller
I *s
3-phase
inverter
Gating
signals
Current controller and
commutation sequencer
FIGURE 6.52 Block diagram of the speed control of the BLDC motor.
Speed
sensor
BLDC
motor
Load
Position
sensor
213
Electric Propulsion Systems
6.3.5
Extend Speed Technology
As discussed above, PM BLDC machines inherently have a short constant
power range due to their rather limited field weakening capability. This is a
result of the presence of the PM field, which can only be weakened through
production of a stator field component, which opposes the rotor magnetic
field. The speed ratio, x, is usually less than 2.12
Recently, the use of additional field windings to extend the speed range of
PM BLDC motors has been developed.1 The key is to control the field current
in such a way that the air gap field provided by PMs can be weakened during
high-speed, constant-power operation. Due to the presence of both PMs and
the field windings, these motors are the so-called PM hybrid motors. The PM
hybrid motor can achieve a speed ratio of around 4. The optimal efficiency
profiles of a PM hybrid motor drive are shown in Figure 6.53.1 However, the
PM hybrid motors have the drawback of a relatively complex structure. The
speed ratio is still not enough to meet the vehicle performance requirement,
especially to an off-road vehicle. Thus a multigear transmission is required.
6.3.6
Sensorless Techniques
As mentioned above, the operation of the BLDC motor drives relies mostly on
position sensors for obtaining the rotor position information so as to properly
perform the turn-on or turn-off of each phase properly.8 The position sensor is
usually either a three-element Hall-effect sensor or an optical encoder. These
position sensors are high-cost, fragile elements. Thus, its presence not only
enhances the cost of the motor drive but also seriously lowers its reliability
and limits its application in some environments, such as the military one.
10
Power (kW)
89 %
88 %
90%
5
92%
88 %
0
2500
5000
Speed (rpm)
7500
FIGURE 6.53 Optimal efficiency profiles of a PM hybrid motor drive.1
10000
214
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Position sensorless technology can effectively continue the operation of the
system in case the position sensors lose their function. This is crucial in some
applications, such as military vehicles.
Several sensorless technologies have been developed. The majority of them
are based on the voltage, current, and back EMF detection. These techniques
can be primarily grouped into four categories:
1. Those using measured currents, voltages, fundamental machine
equations, and algebraic manipulations.
2. Those using observers.
3. Those using back EMF methods.
4. Those with novel techniques not falling into the previous three
categories.
6.3.6.1
Methods Using Measurables and Math
The method consists of two subtypes: (1) those that calculate the flux linkages
using measured voltages and currents and (2) those that utilize a model’s
prediction of a measurable voltage or current, compare the model’s value with
the actual measured voltage or current and calculate the change in position,
which is proportional to the difference between the measured and actual
voltage or current.
The first subtype is seen in some studies.13−20 The fundamental idea is to
calculate the flux linkage from the measured voltage and current
Ψ=
0
t
(V − Ri) dτ.
(6.119)
With the knowledge of initial position, machine parameters, and the flux
linkages’ relationship with rotor position, the rotor position can be estimated.
By determining the rate of change of the flux linkage from the integration
results, the speed can also be determined. An advantage of the flux-calculating
method is that line–line voltages may be used in the calculations and thus no
motor neutral is required.8 This is beneficial, as the most common BLDC
configuration is Y-connected with no neutral.
The second subtype is seen in some other studies.21−24 This method consists
of first developing an accurate d–q model of the machine. Utilizing the measured currents and a d–q transformation, the output voltages of the model are
compared to the measured and transformed voltages. The difference is proportional to the difference in angular reference between the model’s coordinate system and the actual coordinate system, which is the rotor position with
reference to the actual coordinate system’s reference. Conversely, they have
also used the measured voltages and found the differences in the currents. In
either case, the difference between the measured (and transformed) and the
calculated is used as the multiplier in an update equation for the rotor position.
Electric Propulsion Systems
6.3.6.2
215
Methods Using Observers
These methods determine the rotor position and/or speed using observers.
The first of these considered are those utilizing the well-known Kalman filter
as a position estimator.25−30 One of the first of these to appear in the literature was by M. Schroedl in 1988. In his many publications, Schroedl utilized
various methods of measuring system voltages and currents, which could
produce rough estimates of the angular rotor position. The Kalman filtering added the additional refinements to the first estimates of position and
speed. Other observer-based systems include those utilizing nonlinear,31−33
full-order,13,34,35 and sliding-mode observers.15,22,36
6.3.6.3
Methods Using Back EMF Sensing
Using back EMF sensing is the main approach in sensorless control technology of the BLDC motor drive. This approach consists of several methods,
such as (1) the terminal voltage sensing method, (2) the third-harmonic back
EMF sensing method, (3) freewheeling diode conduction, and (4) back EMF
integration.
Terminal voltage sensing: In normal operation of the BLDC motors, the flat
part of a phase back EMF is aligned with the phase current. The switching
instants of the converter can be obtained by knowing the zero crossing of the
back EMF and a speed-dependent period of time delay.37
Since back EMF is zero at rest and proportional to speed, it is not possible
to use the terminal voltage sensing method to obtain a switching pattern at
low speeds. As the speed increases, the average terminal voltage increases
and the frequency of excitation increases. The capacitive reactance in the filters varies with the frequency of excitation, introducing a speed-dependent
delay in switching instants. This speed-dependent reactance disturbs current
alignment with the back EMF and field orientation, which causes problems at
higher speeds. In this method, a reduced speed operating range is normally
used, typically around 1000–6000 rpm. This method is a good method for
the steady state; however, phase differences in the circuits used due to speed
variations do not allow optimal torque per ampere over a wide speed range.
Third harmonic back EMF sensing: Rather than using the fundamentals of the
phase back EMF waveform as in the previous technique, the third harmonic
of the back EMF can be used in the determination of the switching instants
in the wye-connected 120◦ current conduction operating mode of the BLDC
motor.38 This method is not as sensitive to phase delay as the zero voltage
crossing method, since the frequency to be filtered is three times as high. The
reactance of the filter capacitor in this case dominates the phase angle output
of the filter more so than at the lower frequency. This method provides a wider
speed range than the zero-crossing method, does not introduce as much phase
delay as the zero-crossing method, and requires less filtering.
216
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Freewheeling diode conduction: This method uses indirect sensing of the
zero crossing of the phase back EMF to obtain the switching instants of
the BLDC motor.39 In the 120◦ conducting Y-connected BLDC motor, one
of the phases is always open-circuited. For a short period after opening the
phase, there-phase current remains flowing, via a free-wheeling diode, due
to the inductance of the windings. This open-phase current becomes zero
in the middle of the commutation interval, which corresponds to the point
where back EMF of the open phase crosses zero. The largest downfall of this
method is the requirement of six additional isolated power supplies for the
comparator circuitry for each free-wheeling diode.
Back EMF integration: In this method, position information is extracted by
integrating the back EMF of the unexcited phase.40−43 The integration is
based on the absolute value of the open phase’s back EMF. Integration of
the voltage divider scaled-down back EMF starts when the open phase’s
back EMF crosses zero. A threshold is set to stop the integration that corresponds to a commutation instant. As the back EMF is assumed to vary
linearly from positive to negative (trapezoidal back EMF assumed), and this
linear slope is assumed to be speed insensitive, the threshold voltage is kept
constant throughout the speed range. If desired, current advance can be implemented by the change of the threshold. Once the integrated value reaches the
threshold voltage, a reset signal is asserted to zero the integrator output.
This approach is less sensitive to switching noise, automatically adjusts
to speed changes, but the low-speed operation is poor. With this type of
sensorless operation scheme, up to 3600 rpm has been reported.43
6.3.6.4
Unique Sensorless Techniques
The following sensorless methods are completely original and unique. These
range from artificial intelligence methods to variations in the machine structure. The first of the novel methods to be considered are those utilizing artificial intelligence, that is, artificial neural networks (ANNs) and fuzzy logic.
Peters and Harth43 utilized a neural network using the back-propagation
training algorithm (BPN) to act as a nonlinear function implementation
between measured phase voltages and currents, which were inputs, and rotor
position, which was the output. Using the equations in the above method,
the flux linkage can be calculated using the measured voltages, currents, and
system parameters.
Utilizing fuzzy logic, Hamdi and Ghribi44 proposed two fuzzy logic subsystems in an application. Using the conventional equations of phase voltages
and currents, the rotor position can be calculated.8 With the knowledge of
the relationships between these measurables and the rotor position, a fuzzy
Mamdami-type system was developed to produce rotor position estimates.
It was noted that this could have been accomplished just as easily with
look-up tables; however, for the desired resolution the size of the look-up
tables becomes unmanageably large. The second fuzzy system used took as
Electric Propulsion Systems
217
input the estimated rotor position and produced reference current values for
two different drive strategies: unity power factor and maximum torque per
ampere.
In Hesmondhalgh et al.,45 an additional stator lamination with equally
spaced slots around the periphery is added to the end of the machine. Each
of the slots contains a small sensing coil. The local magnetic circuit variations
for each of the sensing coils are affected by the PM rotor’s position. A 20-kHz
signal is injected through the coils. The signal distortions are analyzed at the
terminals of the sensing coils, the second harmonic yielding position information. An artificial saliency was created in Hesmondhalgh et al.45 by attaching
small pieces of aluminum to the surface of the PMs. The flow of eddy currents
in the aluminum acts to increase the reluctance of the various windings’ magnetic circuits, thus causing changes in the winding’s inductances with rotor
position.
6.4 SRM Drives
The SRM drive is considered as an attractive candidate for variable speed
motor drives due to its low cost, rugged structure, reliable converter topology, high efficiency over a wide speed range, and simplicity in control.46,47
These drives are suitable for EVs, HEV traction applications, aircraft
starter/generator systems, mining drives, washing machines, door actuators,
and so on.48–51
The SRM has a simple, rugged, and low-cost structure. It has no PM or winding on the rotor. This structure not only reduces the cost of the SRM, but also
offers high-speed operation capability for this motor. Unlike the induction and
PM machines, the SRM is capable of high-speed operation without the concern
of mechanical failures that result from the high-level centrifugal force. In addition, the inverter of the SRM drive has a reliable topology. The stator windings
are connected in series with the upper and lower switches of the inverter. This
topology can prevent the shoot-through fault that exists in the induction and
permanent motor drive inverter. Moreover, high efficiency over a wide speed
range and control simplicity are known merits of the SRM drive.46,47
A conventional SRM drive system consists of the SRM, power inverter,
sensors such as voltage, current, and position sensors, and control circuitry
such as the DSP controller and its peripherals as shown in Figure 6.54.
Through proper control, high performance can be achieved in the SRM drive
system.46,47 The SRM drive inverter is connected to a DC power supply, which
can be derived from the utility lines through a front-end diode rectifier or
from batteries. The phase windings of the SRM are connected to the power
inverter, as shown in Figure 6.55. The control circuit provides a gating signal
to the switches of the inverter according to particular control strategies and
the signals from various sensors.
218
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Electric
energy
input
Power
converter
Load
SRM
Current
sensor
Control
commands
Position
sensor
Controller
SRD
FIGURE 6.54 SRM drive system.
6.4.1
Basic Magnetic Structure
The SRM has salient poles on both the stator and rotor. It has concentrated
windings on the stator and no winding or PM on the rotor. There are several
configurations for SRM depending on the number and size of the rotor and
stator poles. The configurations of the 8/6 and 6/4 SRMs, which are more
common, are shown in Figure 6.56.
Due to its double saliency structure, the reluctance of the flux path for a
phase winding varies with the rotor position. Also, since the SRM is commonly
designed for high-degree saturation at high phase current, the reluctance of
the flux path also varies as the phase current. As a result, the stator flux linkage,
phase bulk inductance, and phase incremental inductance all vary with the
rotor position and phase current.
The phase voltage equation of the SRM (Figure 6.55) is given by
d λjk ,
dt
m
Vj = Rij +
(6.120)
k=1
a
d'
2
3'
SA1
b
SB1
DB2
SC1
DC2
SD1
DD2
SA2
DB1
SB2
DC1
SC2
DD1
SD2
º
DA2
c'
2'
2
1'
3
b'
c
d
a'
FIGURE 6.55 SRM and its power supply.
DA1
219
Electric Propulsion Systems
Stator
pole
(a)
(b)
a
Rotor
a
b
d′
b
c′
Rotor
pole
c′
c
c
b′
b′
a′
Stator
winding
d
a′
Stator
FIGURE 6.56 Cross section of common SRM configurations: (a) a 6/4 SRM and (b) an
8/6 SRM.
where m is the total number of phases, Vj is the applied voltage to phase j, ij
is the current in phase j, R is the winding resistance per phase, λjk is the flux
linkage of phase j due to the current of phase k, and t is the time. The phase
flux linkage λjk is given by
λjk = Ljk (ik,θ , θ)ik ,
(6.121)
where Ljk is the mutual inductance between phase k and phase j. The mutual
inductance between phases is usually small compared to the bulk inductance
and is neglected in equations.
At a fixed phase current, as the rotor moves from the unaligned to the
aligned position, the reluctance of the flux path reduces due to the reduction
in the air gap. As a result, the phase inductance and flux linkage increases
as the rotor moves. At a fixed rotor position, as the phase current increases,
the flux path becomes more and more saturated. Hence, the reluctance of
the flux path reduces as the phase current increases. As a result, the phase
bulk inductance drops with an increase in the phase current. But the phase
flux linkage still increases as the phase current increases due to the enhancement in the excitation. The variations of the phase bulk inductance and flux
linkage with respect to the phase current and rotor position for an 8/6 SRM
are shown in Figures 6.57 and 6.58, respectively. In those figures, θ = −30◦
and θ = 0◦ represent the unaligned and aligned rotor positions of the referred
SRM, respectively.
220
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Aligned
position
q = 0∞
q = –3∞
q = –6∞
q = –9∞
q = –12∞
0.35
Flux linkage (Wb)
0.3
0.25
0.2
q = –15∞
0.15
q = –18∞
q = –21∞
q = –24∞
q = –27∞
q = –30∞
Unaligned
position
0.1
0.05
0.0
0
1
2
3
4
5
Phase current (A)
6
7
FIGURE 6.57 Variation of phase flux linkage with rotor position and phase current.
Substituting Equation 6.121 into Equation 6.120, one can have
m
m ∂λjk dik
∂λjk dθ
d λjk = Rij +
+
dt
∂ik dt
∂θ dt
k=1
k=1
m
∂(Ljk ik ) dik
∂(Ljk ik )
+
ω
= Rij +
∂ik dt
∂θ
k=1
m ∂Ljk dik
∂Ljk
Ljk + ik
+ ik
ω .
= Rij +
∂ik
dt
∂θ
Vj = Rij +
(6.122)
k=1
Phase, inductance (H)
0.08
0.07
2A
0.06
4A
0.05
6A
0.04
0.03
0.02
0.01
–30
–20
Unaligned
position
–10
0
Aligned
position
10
20
30
Unaligned
position
Rotor angular position (mechanical degree)
FIGURE 6.58 Variation of phase bulk inductance with rotor position and phase current.
221
Electric Propulsion Systems
When only one phase is energized in the operation, Equation 6.122 can be
written as
∂Ljj dij
∂Ljj
Vj = Rij + Ljj + ij
+ ij
ω.
(6.123)
∂ij dt
∂θ
The third term on the right-hand side of Equation 6.123 is the back EMF. The
phase incremental inductance is defined as the derivative of the phase flux
linkage against the phase current as
jj =
∂λjj
∂Ljj
= Ljj + ij
,
ij
∂ij
(6.124)
where jj (i, θ) and Ljj (i, θ) are phase incremental inductance and bulk inductance, respectively. Figure 6.57 shows a typical example of flux linkage varying
with rotor position, θ, and phase current i of an SRM. Figure 6.58 shows typical
variation of phase bulk inductance with rotor position and phase current.
When the magnetic flux is not saturated, the flux linkage varies linearly with
the phase current. The incremental inductance can be viewed as equal to the
phase bulk inductance. However, if the machine is saturated at a certain phase
current and rotor position, the phase incremental inductance does not equal
the phase bulk inductance any more. The variation of the phase incremental
inductance with respect to the phase current and rotor position can be derived
from the variation of the phase linkage with respect to the phase current and
rotor position. The variation of the phase incremental inductance with respect
to the phase current and rotor position for an 8/6 SRM is shown in Figure 6.59.
0.08
2A
Incremental inductance (H)
0.07
0.06
0.05
0.04
4A
0.03
0.02
0.01
–30
6A
–20
–10
0
10
20
30
Rotor angular position (mechanical degree)
FIGURE 6.59 Variation of phase incremental inductance with rotor position and phase current
for a typical 8/6 SRM.
222
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
6.4.2 Torque Production
Torque in SRM is produced by the tendency of the rotor to get into alignment
with the excited stator poles. The analytical expression of the torque can be
derived using the derivative of the coenergy against the rotor position at a
given current.
For a phase coil with current i linking a flux λ, the stored field energy Wf
and the coenergy Wf are indicated as shaded regions as shown in Figure 6.60.
Coenergy can be found from the definite integral:
Wf =
i
0
λ di.
(6.125)
The torque produced by one phase coil at any rotor position is given by
∂Wf
T=
∂θ
.
(6.126)
i=constant
In the case of flux being linear with current, for example, unsaturated field, the
magnetization curve in Figure 6.60 would be a straight line and the coenergy
would be equal to the stored field energy. The instantaneous torque can be
given as
1 dL(θ)
,
(6.127)
T = i2
2
dθ
where L is the unsaturated phase bulk inductance.
Flux
Stored energy, Wf
Coenergy, Wf'
l
di
FIGURE 6.60 Stored field energy and coenergy.
Current, i
223
Electric Propulsion Systems
In the case of saturated phase, the torque cannot be calculated by a simple
algebra equation; instead, an integral equation such as
i
T=
∂L(θ, i)
i di
∂θ
(6.128)
0
is used.
From Equations 6.127 and 6.128, it can be seen that in order to produce positive torque (motoring torque) in SRM, the phase has to be excited when the
phase bulk inductance increases as the rotor rotates. It can also be observed
from Equation 6.127 and 6.128 that the phase current can be unidirectional for
motoring torque production. Hence, low-cost and reliable inverter topology
introduced in the later section can be employed for the SRM drive. Figure
6.61 shows the ideal phase inductance, current, and torque of the SRM. Positive (motoring) torque is produced if the phase is excited when the phase
inductance is increasing as the rotor rotates. Negative torque is generated
if the phase is excited when the phase inductance is decreasing as the rotor
moves.52,53 This implies that the position information is necessary for control
of the SRM drive.
The output torque of an SRM is the summation of torque of all the phases:
Tm =
N
T(i, θ),
i=1
Ideal phase
inductance
Ideal phase current
(motoring operation)
Ideal torque
(motoring operation)
Ideal phase current
(regeneration operation)
Ideal torque
(regeneration operation)
FIGURE 6.61 Idealized inductance, current, and torque profiles of the SRM.
(6.129)
224
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
where Tm and N are the output torque and phase number of the motor. The
relation between the motor torque and mechanical load is usually given by
Tm − Tl = J
dω
+ Bω,
dt
(6.130)
where J, B, and Tl are the moment of inertia, viscous friction, and load torque,
respectively. The relation between position and speed is given by
ω=
6.4.3
dθ
.
dt
(6.131)
SRM Drive Converter
It can be seen from Figure 6.61 that the torque developed by the motor can
be controlled by varying the amplitude and the timing of the current pulses
in synchronism with the rotor position. In order to control the amplitude and
pulse width of the phase current, a certain type of inverter should be used.
The input to the SRM drive is DC voltage, which is usually derived from the
utility through a front-end diode rectifier or from batteries. Unlike other AC
machines, the currents in SRMs can be unidirectional. Hence, conventional
bridge inverters used in AC motor drives are not used in SRM drives. Several
configurations have been proposed for an SRM inverter in the literature,54,55
some of the most commonly used ones are shown in Figure 6.62.
The most commonly used inverter uses two switches and two freewheeling
diodes per phase and is called the classic converter. The configuration of
the classic converter is shown in Figure 6.62a. The main advantage of the
classic converter is the flexibility in control. All the phases can be controlled
independently, which is essential for very high-speed operation where there
will be considerable overlap between the adjacent phase currents.56
The operation of the classic converter is shown in Figure 6.63 by taking
phase-1 as an example. When the two switches S1 and S2 are turned on as in
Figure 6.62a, the DC bus voltage, Vdc , will be applied to the phase-1 winding.
Phase-1 current will increase as it flows through the path consisting of the
Vdc positive terminal, S1 , phase-1 winding, S2 , and the Vdc negative terminal.
By turning off S1 and holding on S2 (i.e., Figure 6.63b), when the phase is
energized, the current freewheels through the S2 and D1 . In this mode, phase1 is not getting or giving energy to the power supply. When S1 and S2 are
turned off (Figure 6.63c), the phase-1 current will flow through D2 , the Vdc
positive terminal, the Vdc negative terminal, D1 , and phase-1 winding. During
this time, the motor phase is subjected to negative DC bus voltage through the
freewheeling diodes. The energy trapped in the magnetic circuit is returned
to the DC link. The phase current drops due to the negative applied phase
voltage. By turning on and off S1 and S2 , the phase-1 current can be regulated.
The half-bridge converter uses 2n switches and 2n diodes for an n-phase
machine. There are several configurations that use less switches; for example,
225
Electric Propulsion Systems
(a)
+
S1
D4
D3
D6
D5
S4
S2
Phase-4
D2
S9
Phase-3
D1
Phase-2
Phase-1
Vdc
S5
S3
D8
D7
S6
S8
–
(b)
(c)
+
+
S2
C
S4
S3
D1
D2
D3
S2
D4
S4
S3
–
(e)
+
Vdc
S1
–
S6
D1
S2
Phase-4
S5
Phase-3
D6
D4
Phase-2
S3
D5
Phase-1
S2
D3
+
Phase-4
D1
Lc
Phase-2
Phase-3
Phase-1
D2
Sc
S4
S1
–
S1
D5
(d)
Vdc
Phase-4
S1
–
Vdc
D4
Phase-3
Phase-4
D3
Phase-2
Phase-3
D2
Phase-1
Phase-2
D1
Phase-1
Vdc
S5
R
D2
D3
S3
C
D4
Dc
S4
FIGURE 6.62 Different inverter topologies for SRM drives: (a) classic half-bridge converter;
(b) R-dump; (c) n + 1 switches (Miller converter); (d) 1.5n switch converter; and (e) C-dump.
the R-dump-type inverter (Figure 6.62b) uses one switch and one diode per
phase. This drive is not efficient; during turn-off, the stored energy of the
phase is charging capacitor C to the bus voltage and dissipating in resistor R.
Also zero voltage mode does not exist in this configuration.
(a) l1
+
l1
(b)
+
S1
l1
D2
Vdc
(c)
+
S1
D2
Vdc
l1
D1
–
S2
l1
D1
–
l1
l1
S1
l1
D1
l1
D2
Vdc
l1
l1
l1
l1
S2
l1
l1
S1
–
l1
FIGURE 6.63 Modes of operation for the classic converter: (a) turn-on mode, (b) zero voltage
mode, and (c) turn-off mode.
226
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
An alternative configuration is (n + 1) switch inverter. In this inverter, all
the phases are sharing a switch and diode so that overlapping operation
between phases is not possible, which is inevitable in high-speed operation
of this motor. This problem has been solved by sharing switches of each couple
nonadjacent phases, as shown in Figure 6.62d. This configuration is limited
to an even number of phases of SRM drives.
One of the popular inverter configurations is C-dump (Figure 6.62e), which
has the advantage of less switches and allows independent phase current
control. In this configuration, during the turn-off time, the stored magnetic
energy is charging capacitor C, and if the voltage of the capacitor reaches a
certain value, for example Vc , it is transferred to the supply through the switch
Sc . The main disadvantage of this configuration is that the negative voltage
across the phase coil is limited to the difference between the voltage across
the capacitor Vc and the system power supply voltage.
6.4.4
Modes of Operation
For SRM, there is a speed at which the back EMF is equal to the DC bus voltage.
This speed is defined as the base speed. Below the base speed, the back EMF
is lower than the DC bus voltage. From Equation 6.125, it can be seen that
when the converter switches are turned on or off to energize or de-energize
the phase, the phase current will rise or drop accordingly. The phase current
amplitude can be regulated from 0 to the rated value by turning on or off the
switches. Maximum torque is available in this case when the phase is turned
on at an unaligned position and turned off at the aligned position and the
phase current is regulated at the rated value by hysteresis or PWM control.
The typical waveforms of the phase current, voltage, and flux linkage of the
SRM below the base speed are shown in Figure 6.64.
Above the base speed, the back EMF is higher than the DC bus voltage. At the rotor position—at which the phase has a positive inductance
slope with respect to the rotor position—the phase current may drop even
if the switches of the power inverter are turned on. The phase current is
limited by the back EMF. In order to build high current and therefore produce high motoring torque in the SRM, the phase is usually excited ahead
of the unaligned position, and the turn-on position is gradually advanced as
the rotor speed increases. The back EMF increases with the rotor speed. This
leads to a decrease in the phase current and hence the torque drops. If the
turn-on position is advanced for building as high as possible a current in the
SRM phase, the maximum SRM torque almost drops as a linear function of
the reciprocal of the rotor speed. The maximum power of the SRM drive is
almost constant. The typical waveforms at high-speed operation are shown
in Figure 6.65.
The advancing of the phase turn-on position is limited to the position at
which the phase inductance has a negative slope with respect to the rotor position. If the speed of the rotor further increases, no phase advancing is available
227
Electric Propulsion Systems
Ideal phase
inductance
Vbus
Phase voltage
0
–Vbus
Imax
Imin
0
Phase current
FIGURE 6.64 Low-speed (below the base speed) operation of SRM.
for building higher current in the phase and the torque of the SRM will drop
significantly.11 The mode is referred to as the natural mode operation. The
torque–speed characteristic of the SRM is shown in Figure 6.66.
6.4.5
Generating Mode of Operation (Regenerative Braking)
Torque in the SRMs is created based on the principle of reaching the minimum
reluctance for the excited phase. Therefore, if the rotor pole is approaching
the excited phase, which means the bulk inductance is increasing, the torque
produced is in the direction of the rotor and it is in motoring mode. But if the
Ideal phase
inductance
Vbus
Phase voltage
0
–Vbus
Imax
Phase current
0
FIGURE 6.65 High-speed (above the base speed) operation of SRM.
228
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
T
Constant
torque
0
Constant
power
Base
speed
Natural
w
FIGURE 6.66 Torque–speed characteristic of SRM.
rotor pole is leaving the stator phase—which means the negative slope of the
bulk inductance—the stator tries to keep it in alignment; the torque produced
is then in the opposite direction of the movement of the rotor, and the SRM
works in the generating mode.
Regenerative braking is an important issue in the propulsion drive of
EVs and HEVs. There is a duality in operation of generating and motoring
modes and the current waveforms in the generating modes are simply the
mirror images of the waveforms in the motoring region around the alignment rotor position.57 The switched reluctance generator (SRG) is a singly
excited machine, so in order to get power from it, it should be excited near
the rotor aligned position and then turned off before the unaligned region
(Figure 6.67).
As in motoring operation, current can be controlled by changing the turnon and turn-off angles and current level while at low speed. Alternatively, at
speeds higher than the base speed, only the turn-on and turn-off angles can
be used for control.
The driving circuit for SRG is similar to the SRMs; one of the common
configurations is shown in Figure 6.68. When the switches are turned on, the
phase gets energy from the supply and capacitor. During the turn-off period,
the freewheeling current from the motor charges up the capacitor and delivers
energy to the load. Since there is no PM in this motor, during the start-up and
initial condition, it needs an external source such as a battery to deliver energy
to the phase; after taking transient time, the capacitor is then charged up to the
output voltage. Depending on the output voltage during phase-on time, both
the capacitor and external source, or just the capacitor, provide the current to
229
Electric Propulsion Systems
(a)
Ideal phase
inductance
Vbus
Phase
voltage
Phase
current
0
–Vbus
Imax
Imin
0
(b)
Ideal phase
inductance
Vbus
Phase
voltage
Phase
current
0
–Vbus
Imax
0
FIGURE 6.67 Low- and high-speed operation in generating mode: (a) low speed operation and
(b) high speed operation.
the load and the phase coil. The external source can be designed to be charged,
or can be disconnected from the system after the system reaches its operating
point.
In the generating region, the back EMF is negative so it helps the phase to
be charged very fast; then during turn-off the back EMF opposes the negative
Is
IL
S5
S9
Ic
S3
S1
D9
S2
D3
S4
D5
S6
FIGURE 6.68 A driving circuit example for the SRG.
D6
Phase-4
D1
D4
Phase-3
D2
Phase-2
Phase-1
Vs
D7
S8
C
D8
ZL
230
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
50
45
40
Turn off = 16
Phase current (A)
35
30
25
Turn off = 13
20
Turn off = 12
15
10
5
0
0.05 0.051 0.052 0.053 0.054 0.055 0.056 0.057 0.058 0.059 0.06
Time (s)
FIGURE 6.69 Effect of the turn-off angle in maximum current level in generating mode in an
8/6 switched reluctance machine.
supply voltage, and it decreases slowly.
di
+ Ri,
dt
di
−VC − e = L
+ Ri,
dt
VC − e = L
e > 0 (during phase-on period).
(6.132)
e > 0 (during phase-off period).
(6.133)
In Equations 6.132 and 6.133, Vc is the bus voltage of the inverter or,
equivalently, the voltage of the bus capacitor and e is the back EMF voltage.
In certain conditions such as high speed and high loads, the back EMF voltage is greater than the bus supply voltage, so the current increases even after
turning off the phase. In addition to uncontrollable torque, this necessitates
an oversized converter, thereby adding to the cost and overall size of the
system. Due to variations in speed of the prime mover, the power electronic
converter should be designed for the worst possible case. This will magnify
the additional cost and size issues. By properly selecting the turn-off angle,
this maximum generating current can be coaxed into the safe region.58 Figure
6.69 shows the effect of turn-off angle in the maximum generating current.
6.4.6
Sensorless Control
Excitation of the SRM phases needs to be properly synchronized with the
rotor position for effective control of speed, torque, and torque pulsation.
Electric Propulsion Systems
231
A shaft position sensor is usually used to provide the rotor position. However, these discrete position sensors not only add complexity and cost to the
system but also tend to reduce the reliability of the drive system and restrict
their application in some specific environments, such as military applications.
Position sensorless technology can effectively continue the operation of the
system, in case the position sensors lose their function. This is crucial in some
applications, such as military vehicles.
Several sensorless control methods have been reported in the literature over
the past two decades.15−28 Most of these techniques are based on the fact that
the magnetic status of the SRM is a function of the angular rotor position.
As the rotor moves from the unaligned position toward the aligned position,
the phase inductance increases from the minimum value to the maximum
value. It is obvious that if the phase bulk inductance can be measured and the
functional relation between the phase bulk inductance and the rotor position
is known, the rotor position can be estimated according to the measured phase
bulk inductance.59
Some sensorless techniques do not use the magnetic characteristic and voltage equation of the SRM directly to sense the rotor position. Instead, these
sensorless control methods are based on the observer theory or synchronous
operation method similar to that applied to conventional AC synchronous
machines.
Generally, the existing sensorless control methods can be classified as
follows:
1. The phase flux linkage-based method.
2. The phase inductance-based method.
3. Modulated signal injected methods.
4. The mutual-induced voltage-based method.
5. Observer-based methods.
6.4.6.1
Phase Flux Linkage-Based Method 60
This method uses the phase voltage and current data of the active phases to
estimate the rotor position. The basic principle of this method is to use the
functional relation between the phase flux linkage, phase current, and rotor
position for rotor position detection. From Figure 6.57, it can be observed that
if the flux linkage and phase current are known, the rotor position can be
estimated accordingly, as shown in Figure 6.70.
The problem with this sensorless control method is the inaccurate estimation of the phase flux linkage at low speed. At high speed (above the base
speed), the phase voltage keeps its positive polarity until the phase is turned
off. The V term dominates in V–Ri, and integration of V–Ri in a relatively short
period will not lead to a huge error in flux estimation. However, at low speed
232
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
V, i
ljj = Ú(Vj – Rij)dt
i
ljj,i
q
l
q
FIGURE 6.70 Flux linkage-based rotor position estimation method.
(below the base speed) the phase voltage changes its polarity from one hysteresis cycle to the next hysteresis cycle. When V–Ri is integrated in a relatively
long period, the phase voltage term cancels itself due to the excursions, while
the Ri term keeps its polarity during the integration period—and becomes
significant after a long time of integration. The error in R or i may lead to a
huge error in the flux estimation in this case. Therefore, this sensorless control
method is only suitable for high-speed operation of SRM.
6.4.6.2
Phase Inductance-Based Method
Similar to the phase flux linkage, the phase bulk and incremental inductances
are both functions of the phase current and rotor position. Hence, they can
also be used for rotor position estimation.
6.4.6.2.1
Sensorless Control Based on Phase Bulk Inductance61
Using the phase flux linkage obtained as shown in Figure 6.70, the phase
inductance can be obtained as
Ljj =
λjj
.
ij
(6.134)
The estimated phase bulk inductance and measured phase current can be
input to a prestored look-up table storing the functional relation between
the phase bulk inductance and phase current and rotor position, to find the
corresponding rotor position. Instead of using a look-up table, one can also
use an analytical model to represent the functional relation between phase
bulk inductance, phase current, and rotor position.61
Like the flux linkage-based method, since integration of V–Ri is used for
phase inductance estimation, this method is only suitable for high-speed operation. Some sensorless control method that can work both at standstill and
low speed, such as the open-loop method, has to be used to start the SRM
233
Electric Propulsion Systems
and bring the rotor speed to a certain level. After the rotor speed has reached
a threshold, the phase flux linkage and/or inductance are calculated using
the integration method and the rotor position is estimated according to the
calculated phase flux linkage and inductance.
6.4.6.2.2 Sensorless Control Based on Phase Incremental Inductance
The position estimation method using the phase incremental inductance utilizes the current and voltage data of the active phase for estimation of the
incremental inductance of this phase, and consequently, of the rotor position.
Neglecting the mutual couplings and, at very low speeds [neglectable
motional-EMF term ij (∂Ljj /∂θ)ω], the incremental inductance can be obtained
from Equations 6.123 and 6.124 as
jj =
Vj − Rij
.
dij /dt
(6.135)
Thus, the phase incremental inductance can be measured from the phase
voltage and current. If the relation between the phase incremental inductance
and rotor position is known, the rotor position can be estimated according to
the estimated phase incremental inductance.
At low phase current—and therefore at unsaturated phase—the phase
incremental inductance can be viewed as equal to the phase bulk inductance,
and monotonically increases as the rotor moves from the unaligned position
to the aligned position. The incremental inductance has a one-to-one relation
with the rotor position in this case. However, at high phase current, and therefore at saturated phase, when the rotor moves from the unaligned position
to the aligned position, the phase incremental inductance may be the same
value at two or more rotor positions.62
Even though the phase incremental inductance does not have a one-to-one
relation with the rotor position at high phase current, it can still be used
for rotor position estimation. Some SRMs are designed with a high degree
saturation such that the phase incremental inductance at the aligned position
has the minimum value at high current. In this case, the phase incremental
inductance at the aligned position is unique; hence, it can be used to detect
the aligned rotor position. This rotor position estimation technique will give
one rotor position at one electrical cycle.
This method does not require any extra sensing circuitry. However, it is
applicable only for very low speeds, less than 10% of the base speed, because
the back EMF term is neglected for calculation of the phase incremental
inductance.
6.4.6.3
Modulated Signal Injection Methods
These methods are to apply a voltage to the idle phase winding and measure
the resultant phase current to detect the phase inductance. This derived phase
234
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
inductance will provide the rotor position information. Both an extra-low
amplitude voltage source and the power converter can be used to apply a voltage to the phase winding. When an extra voltage source is used, a sinusoidal
voltage is usually used for sensing the phase inductance. The phase angle
and the amplitude of the resultant phase current contain the phase inductance; hence the rotor position information can be obtained. This is the idea
behind the amplitude modulation (AM) and phase modulation (PM) methods. When the power converter is used for sensing purpose, a short period
voltage pulse is usually applied to the idle phase and a triangular current is
induced in the corresponding phase. The changing rate of the phase current
contains the phase inductance, and hence the rotor position information. This
is the basic idea of the diagnostic pulse-based method.
6.4.6.3.1
Frequency Modulation Method49,63
This method is used to first generate a train of square wave voltage whose
frequency is in reverse proportion to the instantaneous inductance of the idle
phase. The circuitry used for generating a square wave voltage train whose
frequency is in reverse proportion to the inductance is referred to as an L–F
converter.
To detect the frequency of this square wave voltage train, and hence the
phase inductance, the timer of a microcontroller can be used to count the
frequency of the square wave voltage train. Another approach is to use a
frequency-to-voltage converter (F–V converter) to obtain a voltage proportional to the frequency of the square wave voltage train, and sample this
voltage using an A/D converter. In order to connect the phase winding, which
is in the power circuit, to the sensing circuitry, which is in the control circuit,
two photovoltaic BOSFET switches are used for each phase.
Since the signal used for position estimation in this method is an inductanceencoded frequency signal, this method is referred to as the frequency
modulation (FM) method. This method is easy to implement and is robust.
However, at high-speed operation, the phase energizing current exists—even
when the phase inductance is decreasing as the rotor moves. This restricts
the signal injection to the SRM phases. Another problem with this method is
that it requires additional circuitry for implementation. The cost associated
with this additional circuitry may be a concern in some applications. Furthermore, it is very sensitive to mutual coupling since the current in the active
phase induces voltage in the unenergized phases, which strongly distorts the
probing pulses.
6.4.6.3.2
AM and PM Methods64
The PM and AM techniques are based on the phase and amplitude variations,
respectively, of the phase current due to the time-varying inductance when a
sinusoidal voltage is applied to the phase winding in series with a resistance
R. The current flowing through the circuit in response to the applied voltage
235
Electric Propulsion Systems
is a function of the circuit impedance. Since the coil inductance varies periodically, the phase angle between the current and the applied voltage also varies
in a periodic manner. With a large inductance, the lagging angle of the current wave behind the voltage wave is large and the peak current is small. The
PM encoder technique measures the instantaneous phase angle on a continuous basis, while the AM encoder technique measures the amplitude. These
instantaneous measurements contain the phase inductance information that
can be obtained after passing the signals through a demodulator. The demodulator generates a signal that represents the phase inductance as a function of
the rotor position. Using an inverse function or a conversion table, the rotor
position can be estimated.
Since the PM and AM methods need the injection of a low-amplitude signal
to one of the idle phases, the photovoltaic BOSFET switches are needed to
connect the phase winding to the sensing circuitry.
Like the FM method, signal injection to one of the idle phases is restricted
at high-speed operation where the torque producing current occupies most
of the electrical cycle and makes the signal injection impossible. Another disadvantage of these methods is that they require additional hardware for indirect position sensing. As stated before, it is very sensitive to mutual coupling.
6.4.6.3.3 Diagnostic Pulse-Based Method65
Instead of using an additional voltage source to inject the sensing signal to
the idle phase, the power converter of the SRM drive can be used to provide
a short period voltage pulse to the idle phase, and low-amplitude current is
produced. Therefore, the back EMF, saturation effect, and the voltage drop on
the winding resistance can all be neglected. From Equations 6.123 and 6.124,
the changing rate of the phase current is given as
dij
V
.
=
dt
Ljj
(6.136)
Equation 6.136 indicates that the phase current changing rate contains the
phase inductance, and hence the rotor position information.
Similar to the case of switches being turned on, when the switches connected
to the phase are turned off, the phase current freewheels through the diodes.
The phase voltage equals the negative DC bus voltage; the change rate of the
current has the same expression as Equation 6.136, but with a negative sign.
Either the current growing rate or dropping rate can be used for sensing the
phase inductance. When the current changing rate is found to exceed a threshold that is dictated by the phase inductance at the commutation position, the
phase can be commutated. This method does not require additional hardware for indirect rotor position sensing. However, at high-speed operation,
the phase excitation current occupies the majority of one electrical cycle and
restricts the injection of the testing signal. And like FM, AM, and PM methods,
it is very sensitive to mutual coupling.
236
6.4.6.4
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Mutual-Induced Voltage-Based Method 66
The idea of this method is based on measuring the mutually induced voltage
in an idle phase, which is either adjacent or opposite to the energized phase
of an SRM. The mutual voltage in the “off” phase, induced due to the current
in the active phase, varies significantly with respect to the rotor position.
This mutually induced voltage variation can be sensed by a simple electronic
circuit. If the functional relation between the mutually induced voltage in the
inactive phase due to the current in the active phase and the rotor position
is known, the rotor position information can be extracted from the mutually
measured induced voltage in the inactive phase. This method is only suitable
for low-speed operation. Furthermore, it is very sensitive to noise since the
ratio between induced voltage and system noise is small.
6.4.6.5
Observer-Based Methods 67
In this method, state-space equations are used to describe the dynamic behavior of the SRM drive. An observer is then developed based on these nonlinear
state-space differential equations for estimation of the rotor position. The
input and output of this observer are phase voltage and phase current, respectively. The state variables of this observer are stator flux linkage, rotor position
angle, and rotor speed. The phase current, flux linkage, rotor position, and
rotor speed can be estimated using this observer. The phase current estimated
by this observer is compared to the actual phase current of the SRM, and the
resultant current errors are used to adjust the parameters of the observer.
When the current estimated by the observer matches the actual current, the
observer is considered as a correct representation of the dynamic behavior of
the actual SRM drive and the rotor position estimated by the observer is used
to represent the actual rotor position.
The main disadvantages of these methods are real-time implementation
of complex algorithms, which require a high-speed DSP and a significant
amount of stored data. This increases the cost and speed limitations by the
DSP. However, high resolution in detecting rotor position and applicability
to the whole speed range are some merits of these methods.
6.4.7
Self-Tuning Techniques of SRM Drives
As discussed in previous sections, the SRM drive has a simple and rugged
construction—favorable characteristics for traction application. But its control is very complicated due to the nonlinearity of its magnetic circuit and the
fact that control depends heavily on the mechanical and electrical parameters, such as air gap, resistance, and so on.53 In mass production and
real-world operation, it is important for these parameters to have the exact
values and remain unchanged. For example, the air gap would be changed
due to mechanical vibration wearing, and the resistance in windings and
237
Electric Propulsion Systems
inductance would vary with temperature. These parameter variations would
cause significant degradation of the drive performance if the control system
cannot “know” these variations and implement corresponding corrections
in the control process. Self-tuning techniques are referred to as methods of
updating the control strategy in a control system.
The major purpose of self-tuning control for the SRM drive is to update
the control variables in the presence of motor parameters’ variations, so as to
optimize the torque per ampere.68 There are two approaches to this problem:
the arithmetic mean method and the neural network-based method.
6.4.7.1
Self-Tuning with Arithmetic Method
To optimize the SRM drive performance, it is necessary to maximize torque
per ampere through real-time optimization. The SRM drive control variables
are phase current, turn-on angle, and turn-off angle. In a low-speed region,
hysteresis-type current control is used to keep the commanded current constant. The chopping current band has to be optimally chosen, as there is a
trade-off between the width of the band and the chopping frequency. Assuming the selected band is optimal, maximum torque per ampere can be obtained
by aptly tuning the turn-on angle (θon ) and the turn-off angle (θoff ) of the
phase current excitation. Computer simulations, based on a simple mathematical model, have been performed to prove the existence of a unique (θon ,
θoff ) optimal pair, which gives the maximum torque per ampere for a given
current and speed.31 It has been shown that the optimal values of θon and θoff
are bounded within the following limits:
◦
θmin
on < θon < 0 ,
(6.137)
◦
θmax
off < θoff < 180 ,
(6.138)
where θmin
on is the turn-on angle such that the current reaches the desired value
at 0◦ , and θmax
off is the turn-off angle such that the current reduces to zero at
180◦ .
The intuitive selections for control angles are such as to turn on each phase
exactly at its unaligned position and turn off the phase just before its aligned
position. The optimal θon is not very susceptible to the change in inductance
due to the parameter variations because of the large air gap at the unaligned
position.46 Hence, optimal θon calculated off-line based on the linear model is
sufficient to give the optimal torque per ampere. Therefore, the optimization
problem reduces to calculation of θoff on-line that gives maximum torque per
ampere.
6.4.7.1.1 Optimization with Balanced Inductance Profiles
To minimize the phase current for a given torque and speed, a heuristic search
algorithm for finding the optimum turn-off angle69 can be used in which both
238
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
the reference current and the turn-off angle are varied, while the PI controller
maintains the speed.
The optimization algorithm is explained as follows. Initially the default
turn-off angle is used to reach the commanded speed. Then, the turn-off angle
is reduced in steps. With the turn-off angle variation, the torque will either
decrease or increase, and the PI controller adjusts the phase reference current
accordingly to a new value so that the speed remains at its set value. If the
current reduces with change in the turn-off angle, then the direction of search
is correct and is continued till the current starts increasing with further change
in the turn-off angle. The step size for the turn-off angle can be a function of
the operating point itself.
Once the optimization is completed for a given operating point, the optimum values of control variables—that is, the reference current and the control
angles—can be stored in look-up tables so that the controller can directly pick
up these values if the same operating point is to be reached in future. This
will save some amount of time and effort.
6.4.7.1.2 Optimization in the Presence of Parameter Variations
Initially, when optimization is performed, the reference currents for all the
phases are kept the same, assuming that the phase inductances are balanced.
If there are parameter variations, then different phases will have different
optimal reference currents and turn-off angles. In order to take care of this
problem, once the general optimization is complete, the control variables for
the individual phases are tuned separately, that is, the reference current and
turn-off angle for only one of the phases is varied at a time, while these parameters for the other phases are kept fixed. Finally, when all the phases are tuned,
the optimum reference current and turn-off angle for different phases will be
different if there is any parameter variation. The main advantage of using this
method is that the optimization algorithm does not require any information
about the degree of imbalance present in the inductance profile, which may
change considerably over a period of time.
Figure 6.71 shows the current waveform of the SRM phase with default
values of turn-on and turn-off angles and Figure 6.72 shows the current waveform after applying the self-tune algorithm, which is running almost at the
same operating point. By comparing the two figures, it can be seen that there is
a considerable reduction in the both the amplitude and the width of the phase
current and hence its rms value. The operating speeds are also the same before
and after optimization for both the cases.
6.4.7.2
Self-Tuning Using an ANN
ANNs with highly nonlinear and adaptive structure have been used in many
applications. ANNs have inherent interpolation property, so they are an ideal
candidate for storing turn-on and turn-off angles instead of storing them in
look-up tables. Figure 6.73 shows a three-layer feedforward neural network
239
Electric Propulsion Systems
1
2
-
10.0 ms
Dt
4.15 ms
1 241.0 Hz
Dt
2.5 MS/s
2 DC 22.8 V
10.0 ns < pulse
STOPPED
FIGURE 6.71 Phase current and gating pulse without optimization, terminal voltage: 50 V,
load: 120 W, reference current: 5.5 A, speed: 1200 rpm, conduction angle: 30◦ .
with two inputs—current and speed—and one output—the optimum turnoff angle.70
The proposed self-tuning control technique incorporates a heuristic search
method along with an adaptive-type ANN-based method. The weights of
ANN are initially set to default values. The control technique incorporates
a periodic heuristic search of optimal θoff to verify the accuracy of the θoff
obtained from the ANN. If there is a variation in the inductance profile due
to parameter drift, the optimal θoff obtained from the ANN will no longer be
1
2
10.0 ms
Dt
3.77 ms
2 DC 22.8 V
10.0 ns < pulse
1 265.3 Hz
Dt
2.5 MS/s
STOPPED
FIGURE 6.72 Phase current and gating pulse with optimization, terminal voltage: 50 V, load:
120 W, reference current: 4.7 A, speed: 1200 rpm, conduction angle: 27.25◦ .
240
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
l
qoff
w
FIGURE 6.73 A three-layer feedforward ANN for holding optimal turn-off angles.
valid. This prompts the controller to activate the heuristic search by modifying
the θoff in small steps until the current reaches the minimum value. This
new optimal θoff at that particular operating point is now used to adapt the
weights of the ANN. Hence this novel ANN-based control technique coupled
with the heuristic search learns and adapts to any parameter drift to give the
optimal θoff .
ANNs have been successfully used for many applications in control systems. But the ANN learning algorithm shows a great performance when used
off-line. This means that they have to be fully trained before being applied.
Neural networks with incremental learning capability and stable adaptation
of network parameters are essential for on-line adaptive control. The adaptive
learning is based on the assumption that the ANN to start with is well trained
in such a way that it can perform input/output mapping for the initial training
set with a high degree of accuracy. This can be achieved by training the ANN
with sufficient amount of data to a very low error rate. In this application,
this training can be done off-line as it may require more time.
Now when new training data are obtained, the already trained ANN is used
to generate additional examples. These additional examples with the newly
obtained training data are then used to retrain the current ANN. This ensures
that the original ANN mapping is retained with only a change localized
around the neighborhood of the new training data. This makes the network
gradually adapt to the new data. The above method ensures the stability of
the network weight variations by slow adaptation as the new optimal θoff will
be in the neighborhood of the old value.
Some simulation results70 that show the ability of this algorithm are shown
in Figure 6.74. These results belongs to an 8/6, 12 V, 0.6 kW SRM. This plot
clearly shows the improvement in torque per ampere with optimization,
which is about 13.6%.
6.4.8 Vibration and Acoustic Noise in SRM
Despite the excellent attributes, SRM drives exhibit high levels of torque ripple and audible noise. Indeed, in some noise-sensitive applications such as
241
Electric Propulsion Systems
35
qoff = –5°(optimal)
Torque (N-m)
3
2.5
2
1.5
qoff = –2°
1
0.188 0.19 0.192 0.194 0.196 0.198 0.2
Time (s)
FIGURE 6.74 Developed torque before and after optimization.
domestic products, the problem of acoustic noise and vibration appears to be
particularly important. The acoustic noise in SRM is mainly due to the varying
magnetic forces between the stator and rotor poles, as shown in Figures 6.75
and 6.76.71,72 The tangential and radial components of the electromagnetic
force density in the air gap are given by
Fθ = ν0
Bθ Br dθ,
(6.139)
(Br 2 − B2θ )dθ,
(6.140)
Fr = ν0
where ν0 , Bθ , Br , and θ stand for reluctivity of the air, tangential, and radial
components of the flux density, and rotor position, respectively.
The varying magnetic forces, especially the radial force, cause the deformation of the stator and, therefore, radial vibrations of the stator and acoustic
noise.
Fr
Fr
Fq
FIGURE 6.75 Static profile of the radial component of the force.
242
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Unaligned
position
p
q=
Nr
Fr
Aligned
position
q=0
q (Rotor position)
Fq
q (Rotor position)
FIGURE 6.76 Distribution of radial (Fr ) and tangential (Fθ ) forces.
The results of structural study of SRM show that back iron is the most significant parameter in the dynamic behavior of the stator deformations.60,73
Increasing the back-iron length results in larger natural frequencies and
smaller deformation, which consequently reduce the chance of a mechanical
resonance even at high speeds.
Increasing the air gap length will reduce the radial forces.74 However, it will
vitiate the performance of the SRM. Radial vibration of the stator experiences
a severe acceleration during the turn-off process. This is due to a large magnitude of the attraction forces and their fast rate of change. This is effectively
the point of impact of a hammer on the stator structure. Smoothing of the
radial force during the turn-off process has been found to be the most direct
method for reducing the vibration.75
The current profiling algorithm has to make sure that no negative torque
will be generated. In other words, the phase current has to be completely
removed at or before the aligned position. Also, it must be noted that a high
number of steps in controlling the tail current will increase the switching
losses. Moreover, unconstrained reduction of vibration using this method will
vitiate the performance of the machine. Therefore, study of other objectives
such as efficiency and torque ripple under the proposed control method is an
essential step.71
Fahimi et al.76 showed that, for practical implementation two levels of current will provide a smooth variation in radial force. Therefore, turn-off instant,
the position at which the second current limit is assigned and the position at
which final hard chopping of the phase current occurs, is considered a controlled variable. These parameters are computed at various operating points
using the analytical model of the SRM drive.77
243
Electric Propulsion Systems
6.4.9
SRM Design
SRM has a simple construction. However, this does not mean its design is
simple. Actually, due to the double-salient structure, continuously varying
inductance and high saturation of pole tips, and the fringing effect of pole and
slots, the design of SRM suffers great difficulty in using the magnetic circuit
approach. In most cases, the electromagnetic finite element analysis is used to
determine the motor parameters and performances. Typical electromagnetic
field distributions of an 8/6 SRM are shown in Figure 6.77. Nevertheless,
there are some basic criteria to initialize the design process of SRM for EVs
and HEVs.78,79
6.4.9.1
Number of Stator and Rotor Poles
For continuous rotating, the stator and rotor poles should satisfy some special
conditions, that is, stator poles and rotor poles must be equally distributed
on the circumferences, and pole numbers of the stator and rotor must satisfy
the relationship as
Ns = 2mq,
(6.141)
Nr = 2(mq ∓ 1),
(6.142)
where Ns and Nr are the pole numbers of the stator and rotor, respectively, q is
the phase number of the machine, and m is equal to 1 or 2. In order to reduce
the switch frequency and the minimum inductance, the rotor pole number
is less than the stator pole number, that is, a minus sign is used in Equation
6.142. The most common combination of q, m, Ns , and Nr is listed in Table 6.1.
Four-phase 8/6 and three-phase 6/4 configurations are the most commonly
used SRM structures. A three-phase 6/4 configuration has the advantage of
having more room for phase advancing in high-speed applications. In addition, compared to an 8/6 configuration, this structure will minimize the effects
of mutual coupling between adjacent phases. However, it results in more
(a)
(b)
FIGURE 6.77 Typical electromagnetic field distribution of an 8/6 SRM: (a) aligned position and
(b) unaligned position.
244
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
TABLE 6.1
Common Combination of q, Ns , and Nr
Nr
m=1
m=2
q
Ns
“−”
“+”
Ns
“−”
“+”
3
4
5
6
8
10
4
6
8
8
10
12
12
16
20
10
14
18
14
18
22
torque pulsation due to its torque–angle characteristics, which contains large
dead zones. Furthermore, starting torque can be a problem associated with
this configuration. On the other hand, an 8/6 structure can be used to reduce
the torque ripple and improve the starting torque.11 However, by selecting an
8/6 machine, the cost of the silicon will increase. By increasing the number of
poles per phase (12/8 and 16/12 configuration), one can minimize demerits
of a 6/4 machine while maintaining the same cost of silicon. In this design
study an 8/6 configuration has been selected.
6.4.9.2
Stator Outer Diameter
Stator outer diameter is mainly designed based on the available space given in
desired specifications. In fact, the main compromise has to be made between
the length of the machine and its outer diameter. A pancake (the length of the
machine is less than the stator outer radius) type of the design is subject to
three-dimensional effects of coils’ endings,80 whereas a very long structure
will face cooling and rotor bending issues, which are of particular importance
for large machines.
6.4.9.3
Rotor Outer Diameter
The relationship between the developed torque by SRM and machine
parameters can be represented by the following equation:
T ∝ D2r l(Ni )2 ,
(6.143)
where Dr , Ni , and l are the outer diameter of the rotor, equivalent ampereturn of one phase, and length of the machine. Once the outer diameter of the
SRM is fixed, any increase in rotor outer diameter will result in a reduction
of Ni , thereby reducing the torque developed by SRM. Because of this and
considering the fact that SRM is highly saturated, rotor boring should be
equal or slightly larger than the stator outer radius. It must be noted that the
rotor geometry enhances the moment of inertia and vibrational modes of the
machine.
245
Electric Propulsion Systems
6.4.9.4 Air Gap
Air gap has an important impact on the generated torque and dynamic behavior of the SRM. In fact, by reducing the air gap, inductance at the aligned
position will increase, resulting in higher torque density. On the other hand, a
very small air gap will cause severe saturation in stator and rotor pole areas.74
In addition, mechanical manufacture of a very small air gap might not be feasible. The following empirical formula can be used as a reference for selecting
the air gap in large machines81 :
δ(mm) = 1 +
Ds
,
1000
(6.144)
where δ and Ds are the air gap and the stator outer boring in (mm) and (m),
respectively. By investigating the value of the flux density (B) in the stator and
rotor poles, the level of saturation and consequently air gap can be finalized.
6.4.9.5
Stator Arc
Since the developed torque depends on the area available for the coils, it is
important to design the stator arc in such a way that maximum space for
inserting the coils is provided. A very narrow stator arc will result in the
tangential vibration of the stator pole. Moreover, it reduces the effective region
in the torque–angle characteristics, which increases the torque ripple and
reduces the average torque. An optimal value for the stator arc can be chosen
using the following inequality77 :
0.3
πDR
πDR
≤ λS ≤ 0.35
,
NS
NS
(6.145)
where DR , NS , and λS are, respectively, the rotor diameter, number of stator
poles, and stator arc.
6.4.9.6
Stator Back Iron
For designing the back iron, the following constraints have to be considered:
1. Radial vibration of the stator body has to be minimized.
2. There should be enough space for cooling of the stator.
3. Back iron should be capable of carrying half of the flux existing in the
stator poles without getting saturated.
4. The area available for inserting the coils should not be reduced.
246
6.4.9.7
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Performance Prediction
Clearly, most of the performance requirements are related to the dynamic
performance of the drive and hence call for an overall modeling of the drive
system including control and power electronics considerations. However, in
order to predict the dynamic performance of the drive, static characteristics of the machine (phase inductance and torque–angle profiles) should be
available.
The improved magnetic equivalent circuit (IMEC)82 approach is a shortcut
method that gives an approximation of the steady-state parameters of the
SRM. Indeed, by replacing all of the magnetomotive sources (ampere-turn)
with voltage sources and various parts of the magnetic structure with their
equivalent reluctances, one can perform a magnetic analysis. Furthermore, by
dividing the stator and rotor poles into several smaller portions, accuracy of
Design of SRM
Determination of initial
geometry of the machine
Geometrical
constraints
Selection of material
Analysis of the
electromagnetic field
Torque-speed curve
Post processing to calculate
static characteristics of
the machine
Torque-ripple estimation
Static
requirements
Vibration & acoustic noise
Winding configuration &
control strategy
Efficiency calculation
Dynamic performance
of the SRM
Temperature distribution
Dynamic requirement
End
FIGURE 6.78 Basic design strategy.
Electric Propulsion Systems
247
this method can be arbitrarily improved. It must be noted that finite element
analysis of SRM is a time-consuming procedure. Therefore, the IMEC method
is more appropriate for developing first design examples.
Figure 6.78 depicts a general design strategy for the SRM drive.
References
1. C. C. Chan and K. T. Chau, Modern Electric Vehicle Technology, Oxford University
Press, Oxford, 2001.
2. D. W. Novotny and T. A. Lipo, Vector Control and Dynamics of AC Drives, Oxford
Science Publications, Oxford, ISBN: 0-19-856439-2, 1996.
3. D. C. Hanselman, Brushless Permanent-Magnet Motor Design, McGraw-Hill, New
York, 1994.
4. F. Huang and D. Tien, “A neural network approach to position sensorless control
of brushless DC motors,” in Proceedings of the IEEE 22nd International Conference
on Industrial Electronics, Control, and Instrumentation, Vol. 2, pp. 1167–1170, August
1996.
5. S. Vukosavic, L. Peric, E. Levi, and V. Vuckovic, “Sensorless operation of the
SR motor with constant dwell,” in Proceedings of the 1990 IEEE Power Electronics
Specialists Conference, pp. 451–454, 1990.
6. G. K. Dubey, Power Semiconductor Controlled Drives, Prentice-Hall, Englewood
Cliffs, NJ, 1989.
7. S. R. MacMinn and J. W. Sember, “Control of a switched-reluctance aircraft startergenerator over a very wide speed range,” in Proceedings of the Intersociety Energy
Conversion Engineering Conference, 1989, pp. 631–638.
8. M. Ehsani, Method and Apparatus for Sensing the Rotor Position of a Switched
Reluctance Motor, U.S. Patent No. 5,410,235, April 1995.
9. A. M. Trzynadlowski, The Field Orientation Principle in Control of Induction Motor,
Kluwer Academic Publishers, Dordrecht, 1994.
10. J. F. Gieras and M. Wing, Permanent Magnet Motor Technology, Design and
Applications, Marcel Dekker, New York, 1997.
11. I. Husain, “Minimization of torque ripple in SRM drives,” IEEE Transactions on
Industrial Electronics, 49 (1), 28–39, February 2002.
12. K. M. Rahman and M. Ehsani, “Performance analysis of electric motor drives
for electric and hybrid electric vehicle application,” IEEE Power Electronics in
Transportation, 1996, 49–56.
13. T. Senjyu and K. Uezato, “Adjustable speed control of brushless DC motors
without position and speed sensors,” in Proceedings of the IEEE/IAS Conference on Industrial Automation and Control: Emerging Technologies, pp. 160–164,
1995.
14. A. Consoli, S. Musumeci, A. Raciti, and A. Testa, “Sensorless vector and speed
control of brushless motor drives,” IEEE Transactions on Industrial Electronics, 41,
91–96, February 1994.
15. P. Acarnley, “Sensorless position detection in permanent magnet drives,” IEE
Colloquium on Permanent Magnet Machines and Drives, pp. 10/1–10/4, 1993.
248
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
16. T. Liu and C. Cheng, “Adaptive control for a sensorless permanent-magnet synchronous motor drive,” IEEE Transactions on Aerospace and Electronic Systems, 30,
900–909, July 1994.
17. R. Wu and G. R. Slemon, “A permanent magnet motor drive without a shaft sensor,” IEEE Transactions on Industry Applications, 27, 1005–1011, September/October
1991.
18. T. Liu and C. Cheng, “Controller design for a sensorless permanent magnet
synchronous drive system,” IEE Proc.—B, 140, 369–378, November 1993.
19. N. Ertugrul, P. P. Acarnley, and C. D. French, “Real-time estimation of rotor position in PM motors during transient operation,” IEE Fifth European Conference on
Power Electronics and Applications, pp. 311–316, 1993.
20. N. Ertugrul and P. Acarnley, “A new algorithm for sensorless operation of permanent magnet motors,” IEEE Transactions on Industry Applications, 30, 126–133,
January/February 1994.
21. T. Takeshita and N. Matsui, “Sensorless brushless DC motor drive with EMF
constant identifier,” IEEE International Conference on Industrial Electronics, Control,
and Instrumentation, Vol. 1, pp. 14–19, 1994.
22. N. Matsui and M. Shigyo, “Brushless DC motor control without position
and speed sensors,” IEEE Transactions on Industry Applications, 28, 120–127,
January/February 1992.
23. N. Matsui, “Sensorless operation of brushless DC motor drives,” Proceedings of the
IEEE International Conference on Industrial Electronics, Control, and Instrumentation,
Vol. 2, pp. 739–744, November 1993.
24. N. Matsui, “Sensorless PM brushless DC motor drives,” IEEE Transactions on
Industrial Electronics, 43, 300–308, April 1996.
25. M. Schrodl, “Digital implementation of a sensorless control algorithm for permanent magnet synchronous motors,” Proceedings of the International Conference “SM
100,” ETH Zurich, Switzerland, pp. 430–435, 1991.
26. M. Schrodl, “Operation of the permanent magnet synchronous machine without a mechanical sensor,” IEE Proceedings on the International Conference on Power
Electronics and Variable Speed Drives, pp. 51–56, July 1990.
27. M. Schrodl, “Sensorless control of permanent magnet synchronous motors,”
Electric Machines and Power Systems, 22, 173–185, 1994.
28. B. J. Brunsbach, G. Henneberger, and T. Klepsch, “Position controlled permanent
magnet excited synchronous motor without mechanical sensors,” IEE Conference
on Power Electronics and Applications, Vol. 6, pp. 38–43, 1993.
29. R. Dhaouadi, N. Mohan, and L. Norum, “Design and implementation of an
extended Kalman filter for the state estimation of a permanent magnet synchronous motor,” IEEE Transactions on Power Electronics, 6, 491–497, July 1991.
30. A. Bado, S. Bolognani, and M. Zigliotto, “Effective estimation of speed and rotor
position of a PM synchronous motor drive by a Kalman filtering technique,” in
Proceedings of the 23rd IEEE Power Electronics Specialist Conference, Vol. 2, pp. 951–
957, 1992.
31. K. R. Shouse and D. G. Taylor, “Sensorless velocity control of permanent-magnet
synchronous motors,” in Proceedings of the 33rd Conference on Decision and Control,
pp. 1844–1849, December 1994.
32. J. Hu, D. M. Dawson, and K. Anderson, “Position control of a brushless DC motor
without velocity measurements,” IEE Proceedings on Electronic Power Applications,
142, 113–119, March 1995.
Electric Propulsion Systems
249
33. J. Solsona, M. I. Valla, and C. Muravchik, “A nonlinear reduced order observer
for permanent magnet synchronous motors,” IEEE Transactions on Industrial
Electronics, 43, 38–43, August 1996.
34. R. B. Sepe and J. H. Lang, “Real-time observer-based (adaptive) control of
a permanent-magnet synchronous motor without mechanical sensors,” IEEE
Transactions on Industry Applications, 28, 1345–1352, November/December 1992.
35. L. Sicot, S. Siala, K. Debusschere, and C. Bergmann, “Brushless DC motor control
without mechanical sensors,” in Proceedings of the IEEE Power Electronics Specialist
Conference, pp. 375–381, 1996.
36. T. Senjyu, M. Tomita, S. Doki, and S. Okuma, “Sensorless vector control of brushless DC motors using disturbance observer,” in Proceedings of the 26th IEEE Power
Electronics Specialists Conference, Vol. 2, pp. 772–777, 1995.
37. K. Iizuka, H. Uzuhashi, and M. Kano, “Microcomputer control for sensorless
brushless motor,” IEEE Transactions on Industry Applications, IA-27, 595–601, May–
June 1985.
38. J. Moreira, “Indirect sensing for rotor flux position of permanent magnet AC
motors operating in a wide speed range,” IEEE Transactions on Industry Applications
Society, 32, 401–407, November/December 1996.
39. S. Ogasawara and H. Akagi, “An approach to position sensorless drive for
brushless DC motors,” IEEE Transactions on Industry Applications, 27, 928–933,
September/October 1991.
40. T. M. Jahns, R. C. Becerra, and M. Ehsani, “Integrated current regulation for a
brushless ECM drive,” IEEE Transactions on Power Electronics, 6, 118–126, January
1991.
41. R. C. Becerra, T. M. Jahns, and M. Ehsani, “Four-quadrant sensorless brushless
ECM drive,” in Proceedings of the IEEE Applied Power Electronics Conference and
Exposition, pp. 202–209, March 1991.
42. D. Regnier, C. Oudet, and D. Prudham, “Starting brushless DC motors utilizing
velocity sensors,” in Proceedings of the 14th Annual Symposium on Incremental Motion
Control Systems and Devices, Champaign, IL, Incremental Motion Control Systems
Society, pp. 99–107, June 1985.
43. D. Peters and J. Harth, “I.C.s provide control for sensorless DC motors,” EDN,
pp. 85–94, April 1993.
44. M. Hamdi and M. Ghribi, “A sensorless control scheme based on fuzzy logic for
AC servo drives using a permanent-magnet synchronous motor,” IEEE Canadian
Conference on Electrical and Computing Engineering, pp. 306–309, 1995.
45. D. E. Hesmondhalgh, D. Tipping, and M. Amrani, “Performance and design of an
electromagnetic sensor for brushless DC motors,” IEE Proceedings, 137, 174–183,
May 1990.
46. T. J. E. Miller, Switched Reluctance Motors and Their Control, Oxford Science
Publications, London, 1993.
47. P. J. Lawrenson, J. M. Stephenson, P. T. Blenkinsop, J. Corda, and N. N. Fulton,
“Variable-speed switched reluctance motors,” Proceedings of IEE, 127, Part B (4),
253–265, July 1980.
48. E. Richter, J. P. Lyons, C. A. Ferreira, A. V. Radun, and E. Ruckstatder, “Initial
testing of a 250 kW starter/generator for aircraft applications,” in Proceedings of
the SAE Aerospace Atlantic Conference Expo., Dayton, OH, April 18–22, 1994.
49. M. Ehsani, Phase and Amplitude Modulation Techniques for Rotor Position Sensing in
Switched Reluctance Motors, U.S. Patent No. 5,291,115, March 1994.
250
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
50. D. A. Torrey, “Variable-reluctance generators in wind-energy systems,” in Proceedings of the IEEE PESC ’93, pp. 561–567, 1993.
51. J. M. Kokernak, D. A. Torrey, and M. Kaplan, “A switched reluctance
starter/alternator for hybrid electric vehicles,” in Proceedings of the PCIM ’99,
pp. 74–80, 1999.
52. J. T. Bass, M. Ehsani, and T. J. E. Miller, “Simplified electronics for torque control of
sensorless switched reluctance motor,” IEEE Transactions on Industrial Electronics,
34 (2), 1987.
53. M. Ehsani, Self-Tuning Control of Switched Reluctance Motor Drives System, U.S.
Patent Pending, File Number 017575.0293.
54. M. Ehsani, Switched Reluctance Motor Drive System, U.S. Patent Pending, Filing
Date: January 1997, Serial Number 60/061,087.
55. R. Krishnan, Switched Reluctance Motors Drives: Modeling, Simulation Analysis,
Design and Applications, CRC Press, Boca Raton, FL, 2001.
56. N. Mohan, T. M. Undeland, and W. P. Robbins, Power Electronics—Converters,
Applications, and Design, John Wiley & Sons, New York, ISBN: 0-471-58408-8, 1995.
57. A. Radun, “Generating with the switched-reluctance motor,” in Proceedings of the
IEEE APEC ’94, pp. 41–47, 1994.
58. B. Fahimi, “A switched reluctance machine based starter/generator for more
electric cars,” in Proceedings of the IEEE Electric Machines and Drives Conference,
pp. 73–78, 2000.
59. H. Gao, F. R. Salmasi, and M. Ehsani, “Sensorless control of SRM at standstill,” in
Proceedings of the 2000 IEEE Applied Power Electronics Conference, Vol. 2, pp. 850–856,
2000.
60. J. P. Lyons, S. R. MacMinn, and M. A. Preston, “Discrete position estimator for a
switched reluctance machine using a flux-current map comparator,” U.S. Patent
5140243, 1991.
61. G. Suresh, B. Fahimi, K. M. Rahman, and M. Ehsani, “Inductance based position encoding for sensorless SRM drives,” in Proceedings of the 1999 IEEE Power
Electronics Specialists Conference, Vol. 2, pp. 832–837, 1999.
62. H. Gao, Sensorless Control of the Switched Reluctance Motor at Standstill and NearZero Speed, Ph.D. Dissertation, Texas A&M University, December 2001.
63. M. Ehsani, Position Sensor Elimination Technique for the Switched Reluctance Motor
Drive, U.S. Patent 5072166, 1990.
64. M. Ehsani, I. Husain, S. Mahajan, and K. R. Ramani, “New modulation encoding
techniques for indirect rotor position sensing in switched reluctance motors,”
IEEE Transactions on Industry Applications, 30 (1), 85–91, 1994.
65. G. R. Dunlop and J. D. Marvelly, “Evaluation of a self commuted switched reluctance motor,” in Proceedings of the 1987 Electric Energy Conference, pp. 317–320,
1987.
66. M. Ehsani and I. Husain, “Rotor position sensing in switched reluctance motor
drives by measuring mutually induced voltages,” in Proceedings of the 1992 IEEE
Industry Application Society Annual Meeting, Vol. 1, pp. 422–429, 1992.
67. A. Lumsdaine and J. H. Lang, “State observer for variable reluctance motors,”
IEEE Transactions on Industrial Electronics, 37 (2), 133–142, 1990.
68. K. Russa, I. Husain, and M. E. Elbuluk, “A self-tuning controller for switched
reluctance motors,” IEEE Transactions on Power Electronics, 15 (3), pp. 545–552,
May 2000.
Electric Propulsion Systems
251
69. P. Tandon, A. Rajarathnam, and M. Ehsani, “Self-tuning of switched reluctance motor drives with shaft position sensor,” IEEE Transactions on Industry
Applications, 33 (4), pp. 1002–1010, July/August 1997.
70. A. Rajarathnam, B. Fahimi, and M. Ehsani, “Neural network based self-tuning
control of a switched reluctance motor drive to maximize torque per ampere,”
in Proceedings of the IEEE Industry Applications Society Annual Meeting, Vol. 1, pp.
548–555, 1997.
71. B. Fahimi, Control of Vibration in Switched Reluctance Motor Drive, Ph.D. Dissertation, Texas A&M University, May 1999.
72. D. E. Cameron, J. H. Lang, and S. D. Umans, “The origin and reduction of acoustic
noise an doubly salient variable reluctance motor,” IEEE Transactions on Industry
Applications, IA-28 (6), 1250–1255, November/December 1992.
73. H. Gao, B. Fahimi, F. R. Salmasi, and M. Ehsani, “Sensorless control of the switched
reluctance motor drive based on the stiff system control concept and signature
detection,” in Proceedings of the 2001 IEEE Industry Applications Society Annual
Meeting, pp. 490–495, 2001.
74. B. Fahimi and M. Ehsani, “Spatial distribution of acoustic noise caused by radial
vibration in switched reluctance motors: application to design and control,” in
Proceedings of the 2000 IEEE Industry Application Society Annual Meeting, Rome,
Italy, October 2000.
75. B. Fahimi, G. Suresh, K. M. Rahman, and M. Ehsani, “Mitigation of acoustic noise
and vibration in switched reluctance motor drive using neural network based current profiling,” in Proceedings of the 1998 IEEE Industry Application Society Annual
Meeting, Vol. 1, pp. 715–722, 1998.
76. B. Fahimi and M. Ehsani, Method and Apparatus for Reducing Noise and Vibration in
Switched Reluctance Motor Drives, U.S. Patent pending.
77. G. S. Buja and M. I. Valla, “Control characteristics of the SRM drives—part I:
Operation in the linear region,” IEEE Transactions on Industrial Electronics, 38 (5),
313–321, October 1991.
78. B. Fahimi, G. Suresh, and M. Ehsani, “Design considerations of switched reluctance motors: vibration and control issues,” in Proceedings of the 1999 IEEE Industry
Application Society Annual Meeting, Phoenix, AZ, October 1999.
79. J. Faiz and J. W. Finch, “Aspects of design optimization for switched reluctance
motors,” IEEE Transactions on Energy Conversion, 8 (4), 704–712, December 1993.
80. A. M. Michaelides and C. Pollock, “Effect of end core flux on the performance
of the switched reluctance motor,” IEE on Electronic Power Applications, 141 (6),
308–316, November 1994.
81. G. Henneberger, Elektrische Maschinen I, II, III, RWTH Aachen, Manuscripts at
Institut fuer Elektrische Maschinen, 1989.
82. B. Fahimi, G. Henneberger, and M. Moallem, “Prediction of transient behavior of
SRM drive using improved equivalent magnetic circuit method,” PCIM Conference
Records, pp. 285–291, 1995.
7
Design Principle of Series (Electrical
Coupling) Hybrid Electric Drive Train
The concept of a series hybrid electric drive train was developed from the EV
drive train.1 As mentioned in Chapter 4, EVs, compared with conventional
gasoline- or diesel-fueled vehicles, have the advantages of zero mobile pollutant emissions, multienergy sources, and high efficiency. However, EVs using
present technologies have some disadvantages: a limited drive range due to
the shortage of energy storage in the on-board batteries, limited payload and
volume capacity due to heavy and bulky batteries, and long battery charging
time. The initial objective of developing a series HEV was aimed at extending
the drive range by adding an engine/alternator system to charge the batteries
on-board.
A typical series hybrid electric drive train configuration is shown in
Figure 7.1. The vehicle is propelled by a traction motor. The traction motor
is powered by a battery pack and/or an engine/generator unit. The powers
of both power sources are merged together in a power electronics-based and
controllable electrical coupling device. Many operation modes are available
to choose, according to the power demands of the driver and the operation
status of the drive train system.
Vehicle performance (in terms of acceleration, gradeability, and maximum
speed) is completely determined by the size and characteristics of the traction motor drive. Motor power capability and transmission design are the
same as in the EV design discussed in Chapter 4. However, the drive train
control is essentially different from the pure electric drive train due to the
involvement of the additional engine/generator unit. This chapter will focus
on the design principles of the engine/alternator system, the drive train control, and the energy and power capacity of the battery pack. In this chapter,
the term “peak power source” will replace “battery pack” because, in HEVs,
the major function of the batteries is to supply peaking power and they can
be replaced with other kinds of sources such as ultracapacitors, flywheels, or
combinations.
253
254
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Vehicle
controller
brake
Mechanical
brake
controller
PPS SOC
Engine
control
Peaking
power
source
Engine
controller
Engine
Generator
Electric
coupling
device
Motor control
Engine speed
throttle position
Operation
commands
Motor
controller
Motor
Transmission
Vehicle speed
Mechanical connection
Electrical power
Signals
FIGURE 7.1 Configuration of a typical series hybrid electric drive train.
7.1
Operation Patterns
In series hybrid electric drive trains, the engine/generator system is
mechanically decoupled from the driven wheels as shown in Figure 7.1. The
speed and torque of the engine are independent of vehicle speed and traction
torque demand, and can be controlled to any operating point on its speedtorque plane.2,3 Generally, the engine should be controlled in such a way
that it always operates in its optimal operation region, where fuel consumption and emissions of the engine are minimized (see Figure 7.2). Due to the
mechanical decoupling of the engine from the driven wheels, this optimal
engine operation is realizable. However, it heavily depends on the operating
modes and control strategy of the drive train.
The drive train has several operating modes, which can be used selectively
according to the driving conditions and wish of the driver. These operating
modes are as follows:
1. Hybrid traction mode: When a large amount of power is demanded,
that is, the driver depresses the accelerator pedal deeply, both
engine/generator and peaking power source (PPS) supply their powers to the electric motor drive. In this case, the engine should be
controlled to operate in its optimal region for efficiency and emission reasons as shown in Figure 7.2. The PPS supplies the additional
power to meet the traction power demand. This operation mode can
be expressed as
Pdemand = Pe/g + Ppps ,
(7.1)
Design Principle of Series (Electrical Coupling) Hybrid Electric Drive Train 255
60
Maximum
power
curve
Power (kW)
50
40
30
Optimal
operation
region
Specific fuel
consumption
g/kWh
(efficiency)
25
0
(3
27 4.3
%
0
(3 )
1.
8% 2
) 60
(3
28
3.
0%
0
(3
)
0.
6%
)
70
31
2
0(
7.7
%)
0
35
%)
1.4
2
0(
40
)
.1%
17
(
00
20
5
10
0
500
)
.5%
4
(2
)
)
.3%
(14 12.2%
0
60 00 ( .7%)
7 (10
800 8.57%)
(
1000
1000 1500 2000 2500 3000 3500 4000 4500 5000 5500
Crankshaft (rpm)
FIGURE 7.2 Example of engine characteristics and optimal operating region.
where Pdemand is the power demanded by the driver, Pe/g is the
engine/generator power, and Ppps is the PPS power.
2. Peak power source-alone traction mode: In this operating mode, the peak
power source alone supplies its power to meet the power demand,
that is,
Pdemand = Ppps .
(7.2)
3. Engine/generator-alone traction mode: In this operating mode, the
engine/generator alone supplies its power to meet the power demand,
that is,
Pdemand = Pe/g .
(7.3)
4. PPS charging from the engine/generator: When the energy in the PPS
decreases to a bottom line, the PPS must be charged. This can be
done by regenerative braking or by the engine/generator. Usually,
engine/generator charging is needed, since regenerative braking
charging is insufficient. In this case, the engine/generator power is
divided into two parts: one to propel the vehicle and the other to
charge the PPS. That is,
Pdemand = Pe/g + Ppps .
(7.4)
It should be noticed that the operation mode is only effective when
the power of the engine/generator is greater than the load power
256
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
demand. It should be noted that PPS power is given a negative sign
when it is being charged.
5. Regenerative braking mode: When the vehicle is braking, the traction
motor can be used as a generator, converting part of the kinetic energy
of the vehicle mass into electric energy to charge the PPS.
As shown in Figure 7.1, the vehicle controller commands the operation of
each component according to the traction power (torque) command from the
driver, the feedback from each of the components, and also the drive train and
the preset control strategy. The control objectives are to (1) meet the power
demand of the driver, (2) operate each component with optimal efficiency,
(3) recapture braking energy as much as possible, and (4) maintain the state
of charge (SOC) of the PPS in a preset window.
7.2
Control Strategies
A control strategy is a control rule that is preset in the vehicle controller and
commands the operation of each component. The vehicle controller receives
operation commands from the driver and feedback from the drive train and
all the components, and then makes decisions to use proper operation modes.
Obviously, the performance of the drive train relies mainly on control quality,
in which control strategy plays a crucial role.
In practice, there are a number of control strategies that can be employed in
a drive train for vehicles with different mission requirements. In this chapter,
two typical control strategies are introduced: (1) maximum state-of-charge of
peaking power source (Max. SOC-of-PPS) and (2) engine turn-on and turn-off
(engine on/off) or thermostat control strategies.4
7.2.1
Max. SOC-of-PPS Control Strategy
The target of this control strategy is to meet the power demand commanded
by the driver and, at the same time, maintain the SOC of the PPS at its high
level. The engine/generator is the primary power source, and the PPS is the
secondary source. This control strategy is considered to be the proper design
for vehicles in which performance (speed, acceleration, gradeability, etc.) is
the first concern, such as vehicles with frequent stop–go driving patterns and
military vehicles in which carrying out their mission is the most important
objective. A high SOC level in the PPS will guarantee the high performance
of vehicles at any time.
The Max. SOC-of-PPS control strategy is depicted in Figure 7.3, in which
points A, B, C, and D represent the power demands that the driver commanded in either traction mode or braking mode. Point A represents
the commanded traction power that is greater than the power that the
Design Principle of Series (Electrical Coupling) Hybrid Electric Drive Train 257
Power
Max. traction motor power
A
Ppps-max
Ppps
B
Pe/g
Pcom
Ppps-cha
Pe/g-full
Pcom Pe/g-partial
B---Engine/generator alone traction or PPS charging mode
Ppps-cha--- PPS charging power
Vehicle
speed
Pcom
Pregen
Pcom
Pregen
A---Hybrid traction mode
Pcom--- Commanded power
Pcom--- Power of the peaking power source
Pe/g --- Power of the engine/generator
C---Hybrid braking mode
Pregen --- Regenerative braking power
Pb-mech--- Mechanical braking power
D---Regenerative braking mode
D
Pb-mech Max. regenerative braking power
C
FIGURE 7.3 Illustration of the maximum PPS SOC control strategy.
engine/generator can produce. In this case, the PPS must produce its power
to make up the power shortage of the engine/generator. Point B represents
the commanded power that is less than the power that the engine/generator
produces when operating in its optimal operation region (refer to Figure 7.2).
In this case, two operating modes may be used, depending on the SOC level
of the PPS. If the SOC of the PPS is below its top line, such as less than
70%, the engine/generator is operated with full load. (The operating point
of the engine/generator with full load depends on the engine/generator
design. For details, see the next section.) Part of its power goes to the traction
motor to propel the vehicle and the other part goes to the PPS to increase
the energy level. On the other hand, if the SOC of the PPS has reached its
top line, the engine/generator traction mode alone is supplied, that is, the
engine/generator is controlled to produce power equal to the demanded
power, and the PPS is set at idle. Point C represents the commanded braking
power that is greater than the braking power the motor can produce (maximum regenerative braking power). In this case, a hybrid braking mode is
used, in which the electric motor produces its maximum braking power and
the mechanical braking system produces the remaining braking power. Point
D represents the commanded braking power that is less than the maximum
braking power that the motor can produce. In this case, only regenerative
braking is used. The control flowchart of the Max. SOC-of-PPS is illustrated
in Figure 7.4.
7.2.2
Engine On–Off or Thermostat Control Strategy
The Max. SOC-of-PPS control strategy emphasizes maintaining the SOC of
the PPS at a high level. However, in some driving conditions, such as driving
258
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Traction power
command, Ptraction
Braking power
command, Pbrake
Traction?
No
Yes
Maximum
motor power
Pm-max
Regenerative
braking
If Pbrake > Pm-max
No
Yes
Engine/generator
power, Pe/g
Hybrid traction
(eng./gen. + PPS)
Ptraction < Pe/g
Hybrid braking
No
Yes
SOC of PPS
If SOC < SOPtop
Yes
Eng./gen/
alone traction
No
PPS
charging
FIGURE 7.4 Control flowchart of the Max. SOC-of-PPS control strategy.
for a long time (with a low load) on a highway at constant speed, the PPS can
be easily charged to its full level, and the engine/generator is forced to operate
with power output smaller than its optimum. Hence, the efficiency of the drive
train is reduced. In this case, the engine on–off or thermostat control strategy
would be appropriate. This control strategy is illustrated in Figure 7.5. The
operation of the engine/generator is completely controlled by the SOC of the
PPS. When the SOC of the PPS reaches its preset top line, the engine/generator
is turned off and the vehicle is propelled only by the PPS. On the other hand,
when the SOC of the PPS reaches its bottom line, the engine/generator is
turned on. The PPS gets its charging from the engine/generator. In this way,
the engine can be always operated within its optimal deficiency region.
PPS SOC
PPS SOC top line
Engine operation
PPS SOC bottom line
On
Off
FIGURE 7.5 Illustration of thermostat control.
Design Principle of Series (Electrical Coupling) Hybrid Electric Drive Train 259
7.3 Design Principles of a Series (Electrical Coupling)
Hybrid Drive Train
Successful design of the drive train system means ensuring the vehicle being
capable of achieving the desired performance, such as acceleration, gradeability, high speed, and high operating efficiency. The traction motor drive,
engine/generator unit, PPS, and electrical coupling device are the major
design components of concern. Their design should primarily be considered
at the system level so as to ensure that all the components work harmoniously.
7.3.1
Electrical Coupling Device
As mentioned above, the electrical coupling device is the sole linkage point for
combining the three sources of powers together: engine/generator, PPS, and
traction motor. Its major function is to regulate the power (electric current)
flow between these power sources and sinks. The power (current) regulation
is carried out based on the proper control of the terminal voltages. The simplest structure is to connect the three terminals together directly as shown in
Figure 7.6.
This configuration is the simplest and has the lowest cost. Its major feature
is that the bus voltage is equal to the rectified voltage of the generator and that
of the PPS. The bus voltage is determined by the minimum of the two voltages
above. The power flow is solely controlled by the voltage of the generator. To
deliver its power to the traction motor and/or the PPS, the open circuit voltage
(zero current) of the generator, rectified, must be higher than the PPS voltage.
This can be done by controlling the engine throttle and/or the magnetic field
Driving
command
Vehicle
controller
+
Engine
Generator
Motor controller
(power electronics)
Rectifier
–
+
–
PPS
FIGURE 7.6 Directly connected power source and sink.
Motor
260
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
of the generator. When the engine/generator is controlled to generate the
rectified terminal voltage equal to the open circuit voltage of the PPS, the PPS
does not deliver power and the engine/generator alone powers the electric
motor. When the rectified voltage of the engine/generator is lower than the
PPS voltage, the PPS alone powers the electric motor. In regenerative braking,
the generated bus voltage by the traction motor must be higher than the
PPS voltage. However, the voltage generated by the traction motor is usually
proportional to the rotational speed of the motor. Therefore, the regenerative
braking capability in low speed will be rather limited for this design. It is also
obvious that this simple design requires the engine/generator and the PPS to
have the same rated voltage. This constraint may result in a heavy PPS due
to the high voltage.
Adding a DC/DC converter, and thus releasing the voltage constraints,
may significantly improve the performance of the drive train.5,6 Two alternative configurations are shown in Figures 7.7 and 7.8. In the configuration
of Figure 7.7, the DC/DC converter is placed between the PPS and the DC
bus and the engine–generator–rectifier is connected directly to the DC bus.
In this configuration, the PPS voltage is allowed to be different from the
DC bus voltage, and the rectified voltage of the engine/generator is always
equal to the DC bus voltage. In the configuration of Figure 7.8, the DC/DC
converter is placed between the engine–generator–rectifier and the DC bus
and the PPS is directly connected to the DC bus. Contrary to the configuration of Figure 7.7, the DC/DC converter conditions the rectified voltage of
the engine/generator and the voltage of the PPS is always equal to the DC
bus voltage.
Among these two configurations, the one in Figure 7.7 seems to be more
appropriate. Its advantages over the other one are mainly the following: (1)
changes in the voltage of PPS do not affect the DC bus voltage, (2) the energy
in the PPS can be fully used, (3) the voltage of the DC bus can be maintained by
controlling the engine throttle and/or the magnetic field of the generator, (4)
a low PPS voltage can be used, which may lead to small and light PPS pack
and less cost, and (5) the charging current of PPS can be regulated during
regenerative braking and charging from the engine/generator.
Engine/generator
and
rectifier
+
DC bus
Motor
drive
–
Traction
motor
+
–
DC/DC
converter
PPS
FIGURE 7.7 Configuration with the DC/DC converter on the PPS side.
Design Principle of Series (Electrical Coupling) Hybrid Electric Drive Train 261
DC bus
+
Engine/generator
and
rectifier
DC/DC
converter
Motor
drive
Traction
motor
–
+
–
PPS
FIGURE 7.8 Configuration with the DC/DC converter on the engine/generator side.
It is obvious that the DC/DC converter in this configuration has to be bidirectional. In the case of the rated voltage of the PPS being lower than the DC
bus voltage, the DC/DC converter has to boost the PPS voltage to the level of
the DC bus to deliver its power to the DC bus and buck the DC bus voltage
to the level of the PPS charging voltage to charge the PPS. In regenerative
braking, if the voltage generated by the traction motor at a given low speed
is still higher than the voltage of the PPS, the buck DC/DC converter in the
PPS charging direction is still usable. However, if the voltage generated by
the traction motor at the given low speed is lower than the terminal voltage of the PPS, the DC/DC converter may need to boost the DC bus voltage
to charge the battery. In this case, a buck/boost (step down/step up) converter is needed. The basic functions of the DC/DC required converter are
summarized in Figure 7.9.
Figure 7.10 shows a bidirectional DC/DC converter connected between the
low voltage of PPS and the high-voltage of the DC bus, boosting for PPS discharging (traction) and bucking for PPS charging from the engine/generator
or from regenerative braking.6 In the PPS discharging (traction) mode, switch
S1 is turned off, and switch S2 is turned on and off periodically. In the on period
of S2 , the inductor Ld is charged with energy from the PPS and the load is
powered by the capacitor C, as shown in Figure 7.11a. In the off period of S2 ,
both the PPS and the inductor supply energy to the load and the charging of
capacitor C, as shown in Figure 7.11b.
In the PPS charging mode from the engine/generator or traction motor in
regenerative braking, the DC/DC converter bucks the high voltage of the
DC bus to the low voltage of the PPS. Switch S1 and diode D2 serve as a
Energy flow
PPS traction
PPS charging from e/g
Regenerative braking
PPS discharging
PPS charging
Boost
—
—
Buck
—
Buck or
buck/boost
FIGURE 7.9 Basic functions of the DC/DC converter.
262
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Rectifier
+
Generator
DC/DC
converter
High
voltage
Low +
voltage
S1
D1 S3
Ld
Vpps
Inverter or other DC
source motor drive
D3
+
A
B
VDC
S2
–
D2
C –
S4
Traction
motor
C
D4
FIGURE 7.10 Bidirectional DC/DC with low-voltage PPS and high-voltage DC bus.
(a)
(b)
D1
Ld
Low +
voltage
Vpps
–
–
S2
iL
VDC
C
iL
D1
ic
+
Low +
voltage
Vpps
Ld
ic
+
VDC
C
–
S2
–
FIGURE 7.11 Current flow during the on and off periods of S2 in the PPS discharging mode:
(a) in S2 on period and (b) in S2 off period.
unidirectional buck converter. The current flows during the on and off periods
of S1 are shown in Figure 7.12.
As mentioned above, when the generated voltage by the traction motor at
low speed, in regenerative braking, is lower than the voltage of the PPS, a
bidirectional boost/buck DC/DC converter is required. Such a bidirectional
buck/boost DC/DC converter is shown in Figure 7.13. Its basic operations in
PPS discharging and charging modes are as follows.
In the PPS discharging mode, that is, boosting the PPS voltage to the DC bus
level, switch S1 is always on, S2 and S3 are always off, and S4 are turned on and
off periodically in the same manner as S2 in Figure 7.11. In the PPS charging
mode with the DC bus voltage higher than the PPS voltage in regenerative
Design Principle of Series (Electrical Coupling) Hybrid Electric Drive Train 263
(a)
(b)
ic
S1
Ld
ipps
S1
+
Low +
voltage
C
Vpps
D2
ic
Ld
VDC
–
+
VDC
C
Low
voltage
–
D2
ipps
Vpps
FIGURE 7.12 Current flow during the on and off periods of S1 in PPS charging mode: (a) in S1
on period and (b) in S2 off period.
Rectifier
+
a
Generator
b
c
–
S1
D1
+
Ld
Lower +
voltage
VDC
C
–
S2
Vpps
Inverter or other DC
source motor drive
High
voltage
DC/DC
converter
A
B
C
Traction
motor
D2
–
FIGURE 7.13 A boost/buck DC/DC converter.
braking or engine/generator charging mode, that is, bucking the DC bus
voltage to PPS level, switches S1 , S2 , and S4 are turned off, and S3 are turned
on and off periodically in the same manner as shown in Figure 7.12. In the
PPS charging mode with the DC bus voltage lower than the PPS voltage
(regenerative braking at low speed), that is, boosting the DC bus voltage to
the PPS level, switches S1 and S4 are kept off, S3 on, and S2 is turned on and off
periodically. In the on period of S2 , the inductor Ld is charged by the DC bus
through S3 and S2 In the off period of S2 , both the DC bus and the inductor
charge the PPS through S3 and D1 .
264
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Actually in the DC/DC converter in Figure 7.13, there is a spare function,
which is bucking the PPS voltage to bus voltage if the PPS voltage is higher
than the DC bus voltage. This case would never occur in this application.
7.3.2
Power Rating Design of the Traction Motor
Similar to the pure EV discussed in Chapter 4, the power rating of the electric
motor drive in series HEV is completely determined by vehicle acceleration
performance requirement, motor characteristics, and transmission characteristics (refer Chapter 4). At the beginning of the design, the power rating of
the motor drive can be estimated, according to the acceleration performance
(time used to accelerate the vehicle from zero speed to a given speed), using
the following equation:
Pt =
2
δM 2
1
Vf + Vb2 + Mg fr Vf + ρa CD Af Vf3 ,
2ta
3
5
(7.5)
where M is the total vehicle mass in kg, ta is the expected acceleration time
in s, Vb is the vehicle speed in m/s, corresponding to the motor-based speed
(see Figure 7.14), Vf is the final speed of the vehicle during acceleration in
m/s, g is the gravity acceleration in 9.80 m/s2 , fr is the tire rolling resistance
coefficient, ρa is the air density in 1.202 kg/m3 , Af is the front area of the
vehicle in m2 , and CD is the aerodynamic drag coefficient.
9
8
a
Low gear
Motor speed ratio
x=4
b
High gear
6
c
d
5
Power
4
3
80
f
2
40
FW + Fr
g
1
0
60
Tractive effort
e
0
20
40
Vb1
60
80
100 120 140
Vb2 Vehicle speed (km/h)
160
180
FIGURE 7.14 Speed–torque (power) characteristics of an electric motor.
20
200
Tractive power (kW)
Tractive effort (kN)
7
Design Principle of Series (Electrical Coupling) Hybrid Electric Drive Train 265
The first term in Equation 7.5 represents the power used to accelerate the
vehicle mass, and the second and third terms represent the average power
for overcoming the tire rolling resistance and aerodynamic drag.
Figure 7.14 shows the tractive effort and tractive power versus vehicle speed
with a two-gear transmission. During acceleration, starting from low gear,
the tractive effort follows the trace of a–b–d–e–f. At point f, the electric motor
reaches its maximum speed and the transmission has to be shifted to high gear
for further acceleration. In this case the base speed of the vehicle in Equation
7.5 is Vb1 . However, when a single-gear transmission is used, that is, only
high gear is available, the tractive effort follows the trace of c–d–e–f–g, and
Vb = Vb2 .
It is obvious that for a given final speed during acceleration, such as
100 km/h at point e, the vehicle with a two-gear transmission will have
a short acceleration time, because the tractive effort at low speed in low
gear, represented by a–b–d, is larger than that in higher gear, represented
by a–d.
Figure 7.15 shows an example of the power rating of motor versus speed
ratio, which is defined as the ratio of maximum speed to base speed as shown
in Figure 7.14.
It should be noted that the rated motor power determined by Equation 7.5
is only an estimate for meeting acceleration performance. In some special
applications, such as in off-road military vehicles, the cross-country operation
may be the primary concern. In this case, the traction motor must be powerful
enough to overcome the required maximum grade of an off-road trail. The
Power (kW)
120
110
ta = 10 s
M = 1500 kg
f = 0.001
100
CD = 0.3
Af = 2.0 m2
90
80
70
60
2
3
4
5
6
7
x, Max. speed/base speed
8
9
FIGURE 7.15 Power rating of traction power versus speed ratio of a drive train.
10
266
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Tractive effort and resistance (kN)
70
35° (70.0%)
x = 2, Pm = 758.8 kW
x = 3, Pm = 505.8 kW
60
x = 4, Pm = 379.4 kW
50
x = 5, Pm = 303.5 kW
40
30° (57.7%)
25° (46.6%)
20° (36.4%)
15° (26.8%)
30
10° (17.6%)
20
5° (8.75%)
0° (0%)
x = 6, Pm = 252.9 kW
10
x = 7, Pm = 216.8 kW
x = 2, Pm = 189.7 kW
0
0
20
40
60
80
100
120
Vehicle speed (km/h)
FIGURE 7.16 Tractive effort versus vehicle speed with different speed ratios and motor power.
traction power in hill climbing can be expressed as
1
Pgrade = Mg fr cos α + ρa CD Af V 2 + Mg sin α V(W),
2
(7.6)
where α is the ground slope angle and V is the vehicle speed in m/s, specified
by the gradeability requirement. When the off-road vehicle is climbing its
required maximum slope, 60% or 31◦ , for example, at a speed of 10 km/h in
real operation, the ground is usually unpaved and the rolling resistance is
much larger than those of paved roads because of the road surface deformations. Therefore, in the calculation of motor power required for gradeability,
additional resistance power should be added to reflect this situation.
Based on the specified gradeability requirement of 60% or 31◦ at 10 km/h,
the tractive efforts versus vehicle speeds of a 10-ton military vehicle, with
different extended speed ratios and motor power ratings, can be calculated
by using Equation 7.6 as shown in Figure 7.16. It can be seen that larger
extended speed ratios can effectively reduce the power rating requirement
of the traction motor to meet the gradeability requirement. However, the
speed on the maximum slope will be smaller. The large extended speed
ratio can be implemented either by the motor itself or by a multigear
transmission.
To ensure that the vehicle meets an acceleration requirement, for example,
8 s from zero to 48 km/h, the motor power rating requirement with different
extended speed ratios on hard roads is also calculated by using Equation 7.5.
Figure 7.17 shows the calculation results. It is obvious that motor power rating
Design Principle of Series (Electrical Coupling) Hybrid Electric Drive Train 267
800
Motor power rating (kW)
700
600
Required by gradeability
500
400
300
Required by acceleration
200
100
2
3
4
5
6
7
8
Speed extended speed ratio, x
FIGURE 7.17 Motor power required by acceleration and gradeability along with extended
speed ratios.
is determined by gradeability performance. This means that the power rating
designed to meet the gradeability will naturally meet the acceleration requirement. In engineering design, trade-offs may need to be made between motor
power rating and system complexity to design an appropriate motor extended
speed ratio.
7.3.3
Power Rating Design of the Engine/Generator
As discussed in Chapter 5, the engine/generator in a series hybrid drive train
is used to supply steady-state power in order to prevent the PPS from being
discharged completely. In the design of the engine/generator, two driving
conditions should be considered: driving for a long time at constant speed,
such as highway driving and off-road driving on a soft road, and driving
with a frequent stop–go pattern, such as driving in cities. With the first driving
pattern, the vehicle should not rely on the PPS to support the operation at high
speeds, for example, 130 km/h for on-road vehicles and 60 km/h for crosscountry driving for off-road vehicles. The engine/generator should be able
to produce sufficient power to support vehicle speed. For a frequent stop–
go driving pattern, the engine/generator should produce sufficient power
to maintain the energy store of the PPS at a certain level, so that enough
power can be drawn to support vehicle acceleration and hill climbing. As
mentioned above, the energy consumption of the PPS is closely related to the
control strategy.
268
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
40
M = 1500 kg, fr = 0.01
CD = 0.3, Af = 2.0 m2
Load power (kW)
30
20
10
0
0
20
40
100 120
60
80
Vehicle speed (km/h)
140
160
FIGURE 7.18 Load power of a 1500-kg passenger car at constant speed.
At a constant speed and on a flat road, the power output from the power
source (engine/generator and/or the PPS) can be expressed as
Pe/g
V
1
2
Mg fr + ρa CD Af V (kW),
=
1000ηt ηm
2
(7.7)
where ηt and ηm are the efficiencies of transmission and traction motor,
respectively. Figure 7.18 shows an example of the load power (not including ηt and ηm , curve versus vehicle speed) for a 1500-kg passenger car. It
indicates that the power demand at constant speed is much less than that
for acceleration (refer to Figure 7.15). In this example, about 35 kW (including
losses in the transmission and traction motor, ηt = 0.9, ηm = 0.8, for example)
is needed at a constant speed of 130 km/h.
When the vehicle is driving in a stop-and-go pattern in urban areas, the
power that the engine/generator produces should be equal to or slightly
greater than the average load power in order to maintain balanced PPS energy
storage. The average load power can be expressed as
Pave =
1
T
0
T
1 T
1
dV
Mg fr + ρa CD Af V 2 V dt +
δM
dt,
2
T 0
dt
(7.8)
where δ is the vehicle rotational inertia factor (refer to Chapter 2) and dV/dt
is the acceleration of the vehicle. The first term in Equation 7.8 is the average power that is consumed to overcome the tire rolling resistance and
aerodynamic drag. The second term is the average power consumed in acceleration and deceleration. When the vehicle has the ability to recover all
the kinetic energy of the vehicle, the average power consumed in acceleration and deceleration is zero. Otherwise, it will be greater than zero, as
shown in Figure 7.19.
In the design of an engine/generator system, the power capability should
be greater than, or at least not less than, the power that is needed to support
Design Principle of Series (Electrical Coupling) Hybrid Electric Drive Train 269
Power in acceleration and
deceleration
Average power without
regenerative braking
Average power with
partial regenerative braking
0
Average power with
full regenerative braking
FIGURE 7.19 Average power consumed in acceleration and deceleration with full, partial, and
zero regenerative braking.
the vehicle driving at a constant speed (highway driving) and at average
power when driving in urban areas. In actual design, some typical urban
drive cycles must be used to predict the average power of the vehicle, as
shown in Figure 7.20.
0
30
0
–20
0
200
Average power with full
regenerative braking
400
600
1000 1200 1400
800
Time (s)
FTP 75 urban driving cycle
Speed (km/h)
(c)
Power (kW)
0
40
Average power with zero
regenerative braking
10
–10
20
50
Instantaneous
power
Average power with
zero regenerative
braking
0
Average power with full
regenerative braking
–20
–40
0
100
200
300
400 500 600
Time (s)
FTP 75 highway driving cycle
700
800
(d)
150
100
50
0
100
Instantaneous
power
Average power with
zero regenerative
braking
50
0
–50
–100
0
400
300
Time (s)
US06 driving cycle
200
150
20
100
50
0
40
500
600
Average power with
zero regenerative
braking
Instantaneous
power
0
–20
Average power with full
regenerative braking
100
Speed (km/h)
Power (kW)
20
Instantaneous
power
100
Power (kW)
Speed (km/h)
50
Speed (km/h)
(b)
100
Power (kW)
(a)
–40
0
Average power with full
regenerative braking
100
400
300
Time (s)
ECE driving cycle
200
500
600
FIGURE 7.20 Instantaneous power and average power with full and zero regenerative braking
in typical drive cycles. (a) FTP75 urban, (b) FTP75 highway, (c) US06, and (d) ECE-15.
270
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Engine torque
with full throttle
Engine specific
fuel consumption
a
b
el c
Fu
on
pti
sum
on
inc
s
rea
Engine/generator (torque)
Generator
ing
Engine/generator (rpm)
FIGURE 7.21 Operating point of the engine/generator.
In the engine/generator size design, the operating point at which the
engine/generator produces the above power should be determined. In fact,
there are two possible designs. One approach is to design the engine operating
point at its most efficient point as shown by point a in Figure 7.21. At this operating point, the engine produces the needed power as discussed above. This
design will lead to a somewhat larger engine, since its maximum power will
not be used most of the time. This design has the advantage of more power
being available for special situations. For instance, when the PPS is completely
discharged or has failed, the engine/generator can be operated at a higher
power (point b) to ensure that the vehicle performance has not suffered too
much. The larger engine power can also be used to quickly charge the PPS.
Shifting of the operating point from a to b, as shown in Figure 7.21, will cause
the bus voltage to increase, due to the increase in the speed. With a properly
designed traction motor control, higher voltage will not affect the operation
of the traction motor. On the contrary, the higher DC bus voltage will enable
the motor to produce higher power.
Another design approach is to design the operating point at point b, that is,
close to the engine maximum power, to meet the acceleration and gradeability
requirements as discussed above. This design will lead to a smaller engine.
However, its operating efficiency is somewhat lower than the former design
with no excess power to support the vehicle.
7.3.4
Design of PPS
The PPS must be capable of delivering sufficient power to the traction motor
at any time. At the same time, the PPS must store sufficient energy to avoid
failure of power delivery due to too-deep discharging.
Design Principle of Series (Electrical Coupling) Hybrid Electric Drive Train 271
7.3.4.1
Power Capacity of PPS
To fully utilize traction motor power capacity, the total power of the
engine/generator and PPS should be greater than, or at least equal to, the
rated maximum power of the electric motor. Thus, the power capacity of
the PPS can be expressed as
Ppps ≥
Pm,max
− Pe/g ,
ηm
(7.9)
where Pm,max is the maximum rated power of the motor, ηm is the efficiency
of the motor, and Pe/g is the power of the engine/generator system at its
designed operating point.
7.3.4.2
Energy Capacity of PPS
In some driving conditions, a frequent accelerating/decelerating driving pattern would result in a low SOC in the PPS, thus losing its delivery power. To
properly determine the energy capacity of the PPS, the energy variations in
some typical drive cycles must be known. The energy variation in the PPS
can be expressed as
T
Ppps dt,
(7.10)
ΔE =
0
Energy variation in
the PPS (kWh)
where Ppps is the power of the PPS. Positive Ppps represents charging power
and negative Ppps represents discharging power. It is obvious that the energy
variation in the PPS is closely associated to the control strategy. Figure 7.22
shows an example in which the energy changes in the PPS vary with driving
time in the FTP75 urban driving cycle with the maximum SOC control strategy. Figure 7.22 also shows the maximum amount of energy changes, ΔEmax ,
in the whole drive cycle, if the SOC of the PPS is allowed in the operating
range between SOCtop and SOCbott . The whole energy capacity of the PPS
can be determined by Equation 7.11. The operating range of the SOC of the
PPS depends on the operating characteristics of the PPS. For example, for efficiency reasons, chemical batteries would have an optimal operating range in
0.1
DEmax
0
–0.1
0
200
400
600
800
Time (s)
1000
1200
1400
FIGURE 7.22 Energy variations in an FTP75 urban drive cycle with Max. SOC control strategy.
272
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
the middle (0.4–0.7), and for limited voltage variation reasons, ultracapacitors
would only have a very limited energy change range (0.8–1.0).
Ecap =
7.4
ΔEmax
.
SOCtop − SOCbott
(7.11)
Design Example
Design specification
Parameters
Vehicle total mass: 1500 kg
Rolling resistance coefficient: 0.01
Aerodynamic drag coefficient: 0.3
Front area: 2.0 m2
Transmission efficiency (single gear): 0.9
Performance specification
Acceleration time (from 0 to 100 km/h): 10 ± 1 s
Maximum gradeability: >30% at low speed and >5% at 100 km/h
Maximum speed: 160 km/h
7.4.1
Design of Traction Motor Size
Using Equation 7.5 and assuming the motor drive has a speed ratio of x = 4,
the motor drive power rating can be obtained as 82.5 kW for the specified
acceleration time of 10 s from zero to 100 km/h. Figure 7.23 shows the speed–
torque and speed–power profiles of the traction motor.
7.4.2
Design of the Gear Ratio
The gear ratio is designed so that the vehicle reaches its maximum speed at
the motor maximum speed, that is,
ig =
πnm,max r
,
30Vmax
(7.12)
where nm,max is the maximum motor rpm, Vmax is the maximum speed of the
vehicle in m/s, and r is the radius of the tire. Suppose nm,max = 5000 rpm and
Vmax = 44.4 m/s (160 km/h or 100 mph), and r = 0.2794 m (11 in.); ig = 3.29
is obtained.
Design Principle of Series (Electrical Coupling) Hybrid Electric Drive Train 273
700
140
Torque
x=4
120
500
100
Power
400
80
300
60
200
40
100
20
0
0
Motor power (kW)
Motor torque (Nm)
600
0
500 1000 1500 2000 2500 3000 3500 4000 4500 5000
Motor speed (rpm)
FIGURE 7.23 Characteristics of traction motor versus motor rpm.
7.4.3 Verification of Acceleration Performance
Based on the torque–speed profile of the traction motor, gear ratio, and vehicle parameters, and using the calculation method discussed in Chapters 2
and 4, vehicle acceleration performance (acceleration time and distance versus vehicle speed) can be obtained as shown in Figure 7.24. If the acceleration
time obtained does not meet the design specification, the motor power rating
should be redesigned.
30
300
250
Distance
20
200
15
150
10
100
Time
5
0
0
20
50
40
60
80
100
Vehicle speed (km/h)
120
FIGURE 7.24 Acceleration time and distance versus vehicle speed.
140
Acceleration distance (m)
Acceleration time (s)
25
274
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
7.4.4 Verification of Gradeability
(a)
8
Tractive effort and resistance (kN)
Using the motor torque–speed profile, gear ratio, and vehicle parameters, and
the equations described in Chapters 2 and 4, the tractive effort and resistance
versus vehicle speed can be calculated as shown in Figure 7.25a. Further,
the gradeability of the vehicle can be calculated as shown in Figure 7.25b.
Figure 7.25 indicates that the gradeability calculated is much greater than that
specified in the design specification. This result implies that for a passenger
car, the power needed for acceleration performance is usually larger than
that needed for gradeability; the former determines the power rating of the
traction motor.
7
Tractive effort
a = 25° (46.6%)
6
Resistance (rolling +
aerodynamic +
hill climbing)
a = 20° (36.4%)
5
a = 15° (26.6%)
4
a = 10° (17.6%)
3
a = 5° (8.75%)
2
1
0
a = 0° (0%)
0
20
40
100
60
80
Vehicle speed (km/h)
120
140
160
0
20
40
60
80
100
Vehicle speed (km/h)
120
140
160
(b) 55
50
45
Gradeability (%)
40
35
30
25
20
15
10
5
0
FIGURE 7.25 Tractive effort, resistance, and gradeability of the vehicle versus speed: (a) tractive
effort and resistance and (b) gradeability.
Design Principle of Series (Electrical Coupling) Hybrid Electric Drive Train 275
120
100
Engine power (kW)
On 5% grade road
80
On flat road
60
40
32.5
20
0
0
20
40
60
80
100 120
Vehicle speed (km/h)
140
160
180
FIGURE 7.26 Engine power versus vehicle constant speed on a flat road and a 5% grade road.
7.4.5
Design of Engine/Generator Size
The power rating of the engine/generator is designed to be capable of
supporting the vehicle at a regular highway speed (130 km/h or 81 mph)
on a flat road. Figure 7.26 shows that the engine power needed at 130 km/h
or 81 mph is 32.5 kW, in which energy losses in transmission (90% of efficiency), motor drive (85% of efficiency), and generator (90% of efficiency)
are involved. Figure 7.26 also indicates that 32.5 kW of engine power can
be capable of supporting the vehicle driving at 78 km/h (49 mph) on a 5%
grade road.
Another consideration in the design of the power rating of the
engine/generator is the average power when driving with some typical stopand-go driving patterns as illustrated in Figure 7.20. The typical data in these
drive cycles are listed in Table 7.1.
TABLE 7.1
Typical Data of the Different Drive Cycles
Max.
Speed
(km/h)
FTP75 urban
FTP75 highway
US06
ECE-1
86.4
97.7
128
120
Average
Speed
(km/h)
Average Power with Full
Regen. Braking
(kW)
Average Power without
Regen. Braking
(kW)
27.9
79.6
77.4
49.8
3.76
12.6
18.3
7.89
4.97
14.1
23.0
9.32
276
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Compared with the power needed in Figure 7.25, the average power in these
driving cycles is smaller. Hence, 32.5 kW of engine power can meet the power
requirement in these drive cycles. Figure 7.27 shows the engine characteristics.
The engine would need to supply additional power to support the continuous
nontraction loads, such as lights, entertainment, ventilation, air conditioning,
power steering, brake boosting, and so on. In summary, the engine needs to
(a)
Engine power (kW)
7 14 24 28 35 42 49 54 63
200
Specific fuel
Operating consumption
points
(g/kWh)
(efficiency, %)
180
Engine torque (Nm)
160
a
250 (34.3%)
260 (33%)
270 (31.8%)
280 (30.6%)
140
120
b
100
)
27.7%
310 (
80
%)
(24.5
350
)
21.4%
400 (
60
7.1%)
500 (1 0 (14.3%)
60
)
700 (10.7% (8.6%)
1000
40
20
0
0
1000
2000
3000
4000
Engine speed (rpm)
Engine power (kM)
(b)
4.4
120
8.9 13.3 17.8 22.2 26.7 13.1 35.6 40.0
Operating
points
Specific fuel
consumption
(g/kWh)
(efficiency, %)
100
Engine torque (Nm)
5000
250 (34.3%)
260 (33%)
270 (31.8%)
280 (30.6%)
80
60
7.7%)
310 (2
%)
40
350
(24.5
400
7.1%)
500 (1 (14.3%)
600 700 (10.7%)
1000 (8.6%)
20
0
%)
(21.4
0
1000
2000
3000
4000
Engine speed (rpm)
5000
FIGURE 7.27 Engine characteristics and operating points: (a) operating with best efficiency and
(b) operating with close maximum power.
Design Principle of Series (Electrical Coupling) Hybrid Electric Drive Train 277
produce about 35 kW of power to support the vehicle at 130 km/h on a flat
road, without the need for power assistance from the PPS. This power can
sufficiently meet the average power requirement for the stop-and-go driving
pattern in urban areas.
Figure 7.27a shows the engine fuel consumption map and the minimum fuel
consumption operating point (point a) at which 35 kW of power is produced.
It can be seen that the maximum engine power is about 63 kW at point b.
Another design of the engine power is shown in Figure 7.27b, in which the
engine operating point is designed close to its maximum power to produce
the power demand of 35 kW. The engine size in this design is smaller than
the former design, but the fuel consumption is higher than the former design
at 35 kW power level. As discussed above, this power is for 130 km/h of
constant speed on a flat road. At lower speeds or driving in urban areas in
which the average load power is much less, the latter design may not show
higher fuel consumption than the former.
7.4.6
Design of the Power Capacity of PPS
The sum of the output power of the engine/generator and PPS should be
greater than, or at least equal to, the input power of the traction motor. That is,
Ppps =
Pmotor
82.5
− Pe/g =
− 32.5 = 64.5 kW,
ηmotor
0.85
(7.13)
where 32.5 kW is the power of the engine/generator for traction.
7.4.7
Design of the Energy Capacity of PPS
The energy capacity of the PPS depends heavily on drive cycle and overall control strategy. In this design, because the power capacity of the
engine/generator is much greater than the average load power (refer to
Figure 7.20), the engine on–off (thermostat) control strategy is considered
to be appropriate.
Figure 7.28 shows the simulation results of the above vehicle with engine
on–off control strategy in the FTP75 urban drive cycle. In the simulation,
regenerative braking is involved (see Chapter 13, Regenerative Braking). In
the control, the allowed maximum energy variation in PPS is 0.5 kWh. Suppose that the PPS is allowed to operate in the SOC range of 0.2. Using batteries
as the PPS, operating in the range of 0.4–0.6 of SOC will have optimal efficiency. Using ultracapacitors, 0.2 variation of SOC will limit the terminal
voltage to 10%. The total storage energy in the PPS can be calculated as
Epps =
0.5
ΔEmax
=
= 2.5 kWh.
ΔSOC
0.2
(7.14)
278
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
100
Vehicle power (kW)
50
0
50
Motor power (kW)
0
–50
40
Engine power (kW)
20
0
50
PPS power (kW)
0
–50
2
1
0
Energy change in PPS (kWh)
0
200
400
600
1000
800
1200
1400
Time (s)
FIGURE 7.28 Simulation results in an FTP75 urban drive cycle.
The weight and volume of the PPS are determined by its power capability or
energy capability, depending on the power and energy density ratings of the
PPS. For batteries, power density is usually the determining factor, whereas
for ultracapacitors, energy density is usually the determining factor. A hybrid
100
50
Vehicle speed (km/h)
0
50
Motor power (kW)
0
–50
40
Engine power (kW)
20
0
50
0
PPS power (kW)
–50
2
1
0
Energy change in PPS (kWh)
0
100
200
300
400
Time (s)
500
600
FIGURE 7.29 Simulation results in an FTP75 highway drive cycle.
700
800
Design Principle of Series (Electrical Coupling) Hybrid Electric Drive Train 279
PPS, both battery and ultracapacitors, would be much smaller and lighter than
using any one of the above. For details, refer to Chapter 12, Energy Storage.
7.4.8
Fuel Consumption
The fuel consumption in various drive cycles can be calculated by simulation. In the simulation in this example, the engine as shown in Figure 7.27b
is used. When the engine is on, its power output is around 20 kW, corresponding to its best fuel efficiency operating point. In the FTP75 urban drive
cycle (Figure 7.28), the vehicle has a fuel economy of 5.57 L/100 km or 42.4
miles per gallon (mpg), and in the FTP highway drive cycle (Figure 7.29),
5.43 L/100 km or 43.5 mpg. It is clear that a hybrid vehicle with a similar performance to a conventional vehicle is much more efficient, especially in the
frequent stop-and-go environment. The main reasons are the high operating
efficiency of the engine and the significant amount of braking energy recovered by regenerative braking. Regenerative braking techniques are described
in Chapter 13.
References
1. C. C. Chan and K. T. Chau, Modern Electric Vehicle Technology, Oxford University
Press, New York, 2001.
2. C. G. Hochgraf, M. J. Ryan, and H. L. Wiegman, “Engine control strategy for a
series hybrid electric vehicle incorporating load-leveling and computer controlled
energy management,” Society of Automotive Engineers (SAE) Journal, Paper No.
960230, Warrendale, PA, 2002.
3. M. Ender and P. Dietrich, “Duty cycle operation as a possibility to enhance the
fuel economy of an SI engine at part load,” Society of Automotive Engineers (SAE)
Journal, Paper No. 960227, Warrendale, PA, 2002.
4. M. Ehsani, Y. Gao, and K. Butler, “Application of electric peaking hybrid (ELPH)
propulsion system to a full size passenger car with simulation design verification,”
IEEE Transactions on Vehicular Technology, 48 (6), November 1999.
5. C. C. Chan, “The state of the art of electric and hybrid, and fuel cell vehicles,”
Proceedings of the IEEE, Special issue on Electric, Hybrid and Fuel Cell Vehicles, Vol. 95,
No. 4, April 2007.
6. Jih-sheng (Jason) Lai and D. J. Nelson, “Energy management power converters
in hybrid hlectric and fuel cell vehicles,” Proceedings of the IEEE, Special issue on
Electric, Hybrid and Fuel Cell Vehicles, Vol. 95, No. 4, April 2007.
8
Parallel (Mechanically Coupled) Hybrid
Electric Drive Train Design
Unlike the series hybrid drive train, the parallel or mechanically coupled
hybrid drive train has features that allow both the engine and the traction
motor to apply their mechanical power in parallel directly to the driven
wheels. As mentioned in Chapter 5, the mechanical coupling has two forms:
torque and speed couplings. When using conventional IC engines as the primary power source, torque coupling is more appropriate since the IC engine
is essentially a torque source.
The major advantages of a torque-coupling parallel configuration over a
series configuration are (1) nonnecessity of a generator, (2) a smaller traction
motor, and (3) only part of the engine power going through multipower conversion. Hence, the overall efficiency can be higher than in the series hybrid.1
However, control of the parallel hybrid drive train may be more complex than
that of the series hybrid drive train, because of the simultaneous mechanical
coupling between the engine and the driven wheels.
There are many possibilities of configurations in a parallel hybrid drive
train, as mentioned in Chapter 5. The design methodology for one configuration may not be applicable to others. Each particular configuration may be
only applicable to the specified operation environment and mission requirement. This chapter will focus on the design methodology of parallel drive
trains with torque coupling, which operates with the electrically peaking principle. That is, the engine supplies its power to meet the base load (operating at
a given constant speed on flat and mild grade roads, or the average of the load
of a stop-and-go driving pattern) and the electric motor supplies the power to
meet the peak load requirement. Other options, such as a mild hybrid drive
train, are discussed in Chapter 11.
8.1 Drive Train Configuration and Design Objectives
The drive train structure of the parallel (torque coupling) hybrid vehicle is
shown in Figure 8.1. The control system of the drive train consists of a vehicle
controller, an engine controller to control the engine power, an electric motor
281
282
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Mechanical
brake
controller
Vehicle
controller
PPS SOC
Motor
control
signal
Engine
control
Engine
controller
Engine
control
Engine speed
throttle position
Operation
command
Peaking
power
source
Motor
controller
Motor
Torque
coupler
Clutch
Transmission
Engine
Vehicle speed
Mechanical connection
Electrical power
Signals
FIGURE 8.1 Configuration of the parallel torque-coupling hybrid drive train.
controller, and, perhaps, a mechanical brake controller and clutch controller.
The vehicle controller is the highest level controller. It receives the operation command from the driver through the accelerator and brake pedals, and
other operating variables of the vehicle and its components, which includes
vehicle speed, engine speed and throttle position, SOC of the PPS, and so on.
By processing all the signals received, based on the embedded drive train
control algorithm, the vehicle controller generates control commands and
sends the commands to the corresponding component controllers. The component controllers control the corresponding components to carry out the
commands coming from the vehicle controller. Since the torque coupler is
uncontrollable, the power flow in the drive train can only be controlled by
controlling the power sources, that is, the engine, traction motor, clutch, and
mechanical brake.
In the drive train design, the important factors are the power of the engine,
motor and PPS and its energy capacity, transmission, and more importantly
the control strategy of the drive train. The design objectives are as follows: (1) satisfying the performance requirements (gradeability, acceleration,
and maximum cruising speed), (2) achieving high overall efficiency whenever possible, (3) maintaining the SOC of PPS at reasonable levels during
driving on highways and in urban areas without the need of charging the
PPS from outside the vehicle, and (4) recovering brake energy as much as
possible.
Parallel (Mechanically Coupled) Hybrid Electric Drive Train Design
283
8.2 Control Strategies
The available operation modes in a parallel torque-coupling hybrid drive
train, as mentioned in Chapter 5, mainly include the following: (1) enginealone traction, (2) electric-alone traction, (3) hybrid traction (engine plus
motor), (4) regenerative braking, and (5) PPS charging from the engine. During operation, proper operation modes should be used so as to meet the
traction torque requirement, achieve high overall efficiency, maintain a reasonable level of SOC of the PPS, and recover braking energy as much as
possible.2–6
The overall control scheme is schematically shown in Figure 8.2. It consists of the vehicle controller, engine controller, electric motor controller, and
mechanical brake controller. The vehicle controller is at the highest level. It
collects data from the driver and all the components, such as the desired
torque from the driver, vehicle speed, SOC of the PPS, engine speed and
throttle position, electric motor speed, and so on. Based on these data, the component characteristics and the preset control strategy, the vehicle controller
gives its control signals to each component controller. Each component controller then controls the operation of the corresponding component to meet
the requirement of the drive train.
Accelerator
pedal
signal
Traction
mode
Brake
pedal
signal
Vehicle speed
PPS SOC
Vehicle controller
Engine power
command
Motor
Mechanical power
power
command
command
Engine controller
Motor
controller
Engine
Electric
motor
Engine
power
Motoring
power
Transmission
Braking
mode
Mechanical brake
controller
Regenerative
braking power
+
+
Wheels
Wheels
Mechanical
brake
Mechanical
braking power
FIGURE 8.2 Overall control scheme of the parallel torque-coupling hybrid drive train.
284
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
The vehicle controller plays a central role in the operation of the drive train.
The vehicle controller should fulfill various operation modes, according to
the data collected from components and the driver’s command, and should
give the correct control command to each component controller. Hence, the
control strategy in the vehicle controller is the key in the success of the drive
train operation.
8.2.1
Max. SOC-of-PPS Control Strategy
When a vehicle is operating in a stop-and-go driving pattern, the PPS must
deliver its power to the drive train frequently. Consequently, the PPS tends
to be discharged quickly. In this case, maintaining a high SOC in the PPS is
necessary to ensure that the PPS is capable of delivering sufficient power to
the drive train to support the vehicle’s frequent acceleration. The basic rules
in this control strategy are using the engine as the primary power source
as much as possible and charging the PPS whenever the engine has excess
power over that required for propulsion, without the PPS SOC exceeding its
full charge limit.3
The maximum control strategy can be explained by Figure 8.3. In this figure, the maximum power curves for hybrid traction (engine plus electric
motor), engine-alone traction, and electric motor-alone traction and regenerative braking are plotted against vehicle speed. Power demands in different
conditions are also plotted, represented by points A, B, C, and D.
1
Traction power
A
2
Pm
PL
a
Pe
Braking power
0
b
Pmc
B
Pe PL
Veb
PL
Pmb
Pmb PL
C
Pmf d
3
4
1: Maximum power with hybrid mode
2: Maximum power with electric alone traction
3: Engine power on its optimum operating line
4: Engine power with partial load
5: Maximum generative power of electric motor
Vehicle PL : Load power, traction or braking
speed Pe : Engine power
Pm: Motor traction power
Pmb: Motor braking power
Pmc: PPS charging power
5
Pmf : Mechanical braking power
Veb : Vehicle speed corresponding to the engine
minimum rpm
D
FIGURE 8.3 Demonstration of various operating modes based on power demand.
Parallel (Mechanically Coupled) Hybrid Electric Drive Train Design
285
The operation modes of the drive train are explained below.
Motor-alone propelling mode: When the vehicle speed is less than a preset
value Veb , which is considered to be the bottom line of the vehicle speed
below which the engine cannot operate stably, or operates with high fuel
consumption or high emissions, the electric motor alone propels the vehicle. Meanwhile the engine is shut down or idles with the clutch open. The
engine power, electric traction power, and PPS discharge power can be
written as
Pe = 0,
(8.1)
Pm = PL ,
(8.2)
Ppps-d =
Pm
.
ηm
(8.3)
where Pe is the engine power output, PL is the propelling power commanded
by the driver, from the accelerator pedal, Pm is the power output of the electric
motor, Ppps-d is the PPS discharge power, and ηm is the motor efficiency.
Hybrid propelling mode: When the commanded propelling power, PL , by the
driver, represented by point A in Figure 8.3, is greater than the power that
the engine can produce, both the engine and electric motor must deliver their
power to the driven wheels at the same time. In this case, engine operation is
set on its optimum operation line (point a) by controlling the engine throttle to
produce power Pe . The remaining power demand is supplied by the electric
motor. The motor power output and PPS discharge power are
Pm = PL − Pe ,
Ppps-d =
Pm
.
ηm
(8.4)
(8.5)
PPS charge mode: When the commanded propelling power, PL , represented
by point B in Figure 8.3, is less than the power that the engine can produce
while operating on its optimum operation line, and the SOC of the PPS is
below its top line, the engine is operated on its optimum operating line (point
b), producing its power Pe . In this case, the electric motor is controlled by its
controller to function as a generator, powered by the remaining power of the
engine. The input power to the electric motor and PPS charge power are
Pm = (Pe − PL )ηt,e,m ,
Ppps-c = Pm ηm ,
(8.6)
(8.7)
where ηt,e,m is the transmission efficiency from the engine to the electric motor.
Engine-alone propelling mode: When the commanded propelling power, represented by point B in Figure 8.3, is less than the power that the engine can
286
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
produce while operating on its optimum operation line, and the SOC of the
PPS has reached its top line, the engine-alone propelling mode is used. In
this case, the electric system is shut down, and the engine is operated to
supply its power that meets the load power demand. The power output
curve of the engine with a partial load is represented by the dashed line
in Figure 8.3. The engine power, electric power, and battery power can be
expressed by
Pe = PL ,
(8.8)
Pm = 0,
(8.9)
Ppps = 0.
(8.10)
Regenerative-alone brake mode: When the vehicle experiences braking and
the demanded braking power is less than the maximum regenerative braking
power that the electric system can supply (as shown in Figure 8.3 by point
C), the electric motor is controlled to function as a generator to produce its
braking power that equals the commanded braking power. In this case, the
engine is shut down or set idling. The electric power output from the motor
and PPS charging power are
Pmb = PL ηm ,
Ppps-c = Pmb .
(8.11)
(8.12)
Hybrid braking mode: When the demanded braking power is greater than
the maximum regenerative braking power that the electric system can supply (as shown in Figure 8.3 by point D), the mechanical brake must be applied.
In this case, the electric motor should be controlled to produce its maximum regenerative braking power, and the mechanical brake system handles
the remaining portion. The motor output power, PPS charging power, and
mechanical braking power are
Pmb = Pmb,max ηm ,
Ppps-c = Pmb ,
Pmf = PL − Pmb .
(8.13)
(8.14)
(8.15)
It should be noted that for good braking performance, the braking forces on
the front and rear wheels should be proportional to their normal load on
the wheels. Thus, braking power control will not be exactly that mentioned
above (for more details, see Chapter 13, Regenerative Braking). The control
flowchart of the Max. SOC-of-PPS is illustrated in Figure 8.4.
The major objective of this control strategy is using the engine as the vehicle
primary mover as much as possible, with no pure electric traction, when the
vehicle speed is higher than a present value. This control strategy minimizes
Parallel (Mechanically Coupled) Hybrid Electric Drive Train Design
Maximum
motor power
Pm-max
Traction power
command, Ptc
Braking power
command, Pbc
Traction?
No
Vehicle speed, V
V < Ver?
Yes
No
Pe-opt
If Ptc > Pe-opt
If Pbc > Pm-max
Yes
Yes
Yes
287
Electric alone
traction mode
No
Regenerative
alone braking
mode
Hybrid
braking mode
Pe-pot–The engine power while operating
on its optimum operating line
Hybrid traction
mode
No
SOC of PPS
If SOC < SOCtop
No
Engine alone
traction mode
Yes
PPS charge
mode
FIGURE 8.4 Flowchart of Max. SOC-of-PPS control strategy.
the part of engine energy that cycles through the electric motor and the PPS.
This may reduce the engine energy transmission losses. However, when the
PPS is fully charged and the vehicle load is small, the engine will be throttled
down to meet the small load power. In this case, the engine will suffer a low
operating efficiency.
8.2.2
Engine On–Off (Thermostat) Control Strategy
When the vehicle operates in a state in which the load power is less than
the power that the engine produces with optimal operating efficiency, and
the PPS is fully charged, the Max. PPS SOC control strategy will force the
engine to operate away from its optimal operating region. Consequently, the
overall efficiency of the vehicle would suffer. In this situation, the engine
on–off (thermostat) control strategy may be used. In the engine on–off control
strategy, the operation of the engine is controlled by the SOC of the PPS, as
shown in Figure 8.5.
During the engine on period, the control is Max. PPS SOC strategy, in which
the engine is always operated on its optimal curve. When the SOC of the PPS
reaches its top line, the engine is turned off and the vehicle is propelled only
by the electric motor. When the SOC of the PPS reaches its bottom line, the
engine is turned on and the control again goes into Max. SOC-of-PPS.
It can be seen that this control strategy uses the electric motor as its primary
power sources. The engine either operates in its optimal region or stops. Thus,
the average operating efficiency of the engine is optimized. However, contrary
288
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
PPS SOC
PPS SOC top line
Engine operation
PPS SOC bottom line
On
Off
FIGURE 8.5 Illustration of engine on–off control strategy.
to the Max. SOC-of-PPS control strategy, the engine energy that goes through
the electric motor and the PPS is also maximized, which may cause more
energy losses in energy conversion.
It should be noted that with this control strategy, the electric motor has to
have sufficient power to meet the vehicle peaking power during the engine
off periods.
8.2.3
Constrained Engine On–Off Control Strategy
The constrained engine on–off control strategy is the trade-off between the
Max. SOC-of-PPS and engine on–off control strategies. The principle of this
control strategy is to add the engine on and off operation in the Max. SOCof-PPS control strategy. The control will be exactly the Max. SOC-of-PPS,
when the vehicle speed is less than Veb , and the commanded traction powers
are at points A, C, and D as shown in Figure 8.3. However, when the commanded traction power is at point B shown in Figure 8.3, that is, less than
the engine power with optimal efficiency, the engine can be operated with
optimal throttle, partial throttle, or turned off, depending on the SOC of the
PPS. This control strategy is explained by using the diagram of Figure 8.6.
The engine torque or power is divided into three special regions—the large
torque area, TL, the medium torque area, TM, and the small torque area, TS—
as shown in Figure 8.6a. These three torque areas are separated by the torque
curves Te-l , Te-m , and Te-s . These three curves may be generated by three
special throttle openings. In Figure 8.6a, the isofuel consumption curves are
also plotted. Similarly, the SOC of the PPS is also divided into three regions—
high, medium, and low—as shown in Figure 8.6b. The engine control is based
on the real-time commanded traction torque TL and the SOC of the PPS. The
suggested control strategy is illustrated in Figure 8.7.
When the commanded traction torque is in the TL area, as shown by point
A in Figure 8.6a, and if the SOC of the PPS is in the medium or high region, the
engine is controlled to produce its torque equal to the commanded traction
Parallel (Mechanically Coupled) Hybrid Electric Drive Train Design
289
Veb
C
Large
engine
torque
area,
Media TL
engine
torque
area,
TM
Small
engine
torque
area,
TS
Te-u
Te-m
(b)
High
PPS SOC
Tchgll
Tchgmm
c
Maximum torque
with fuel throttle
Iso-fuel
consumption
curve
Tchgsl
B
Tchgsm
rease
Fuel consumptio
n inc
Engine turn off area, E-off
Engine torque
a
A
b
Tchgml
(a)
Medium
Low
Te-b
Engine speed
Engine torque regions
0
PPS SOC regions
FIGURE 8.6 Illustration of the constrained engine on and off control strategy: (a) engine
operation regions and (b) PPS SOC regions.
torque and no additional torque is produced to charge the PPS. However, if
the SOC of the PPS is in the low region, the engine needs to produce more
torque to charge the PPS. In this case, the engine is controlled to operate on its
optimal efficiency line, represented by point a. The charging torque is Tchgll
as shown in Figure 8.6a.
PPS SOC
Commanded
torque, TL
Low
Medium
High
In small area
(point C)
Te = Tc
Te = Tb
Tchgsl = Tb – Tc Tchgsm = Tc – Tc
Te = 0
Tchgsh = 0
In medium area
(point B)
Te = Ta
Te = Tb
Tchgml = Ta – TB Tchgmm = Tb – TB
Te = TB
Tchgmh = 0
In large area
(point A)
Te = Ta
Tchgll = Ta – TA
Te = TA
Tchglh = 0
Te = TA
Tchglh = 0
TA, TB, TC ---
Commanded traction torques in large, medium, and low torque areas,
corresponding to point A, B, and C in Figure 8.6(a)
Ta, Tb, Tc --- Torques that engine is controlled to produce, corresponding to point a, b, and c
Tchgxx --- PPS charging torque, footnote xx = Ih--- large torque, low SOC, and so on
Te --- Engine torque
FIGURE 8.7 Engine torque control strategy with different commanded traction torque and PPS
SOC.
290
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
When the commanded traction torque is in the TM area, as shown by point
B in Figure 8.6a, and if the SOC of the PPS is in the high region, the engine is
controlled to produce its torque equal to the commanded traction torque with
no additional engine torque to charge the PPS. Otherwise if the SOC of the PPS
is in the medium region, the engine torque is controlled on the top boundary
line of this area, as shown by point b in Figure 8.6a. The PPS charging torque
is Tchgmm . However, if the SOC of the PPS is in the low region, and in order to
quickly bring the SOC of the PPS to the medium level, the engine is controlled
to operate on the optimal efficiency line as shown by point a in Figure 8.6a.
The PPS charging torque is Tchgml .
When the commanded traction torque is in the TS area as shown by point
C in Figure 8.6a, and if the SOC of the PPS is in the high region, the engine
is shut down and the electric motor alone propels the vehicle. If the SOC of
the PPS is in the medium region, the engine is controlled to operate on the
upper boundary line of this area, as shown by point c in Figure 8.6a. The
PPS charging torque is Tchgsm . However, if the SOC of the PPS is in the low
region, the engine is controlled to operate at point a, as shown in Figure 8.6a,
to quickly bring the SOC of the PPS to the medium region. The PPS charging
torque is Tchgsl .
Figure 8.7 summarizes engine control in all the commanded engine traction
torque areas and SOC regions of the PPS.
8.2.4
Fuzzy Logic Control Technique
The above engine and electric motor control strategy can be further developed by using fuzzy logic control methods, based on the commanded traction
torque and SOC of the PPS. In fuzzy logic language, input variables of the
commanded traction torque and SOC of the PPS are described by linguistic
values as high (H), medium (M), and low (L). The output variables of the
commanded engine torque are described as high (H), medium (M), low (L),
and a crisp value zero (Z). Similarly, the electric motor torques are described
as negative high (NH), negative medium (NM), negative low (NL), zero (Z,
a crisp value), positive low (PL), positive medium (PM), and positive high
(PH). Positive torque is for traction and negative is for generating. The control
rules are very similar to those described in Figure 8.7. The block diagram of
the fuzzy logic control is shown in Figure 8.8.
In Figure 8.8, only the engine torque is determined by the fuzzy logic control
rules, based on the SOC of the PPS and the commanded traction torque. The
motor torque will be obtained from the commanded traction torque and the
engine torque is obtained from fuzzy logic, that is,
Tm = Tct − Te .
Since this control strategy is for the operation mode in which the commanded traction torque, Tct , is smaller than the maximum torque that the
291
Parallel (Mechanically Coupled) Hybrid Electric Drive Train Design
Rule base
Tct : Commanded traction torque
Te : Engine torque
Tm : Motor torque
mTct : Membership of Tct
mSOC : Membership of PPS SOC
mTe : Membership of Te
SOC
Tct Te
L
M
H
PPS SOC
Tct
mSOC
Fuzzification
mTct
L
M
H
H
M
H
H
H
H
Z
M
H
mTe
Inference
(Control rules)
Defuzzification
Te
Tm = Tct – Te
Tm
FIGURE 8.8 Block diagram of fuzzy logic control.
engine can produce with its optimal efficiency curve, as shown by point B
in Figure 8.3, the commanded traction torque, Tct , and the engine torque, Te ,
have the same boundary, that is, from zero to maximum. Thus Tct and Te
have the same membership function. A possible one is shown in Figure 8.9a,
and a possible membership function of the SOC of PPS is shown in Figure
8.9b. The standard procedure to solve fuzzy logic control problems will not
be discussed here. Readers should refer the associated references.7–11
It has to be noted that the threshold or fuzzy logic control strategies discussed above are based on the experience and knowledge about the drive
train operation. Real-time operation is the most reliable way of obtaining this
(a)
Medium
Low
High
mTct and mTe
1
0
0
10
20
30
40
50
60
70
80
90 100%
Tct and Te
(b)
Medium
Low
High
mSOC
1
0
0
10
20
30
40
50
60
70
80
90 100%
SOC
FIGURE 8.9 Membership functions of (a) Tct , Te and (b) SOC.
292
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
experience and knowledge. However, it requires a long time and involves high
cost. Another effective way is by using simulation technologies to tune the
control parameters iteratively until achieving optimal parameters for the specified operation environment. This work may be time consuming. It is easier
to use off-line optimization technologies than iteratively running rules/fuzzy
logic-based simulation to obtain the optimal control parameters. Dynamic
programming is one of these techniques.12,13
8.2.5
Dynamic Programming Technique
The basic idea of the control algorithm using the dynamic programming technique considers the dynamic nature of the HEV system when performing the
optimization. Furthermore, the optimization is with respect to the time horizon, rather than for an instant in time,12,13,14 that is, for the whole driving
cycle.
Contrary to the rule-based fuzzy logic control algorithm, the dynamic optimization approach usually relies on a drive train model to compute the best
control strategy. The model can be either analytical or numerical. For a given
driving cycle, the optimal operating strategy to achieve the best fuel economy
can be obtained by solving a dynamic optimization problem. The problem
formulation is described in the following.12,13
In the discrete-time format, a model of the HEV can be expressed as
x(k + 1) = f (x(k), u(k)),
(8.16)
where u(k) is the vector of control variables such as engine throttle opening,
desired motor torque, gear shift command of the transmission, and so on and
x(k) is the vector of state variables of the system, which is the response to
control variables u(k). The goals of the optimization are to find the optimal
control input u(k), to minimize the total fuel consumption, or the combination
of total consumption and total emission over a given driving cycle.14 The total
fuel consumption or the combined fuel consumption and emission is defined
as a cost function to be minimized. In the following expression, only the total
fuel consumption is included in the cost function:
J = Fuel =
N−1
L(x(k), u(k)),
(8.17)
k=0
where N is the time length of the driving cycle and L is the instantaneous fuel
consumption rate, which is a function of the system state x and input u.
In the minimizing procedure of Equation 8.17, some constraints have to be
imposed to ensure that all the operating parameters are within their valid
Parallel (Mechanically Coupled) Hybrid Electric Drive Train Design
293
ranges. These constraints include
ωe-min ≤ ωe ≤ ωe-max ,
(8.18)
0 ≤ Te ≤ Te-max ,
(8.19)
0 ≤ ωm ≤ ωm-max ,
(8.20)
Tm-min ≤ Tm ≤ Tm-max ,
(8.21)
SOCmin ≤ SOC ≤ SOCmax ,
(8.22)
where ωe is the engine angular velocity; ωe-min and ωe-max are the specified
minimum and maximum engine angular velocities, respectively; Te is the
engine torque, which must be greater than or equal to zero and smaller than or
equal to its maximum torque at the corresponding angular velocity; ωm is the
motor angular velocity, which is defined in the range of zero to its maximum;
Tm is the motor torque; Tm-min is the minimum motor torque, which may be
the maximum generating torque (negative); Tm-max is the maximum motor
torque in traction; and SOC is the SOC of the PPS, which is constrained to
the range of its bottom SOCmin and top SOCmax levels. In order to sustain the
charge of the PPS (PPS SOC at the end of the driving cycle is not lower than
that at the beginning of the driving cycle), a final state constraint for the SOC
of the PPS should be imposed. Thus a soft terminal constraint on PPS SOC
(quadratic penalty function) is added to the cost function as follows12,13 :
J = Fuel =
N−1
L(x(k), u(k)) + G(x(N)),
(8.23)
k=0
where G(x(N)) = α(SOC(N) − SOCf )2 represents the penalty associated with
the error in the SOC at the end of the driving cycle; SOCf is the desired SOC
at the end of the driving cycle, which may be set equal to the SOC at the
beginning of the driving cycle, and α is a weight factor.
Dynamic programming is well known for requiring computations that grow
exponentially with the number of states.12,15 Therefore, a simplified vehicle
and components model is necessary. The engine, electric motor, PPS, transmission, and so on may need to be reduced to static models with look-up
tables for I/O mapping and efficiencies.
Standard procedures to solve the above optimization problem based on
Bellman’s principle of optimality are given by Lin et al.12 and Betsekas.15 The
dynamic programming algorithm is presented as follows:
Step N − 1:
∗
(x(N − 1)) = min [L(x(N − 1), u(N − 1)) + G(x(N))] .
JN−1
u(N−1)
(8.24)
294
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Step k, for 0 ≤ k < N − 1:
∗
Jk∗ (x(k)) = min L(x(k), u(k)) + Jk+1
(x(k + 1)) .
(8.25)
u(k)
The recursive equation is solved backward from step N − 1 to zero. Each
minimization in a given driving cycle is performed subject to the constraints
imposed by Equations 8.18 through 8.22.
The standard method to solve a dynamic programming problem numerically is to use quantization and interpolation.12,13,16 The states and control
values are first quantized into finite grids. At each step of the dynamic programming algorithm, the function Jk (x(k)) is evaluated only at the grid points.
If the next states, x(k + 1), do not fall exactly on to a quantized value, function
∗ (x(k + 1)) in Equation 8.25
interpolation is used to determine the value of Jk+1
as well as G(x(N)) in Equation 8.24.
The dynamic programming procedure produces an optimal, time-varying,
state-feedback control policy that is stored in a table for each of the quantized states and time stages, that is, u∗ (x(k), k). This function is then used as a
state-feedback controller in the simulations. It should be noted that dynamic
programming creates a family of optimal paths for all possible initial conditions. In this application, once the initial SOC is given, the optimal policy
will find an optimal way of bringing the final SOC back to the terminal value
(SOCf ) while achieving the minimal fuel consumption.
Although the dynamic programming approach provides an optimal solution for minimizing the fuel consumption, the resulting control policy is not
implementable in real driving conditions because the optimal policy requires
120
Hybrid
Engine only
Recharging
Motor only
Power demand (kW)
100
Region B
Region A
80
60
40
Region C
20
0
0
50
100
150
200
Trans input speed (rad/s)
250
300
FIGURE 8.10 Operating points of dynamic programming optimization over UDDSHDC
cycle.12
Parallel (Mechanically Coupled) Hybrid Electric Drive Train Design
295
a knowledge of the future speed and load profiles of the vehicle. Nevertheless, analytic optimal policies determined through dynamic programming can
provide insight into how fuel economy improvement is achieved. Figure 8.10
shows an example of the operating points of different operating modes that
can be used to refine the rules/fuzzy logic-based control strategy as discussed
above.
8.3 Parametric Design of a Drive Train
Parameters of the parallel (torque coupling) hybrid drive train such as engine
power, electric motor power, gear ratios of transmission, and power and
energy capacity of the PPS are key parameters and exert considerable influence on vehicle performance and operation efficiency. However, as initial
steps in the drive train design, these parameters should be estimated based
on vehicle performance requirements. Such parameters should also be refined
by more accurate simulations.
In the following sections, the parameters of a passenger car are used
in calculations. These parameters are vehicle mass, M = 1500 kg; rolling
resistance coefficient, fr = 0.01; air density, ρa = 1.205 kg/m3 ; front area,
Af = 2.0 m2 ; aerodynamic drag coefficient, CD = 0.3; radius of driven wheels,
r = 0.2794 m; transmission efficiency from engine to drive wheels, ηt,e = 0.9;
and transmission efficiency from motor to drive wheels, ηt,m = 0.95.
8.3.1
Engine Power Design
The engine should be able to supply sufficient power to support vehicle operation at normal constant speeds on both a flat and a mild grade road without
the help of the PPS. The engine should also be able to produce an average
power that is larger than the average load power when the vehicle operates
with a stop-and-go operating pattern.
As a requirement of normal highway driving at a constant speed on a flat
or a mild grade road, the power needed is expressed as
V
1
Mg fr + ρa CD Af V 2 + Mgi (kW).
(8.26)
Pe =
1000ηt,e
2
Figure 8.11 shows the load powers of a 1500-kg example passenger car, along
with vehicle speed, on a flat road and a road with 5% grade. It is seen that
on a flat road, a speed of 160 km/h (100 mph) needs a power of 43 kW. For a
comprehensive analysis, the power curves of a 43-kW engine with a multigear
transmission are also plotted in Figure 8.11. From Figure 8.11, it can also be
seen that on a 5% grade road, the vehicle can reach maximum speeds of about
103 km/h and 110 km/h with fourth gear and third gear, respectively.
296
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
50
Engine power
Resistance
power on a
5% grade
road
Resistance
power on a
flat road
45
40
Power (kW)
35
30
25
20
4th gear
3rd gear
15
2nd gear
1st gear
10
5
0
0
20
40
80
60
100 120
Vehicle speed (km/h)
140
160
180
FIGURE 8.11 Engine power required at constant speed on a flat road and a 5% grade road.
Figure 8.12 is the same diagram as shown in Figure 8.11, with the addition
of the engine fuel consumption map at each gear. This figure can be used
to analyze the influence of transmission gears on vehicle performance, such
as acceleration, gradeability, and fuel consumption of the vehicle (for more
details, refer to the section of transmission design).
The above-designed engine power should be evaluated so that it meets the
average power requirement while driving in a stop-and-go pattern. In a drive
cycle, the average load power of a vehicle can be calculated by
Pave =
1
T
0
T
1
dV
Mgfr V + ρa CD Af V 3 + δMV
2
dt
dt.
(8.27)
The average power varies with the degree of regenerative braking. The two
extreme cases are the full and zero regenerative braking cases. Full regenerative braking recovers all the energy consumed in braking and the average
power is calculated by Equation 8.27, where negative dV/dt (deceleration)
will tend to reduce the average power, Pave . However, when the vehicle has
no regenerative braking, the average power is larger than that with full regenerative braking, which can be calculated from Equation 8.27 in such a way
that when the instantaneous power is less than zero, it is given a zero.
Figure 8.13 shows vehicle speed, instantaneous load power, and average
powers with full regenerative braking and zero regenerative braking in some
typical drive cycles for a 1500-kg passenger car.
In the engine power design, the average power that the engine produces
must be greater than the average load power as shown in Figure 8.13. In a
Parallel (Mechanically Coupled) Hybrid Electric Drive Train Design
Resistance
power on a
5% grade
road
Engine power
with four gears
50
297
Resistance
power on a
flat road
45
40
Power (kW)
35
30
25
20
A
15
10
5
0
0
20
40
80
60
100 120
Vehicle speed (km/h)
140
160
180
FIGURE 8.12 Engine power required at constant speed on a flat road and a 5% grade road with
the engine fuel consumption map at each gear.
parallel drive train, the engine is mechanically coupled to the driven wheels.
Hence, engine rotating speed varies with vehicle speed. On the other hand,
engine power with full throttle varies with engine rotating speed. In other
words, engine power with full throttle is associated with vehicle speed. Thus,
the determination of engine power to meet the average power in a drive cycle
is not as straightforward as in a series hybrid, in which the engine operating
point can be fixed. The average power that the engine produces with full
throttle can be calculated by
Pmax-ave =
1
T
0
T
Pe (V) dt,
(8.28)
where T is the total time in the drive cycles and Pe (V) is the engine power
with full throttle, which is a function of vehicle speed when the gear ratio of
the transmission is given, as shown in Figures 8.11 and 8.12.
The possible operating points of the engine with full throttle and the maximum possible average powers in some typical drive cycles are shown in
Figure 8.14, in which the maximum engine power is 43 kW and transmission
is single gear (fourth gear only in Figures 8.11 and 8.12). If a multigear transmission is used, the maximum average power of the engine is greater than
that used in a single-gear transmission (refer to Figures 8.11 and 8.12). Comparing the maximum possible average powers to the load average powers
in the typical driving cycles as shown in Figure 8.13, it is concluded that the
298
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Speed (km/h)
50
0
30
Instantaneous power
Power (kW)
20
Average power with
zero regenerative
braking
10
0
–10
–20
0
200
400
600
Speed (km/h)
(b)
100
100
Power (kW)
(a)
20
Average power without
regenerative braking
800 1000 1200 1400
50
0
40
Instantaneous power
0
Average power without
regenerative braking
–20
–40
0
100
200
Time (s)
FTP 75 urban driving cycle
400
500
600
700
800
(d)
Speed (km/h)
150
100
50
0
100
Instantaneous power
50
Average power with
zero regenerative
braking
Power (kW)
Speed (km/h)
300
Time (s)
FTP 75 urban driving cycle
(c)
Power (kW)
Average power with
zero regenerative
braking
0
Average power without
regenerative braking
–50
–100
0
100
200
300
400
Time (s)
US06 driving cycle
500
600
150
100
50
0
40
Average power with
zero regenerative
braking
20
Instantaneous power
0
–20
–40
Average power without
regenerative braking
0
100
200
300
400
500
600
Time (s)
ECE driving cycle
FIGURE 8.13 Instantaneous power and average power with full and zero regenerative braking
in typical drive cycles: (a) FTP75 urban, (b) FTP75 highway, (c) US06, and (d) ECE-15.
engine power designed is sufficient to support a vehicle operating in these
typical drive cycles.
8.3.2 Transmission Design
Multigear transmission can effectively increase the tractive effort of driven
wheels, from the engine torque, especially at a low to medium speed range
(refer to Figures 8.11 and 8.12). The direct benefit of the increased tractive
effort is the reduction of acceleration time and enhanced gradeability with
a given motor power rating. In other words, the motor power rating can be
reduced to meet the given acceleration performance and gradeability. Another
benefit is the large remaining engine torque for charging the PPS, in addition
to propelling the vehicle. Thus the SOC of the PPS can be brought back to a
high level quickly. However, a multigear transmission will add complexity
to the drive train, especially for the control system in which a shifting control
module must be added.
In real engineering, careful analysis is needed to make the decision of
using a single-gear or multigear transmission. In the example mentioned
above, the maximum average engine power in the typical driving cycle with
Parallel (Mechanically Coupled) Hybrid Electric Drive Train Design
(b)
45
40 Maximum average
power:
35
14.5 kW
30
Engine power-speed
25
curve with full throttle
20
15
Operating points
10
with full throttle
5
20 40 60 80 100 120 140 160 180
Engine power (kW)
Engine power (kW)
(a)
Vehicle speed (km/h)
FTP75 urban driving cycle
45
40 Maximum average
power:
35
28.3 kW
30
Engine power-speed
25
curve with full throttle
20
15
Operating points
with full throttle
10
5
20 40 60 80 100 120 140 160 180
Vehicle speed (km/h)
FTP75 urban driving cycle
(d)
45
Maximum average
40
power:
35
32.2 kW
30
Engine power-speed
curve with full throttle
25
20
Operating points
15
with full throttle
10
5
20 40 60 80 100 120 140 160 180
Engine power (kW)
(c)
Engine power (kW)
299
Vehicle speed (km/h)
US06 driving cycle
45
Maximum average
40
power:
35
22.1 kW
30
Engine power-speed
25
curve with full throttle
20
Operating points
15
with full throttle
10
5
20 40 60 80 100 120 140 160 180
Vehicle speed (km/h)
ECE-1 driving cycle
FIGURE 8.14 Maximum possible operating points of the engine and maximum average power
in typical drive cycles: (a) FTP75 urban, (b) FTP75 highway, (c) US06, and (d) ECE-15.
a single-gear transmission, as shown in Figure 8.14, is significantly larger
than the average load power even without regenerative braking as shown in
Figure 8.13. Thus, the engine is considered powerful enough with a singlegear transmission if it meets the sustainability of the SOC of PPS. Of course,
if a small engine is mandatorily used, a multigear transmission would be
necessary.
The operating efficiency of the engine would not be expected to be significantly improved by using a multigear transmission. By referring to
Figure 8.12, it can be seen that at most engine speeds, the engine will have
higher operating efficiency at the highest gear (fourth gear in Figure 8.12)
than with lower gears. However, with the highest gear, the low-speed range,
in which the engine cannot stably operate and pure electric traction is used,
is larger than that with other gears.
8.3.3
Electric Motor Drive Power Design
In HEV, the major function of the electric motor is to supply peak power to
the drive train. In motor power capacity design, acceleration performance
and peak load power in typical drive cycles are the major concerns.17
300
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
It is difficult to directly design the motor power from the acceleration performance specified. This is because we have two power sources and their
maximum power relationship with the vehicle speed. An effective approach
is to initially make an estimate of the motor power capacity based on the specified acceleration performance, and then complete the final design through
iterative simulations.
As an initial estimate, one can assume that the steady-state load (rolling
resistance and aerodynamic drag) is handled by the engine and the dynamic
load (inertial load in acceleration) is handled by the motor. With this assumption, acceleration is directly related to the torque output of an electric
motor by
dV
Tm itm ηtm
= δm M
,
r
dt
(8.29)
where Tm is the motor torque, itm is the gear ratio from the motor to the driven
wheels, where a single transmission is supposed, ηtm is the transmission efficiency from the motor to the driven wheels, and δm is the rotating inertia
factor associated with the motor (refer to Chapter 2).
Using the output characteristics of the electric motor shown in Figure 8.15,
and a specified acceleration time, ta , from zero speed to a final high speed,
Vf , and referring to Chapter 4, the motor power rating is expressed as
Pm =
δm M
(V 2 + Vb2 ).
2ηtm ta f
(8.30)
Tractive effort
For a 1500-kg passenger car with a maximum speed of 160 km/h, a base
speed of 50 km/h, a final acceleration speed of 100 km/h, acceleration time
ta = 10 s, and δm = 1.04, the power rating of the electric motor is about 74 kW.
Vb
Vf
0
0
Vehicle speed
FIGURE 8.15 Tractive effort versus vehicle speed of an electric-motor-driven vehicle.
Parallel (Mechanically Coupled) Hybrid Electric Drive Train Design
301
It should be noted that the motor power obtained above is somewhat
overestimated. Actually, the engine has some remaining power to help the
motor to accelerate the vehicle as shown in Figures 8.11 and 8.12. This is also
seen in Figure 8.16, in which vehicle speed, engine power with full throttle,
and resistance power (rolling resistance, aerodynamic drag, and power losses
in transmission) with a multigear and single-gear transmission are plotted
along acceleration time. The average remaining power of the engine, used to
accelerate the vehicle, can be expressed as
Pe,a
1
=
ta − t i
ta
ti
(Pe − Pr ) dt,
(8.31)
where Pe and Pr are engine power and resistance power, respectively. It should
be noted that the engine power transmitted to the driven wheels is associated
with the gear number and gear ratios of the transmission. It is obvious from
Figures 8.11 and 8.12 that a multigear transmission will effectively increase the
remaining power at the driven wheels, therefore reducing the motor power
required for acceleration.
Figure 8.16 shows the engine remaining power over that required for
overcoming the rolling resistance and aerodynamic drag with a multigear
and single-gear transmission during acceleration. This figure indicates that
around 17- and 22-kW engines with a single-gear and multigear transmission,
respectively, are available for assisting the motor in acceleration. Finally, the
motor power rating is 74 − 17 = 57 kW for a single-gear transmission and
74 − 22 = 52 kW for a multigear transmission.
When the power ratings of the engine and electric motor as well as the
transmission are initially designed, a more accurate calculation needs to be
120
100
Vehicle speed (km/h)
80
Engine power
Engine power
with single-gear
with multigear
transmission (kW) transmission (kW)
60
40
a
tion with
r accelera (kW )
fo
r
e
w
o
a
Engine p ear transmission
tion with
multig
accelera sion (kW )
r
fo
r
e
w
o
is
Engine p gle-gear transm
sin
20
0
0 ti
2
4
6
Time (s)
8
Resistance
power (kW)
ta
10
12
FIGURE 8.16 Vehicle speed, engine power, and resistance power versus acceleration time.
302
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
performed to evaluate vehicle performance, mainly maximum speed, gradeability, and acceleration. Maximum speed and gradeability can be obtained
from the diagram of tractive effort and resistance versus vehicle speed. This
diagram can be created by using the methods discussed in Chapter 2.
The diagram (as shown in Figure 8.17) shows the design results of the
example passenger car with a single-gear transmission. It indicates that the
maximum gradeability of the vehicle is about 42% or 22.8◦ at a vehicle speed
of about 40 km/h (point A). At a vehicle speed of 100 km/h, the gradeability of the vehicle with full hybrid, motor-alone and engine-alone traction are
18.14% or 10.28◦ (point B), 10.36% or 5.91◦ (point C), and 4.6% or 2.65◦ (point
D), respectively. The maximum speed of the vehicle is round 160 km/h with
engine-alone traction, which is dictated by the engine power point (E). However, if the engine and motor top speed can extend beyond this vehicle speed,
the vehicle maximum speed with hybrid mode and motor-alone mode can be
extended further.
Figure 8.18 shows the acceleration performance for the example passenger
car with a single-gear transmission. It indicates that 10.7 s are used and 167 m
are covered for accelerating the vehicle from zero speed to 100 km/h.
The calculation results for vehicle performance shown in Figures 8.17
and 8.18 indicate that the design of the engine and motor power capacities
are appropriate.
8.3.4
PPS Design
The PPS design mainly includes the design for power capacity and energy
capacity. The power capacity design is somewhat straightforward. The
8
Tractive effort on driven wheels, kN
Engine + motor
7
a = 25° (46.6%)
6
A
a =20° (36.4%)
5
Motor alone
a=15° (26.8%)
4
a = 10° (17.6%)
B
3
2
C
Engine alone
D
1
E
0
0
20
40
60
80
100
120
140
160
a =5° (8.7%)
Resistance:
a = 0 (0%)
Maximum
speed
180
Vehicle speed (km/h)
FIGURE 8.17 Tractive effort and resistance on a slope versus vehicle speed.
40
400
35
350
30
300
Distance
25
250
20
200
15
150
10
100
Time
5
0
303
Acceleration distance (m)
Acceleration time (s)
Parallel (Mechanically Coupled) Hybrid Electric Drive Train Design
50
0
20
40
60
80
100
Vehicle speed (km/h)
120
0
140
FIGURE 8.18 Acceleration time and distance versus vehicle speed.
terminal power of the PPS must be greater than the input electric power
of the electric motor, that is,
Ps ≥
Pm
,
ηm
(8.32)
where Pm and ηm are motor power rating and efficiency, respectively.
The energy capacity design of the PPS is closely associated with the electrical energy consumption in various driving patterns—mainly the full load
acceleration and typical urban driving cycles.
During the acceleration period, the energies drawn from the PPS and the
engine can be calculated along with the calculation of acceleration time and
distance by
ta
Pm
dt
(8.33)
Epps =
0 ηm
and
Eengine =
0
ta
Pe dt,
(8.34)
where Epps and Eengine are the energies drawn from the PPS and the engine,
respectively, and Pm and Pe are the powers drawn from the motor and the
engine, respectively. Figure 8.19 shows the energies drawn from the PPS and
the engine in the period of full acceleration along the vehicle speed for the
example passenger car. At an end speed of 120 km/h, about 0.3 kW h energy
is drawn from the PPS.
The energy capacity of the PPS must also meet the requirement while driving in a stop-and-go pattern in typical drive cycles. In other words, the energy
304
0.35
40
0.3
35
30
Energy from
the PPS
0.25
25
0.2
Acceleration
Time
0.15
0.1
20
15
10
0.50
0
0
20
100
60
80
Vehicle speed (km/h)
40
Acceleration distance (m)
Energy consumption (kWh)
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Energy from 5
the engine
0
120
140
FIGURE 8.19 Energies drawn from the PPS and engine during full acceleration period.
in the PPS will not be fully discharged. The energy changes in the PPS can be
obtained by
t
Evar =
Ppps-ch − Ppps-disch dt,
(8.35)
0
where Ppps-ch and Ppps-disch are the instantaneous charging and discharging
power of the PPS. With a given control strategy, the charging and discharging
power of the PPS can be obtained by a drive train simulation in a typical
driving cycle (refer to Section 8.4).
Figure 8.20 shows the simulation results of the example passenger car in an
FTP75 urban drive cycle with maximum SOC control strategy. It can be seen
that the maximum energy change in the PPS is about 0.11 kWh, which is less
than that in full load acceleration (0.3 kWh). Thus, the energy consumption
in full load acceleration determines the energy capacity of the energy storage.
This conclusion is only valid for the Max. SOC control strategy and FTP75
urban driving cycle. When other control strategies and driving cycles are
used, the conclusion may be different.
In fact, not all the energy stored in the PPS can be fully used to generate
sufficient power for the drive train. In the case of using batteries as the PPS,
low SOC will severely limit their power output, and will, at the same time, lead
to low efficiency, due to an increase in internal resistance. In the case of using
ultracapacitors as the PPS, low SOC will result in low terminal voltage that
will affect the performance of the traction motor. Similarly when a flywheel is
used, low SOC means the low flywheel speed at which the terminal voltage
of the electric machine, functioning as the energy exchange port, is low. Thus
only part of the energy stored in the PPS can be available for use, which can
Engine power Vehicle speed
(kW)
(km/h)
100
Energy Variation Motor power
in the PPS (kWh)
(kW)
Parallel (Mechanically Coupled) Hybrid Electric Drive Train Design
20
305
50
0
40
20
0
0
–20
1.2
0.11 kWh
1
0.8
0
200
400
600
800
Time (s)
1000
1200
1400
FIGURE 8.20 Vehicle speed, engine power, motor power, and energy variation in PPS storage
in an FTP75 urban drive cycle with the Max. SOC control strategy.
be represented by the SOC or state-of-energy. The energy capacity of the PPS
can be obtained from
Edis-max
Ec-pps =
,
(8.36)
SOCt − SOCb
where Edis-max is the allowed maximum energy discharging from the PPS,
and SOCt and SOCb are the top line and bottom line of the SOC of the PPS. In
the example, Edis-max = 0.3 kWh and assume that 30% of the total energy of
the PPS is allowed to be used; then the minimum energy capacity of the PPS
is 1 kWh.
8.4 Simulations
When all the major components have been designed, the drive train should
be simulated by using a simulation program. The simulation in typical drive
cycles can produce a great deal of useful information about the drive train,
such as engine power, motor power, energy changes in the PPS, engine
operating points, motor operating points, fuel consumption, and so on.
Figure 8.20 shows the vehicle speed, engine power, motor power, and
energy changes in the PPS along with the driving time for the example passenger car in the FTP75 urban drive cycle. Figures 8.21 and 8.22 show the
306
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Engine power (kW)
5
120
10 15
20 25 30 35 40
+ Operating points
100
Engine torque (Nm)
45
250
260
80
270
280
60
310
350
40
400
500
600
700 600
1000
bsfc(g/kWh)
20
0
0
500 1000 1500 2000 2500 3000 3500 4000 4500 5000 5500
Engine speed (rpm)
FIGURE 8.21 Engine operating points overlap its fuel consumption map in an FTP75 urban
drive cycle with maximum SOC control strategy.
engine and motor operating points, respectively. The simulation results of
the fuel economy of the example passenger car are 4.66 L per 100 km or 50.7
miles per gallon (mpg) when the engine is turned off during the period of
standstill and braking, and 5.32 L per 100 km or 44.4 mpg when the engine is
set at idle during the period of standstill and braking.
500
+ Operating points
400
Motor torque (Nm)
300
200
100
0
–100
–200
–300
–400
–500
0
500 1000 1500 2000 2500 3000 3500 4000 4500 5000
Motor speed (rpm)
FIGURE 8.22 Motor operating points in an FTP75 urban drive cycle with maximum SOC control
strategy.
Parallel (Mechanically Coupled) Hybrid Electric Drive Train Design
307
References
1. M. Ehsani, K. L. Butler, Y. Gao, K. M. Rahman, and D. Burke, “Toward a sustainable
transportation without sacrifice of range, performance, or air quality: The ELPH
car concept,” International Federation of Automotive Engineering Society Automotive
Congress, Paris, France, September/October 1998.
2. M. Ehsani, Y. Gao, and K. Butler, “Application of electric peaking hybrid (ELPH)
propulsion system to a full size passenger car with simulation design verification,”
IEEE Transactions on Vehicular Technology, 48 (6), November 1999.
3. Y. Gao, K. M. Rahman, and M. Ehsani, “The energy flow management and battery
energy capacity determination for the drive train of electrically peaking hybrid,”
Society of Automotive Engineers (SAE) Journal, Paper No. 972647, Warrendale, PA,
1997.
4. Y. Gao, K. M. Rahman, and M. Ehsani, “Parametric design of the drive train of an
electrically peaking hybrid (ELPH) vehicle,” Society of Automotive Engineers (SAE)
Journal, Paper No. 970294, Warrendale, PA, 1997.
5. C. Liang, W. Weihua and W. Qingnian, “Energy Management Strategy and
Parametric Design for Hybrid Electric Military Vehicle,” SAE paper 2003-01-0086.
6. P. Pisu, and G. Rizzoni, “A comparative study of supervisory control strategies
for hybrid electric vehicles,” IEEE Transaction on Control Systems Technology, 15 (3),
May 2007.
7. H.-D. Lee and S.-K. Sul, “Fuzzy-logic-based torque control strategy for paralleltype hybrid electric vehicle,” IEEE Transaction on Industrial Electronics, 45 (4),
August 1998.
8. G. Shi, Y. Jing, A. Xu, and J. Ma, “Study and simulation of based-fuzzy-logic parallel hybrid electric vehicles control strategy,” Proceedings of the Sixth International
on Intelligent Systems Design and Application (ISDA’06), 2006.
9. R. Langari and J.-S. Won, “Intelligent energy management agent for a parallel
hybrid vehicle—part I: System architecture and design of the driving situation identification process,” IEEE Transactions on Vehicular Technology, 54 (3),
May 2005.
10. R. Langari and J.-S. Won, “Intelligent energy management agent for a parallel hybrid vehicle—part II: Torque distribution, charge sustenance strategies,
and performance results,” IEEE Transactions on Vehicular Technology, 54 (3),
May 2005.
11. T. Heske and J. N. Heske, Fuzzy Logic for Real World Design, Annabooks, ISBN:
0-929392-24-8, 1996.
12. C.-C. Lin, H. Peng, J. W. Grizzle, and J.-M. Kang, “Power management strategy
for a parallel hybrid electric truck,” IEEE Transactions on Control System Technology,
11 (6), November 2003.
13. C.-C. Lin, J.-M. Kang, J. W. Grizzle, and H. Peng, “Energy management strategy
for a parallel hybrid electric truck,” Proceedings of the American Control Conference,
Arlington, VA, June 25–27, 2001.
14. C.-C. Lin, H. Peng, S. Jeon, and J. M. Lee, “Control of a hybrid electric truck based
on driving pattern recognition,” Proceedings of the 2002 Advanced Vehicle Control
Conference, Hiroshima, Japan, September 2002.
15. D. P. Betsekas, Dynamic Programming and Optimal Control, Athena Scientific,
1995.
308
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
16. C.-C. Lin, Z. Filipi, L. Louca, H. Peng, D. Assanis, and J. Stein, “Modeling and
control of a medium-duty hybrid electric truck,” International Journal of Heavy
Vehicle Systems. 11 (3/4), 2004.
17. Y. Gao, H. Moghbelli, and M. Ehsani, “Investigation of proper motor drive characteristics for military vehicle propulsion,” Society of Automotive Engineers (SAE)
Journal, Paper No. 2003-01-2296, Warrendale, PA, 2003.
9
Design and Control Methodology of
Series–Parallel (Torque and Speed Coupling)
Hybrid Drive Train
As discussed in Chapter 5, the series–parallel or, more accurately, the
torque/speed-coupling hybrid drive train has some advantages over the
series (electrical coupling) and parallel (single torque or speed coupling) drive
trains. The torque and speed couplings in this drive train free the engine from
the driven wheels in the torque and speed constraints. Consequently, the
instantaneous engine torque and speed can be independent of the load torque
and speed of the vehicle. Therefore, the engine can be operated in its highefficiency region in a similar way as that of the series (electrical coupling)
drive train. On the other hand, part of the engine power is directly delivered to the driven wheels without experiencing multiform conversion.1–3
This feature is more similar to the parallel (torque or speed coupling) drive
train.
As discussed in Chapter 5, the series–parallel hybrid drive train can be
composed of speed-coupling devices such as planetary gears and transmotors as shown in Figures 5.22 through 5.24. All these configurations have
similar features, design, and control principles. This chapter focuses on the
design and control principles of the configuration, which uses the planetary
gear unit as its speed-coupling device as shown in Figure 5.22. For a more
detailed example of a commercial series–parallel vehicle, see the Appendix in
this book.
9.1 Drive Train Configuration
9.1.1
Speed-Coupling Analysis
A series–parallel hybrid drive train can be formed by using both torque and
speed coupling. The well-known torque-coupling devices are mostly gear set,
sprocket-chain set, or pulley-belt set.4,5 However, speed-coupling devices are
less familiar to the reader and more complex. The operating characteristics of
309
310
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
planetary gear functioning as a speed-coupling device are discussed in detail
as follows.
A mechanical planetary unit has the structure as shown in Figure 9.1. It
consists of a sun gear labeled s in Figure 9.1, a ring gear labeled r, several
planetary gears labeled p (usually three or four for force balance), and a yoke
labeled y, which is hinged to the centers of the planetary gears. As discussed
in Chapter 5, the speeds in rpm of the sun gear, ns , ring gear, nr , and yoke, ny ,
have the relationship
ny =
ig
1
ns +
nr ,
1 + ig
1 + ig
(9.1)
where ig is the gear ratio defined as Rr /Rs as shown in Figure 9.1. The speeds
ns , nr , and ny are defined as positive in the direction shown in Figure 9.1.
Defining kys = (1 + ig ) and kyr = (1 + ig )/ig , Equation 9.1 can be further
expressed as
1
1
ny =
ns +
nr .
(9.2)
kys
k yr
Neglecting the energy losses in steady-state operation, the torques acting on
the sun gear, ring gear, and yoke have the relationship
Ty = −kys Ts = −kyr Tr .
(9.3)
Equation 9.3 indicates that the torques acting on the sun gear, Ts , and ring
gear, Tr , always have the same sign; in other words, they have to be always
in the same direction. However, the torque acting on the yoke, Ty , is always
in the opposite direction of Ts and Tr . Equation 9.3 also indicates that with
ig > 1, which is the general case since Rr > Rs , Ts is the smallest, Ty is the
largest, and Tr is in between. This means that the torque acting on the yoke
is balanced by torques acting on the sun gear and ring gear.
nr , Tr
r
y
Rs
Ry
s
Rr
ny
p
FIGURE 9.1 Planetary gear unit used as a speed coupling.
ns , T s
Ty
311
Design and Control Methodology of Hybrid Drive Train
Element fixed
Speed
Torque
Sun gear
ny =
1
n
kyr r
Ty = –kyrTr
Ring gear
ny =
1
n
kys s
Ty = –kysTs
Yoke
ns = –
kys
n
kyr r
Ts =
kyr
T
kys r
FIGURE 9.2 Speed and torque relationships while one element is fixed.
When one element among the sum gear, ring gear, and yoke is locked to
the vehicle frame, that is, one degree of freedom of the unit is constrained,
the unit becomes a single-gear transmission (one input and one output); the
speed and torque relationship, with one element fixed, is shown in Figure 9.2.
In composing a hybrid drive train with the planetary gear unit as a speed
coupling, there are many options as shown in Figure 9.3. To reduce the torque
capacity requirement, therefore reducing the motor/generator physical size
and weight, connecting the motor/generator to the sun gear of the planetary
gear unit may be the appropriate choice. The engine may be either connected
to the yoke or to the ring gear as shown in Figure 9.3a and b. In the former
design (Figure 9.3a), the torques of the engine and motor/generator have the
relationship
(9.4)
Te = −kys Tm/g ,
where Te and Tm/g are the torques acting on the yoke and sun gear, produced
by the engine and the motor/generator. Te and Tm/g have opposite directions.
The engine operates in the first quadrant, and the motor operates in the third
and fourth quadrants as shown in Figure 9.4. Since the motor/generator has
to produce its torque to balance the engine maximum torque at any speed, the
motor/generator must have a constant maximum torque in its whole speed
range as shown in Figure 9.4.
In the latter design (Figure 9.3b), the engine torque and motor/generator
torque have the relationship
Te =
kyr
Tm/g .
kys
(9.5)
The engine and motor/generator operating areas are shown in Figure 9.5.
Comparing Figure 9.4 with Figure 9.5, with the same maximum engine
torque, the former design (Figures 9.3a and 9.4) results in a smaller
motor/generator (smaller motor/generator torque). However, the torque
from the ring gear, delivered to the driven wheel through the gear box, is
smaller than the engine torque (similar to the overdrive gear in the conventional transmission). Nevertheless, the gear ratio of the gear box can be
Gear
box
Ring
gear
Driven
wheels
Motor/
generator
Ring
gear
Driven
wheels
Gear
box
Sun gear—Driven wheels
Ring gear—Motor/generator
Yoke—Engine
Motor/
generator
Yoke
Sun
gear
Traction
motor
Ring
gear
Engine
Yoke
Sun
gear
Motor/
generator
Ring
gear
Driven
wheels
Gear
box
Sun gear—Driven wheels
Ring gear—Engine
Yoke—Motor/generator
Engine
Yoke
Sun
gear
Traction
motor
Sun gear—Motor/generator
Ring gear—Engine
Yoke—Driven wheels
Driven
wheels
Gear
Box
Traction
motor
Motor/
generator
(e)
(b)
(f )
Ring
gear
Gear
Box
Yoke
Sun
gear
Driven
wheels
Ring
gear
Motor/
generator
Yoke
Sun
gear
Sun gear—Engine
Ring gear—Motor/generator
Yoke—Driven wheels
Driven
wheels
Gear
box
Traction
motor
Engine
Engine
Sun gear—Engine
Ring gear—Driven wheels
Yoke—Motor/generator
Traction
motor
Motor/
generator
(c)
FIGURE 9.3 Possible configurations of the torque- and speed-coupling hybrid drive train with connections of (a) motor/generator to sun gear, drive
wheel to ring gear and engine to yoke, (b) motor/generator to sun gear, engine to ring gear and drive wheel to yoke, (c) engine to sun gear, drive wheel to
ring gear and motor/generator to yoke, (d) drive wheel to sun gear, motor/generator to ring gear and engine to yoke, (e) drive wheel to sun gear, engine
to ring gear and motor/generator to yoke, and (f) engine to sun gear, motor/generator to ring gear and drive wheel to yoke.
Engine
(d)
Yoke
Sun
gear
Sun gear—Motor/generator
Ring gear—Driven wheels
Yoke—Engine
Traction
motor
Engine
(a)
312
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
313
Design and Control Methodology of Hybrid Drive Train
Te , Tm/g
Te-max
we , wm/g
Tm/g-max =
Te-max
kys
Engine connected to yoke
Motor/generator connected to sun gear
FIGURE 9.4 Operating areas of the engine and motor/generator.
designed to meet the tractive torque requirement. Further discussion in the
following section will be based on this design (Figures 9.3a and 9.4).
9.1.2
Drive Train Configuration
Figure 9.6 shows the detailed configuration of a series–parallel (torque/speed
coupling) drive train.6 The planetary gear unit constitutes the speed coupling that connects an engine and a motor/generator together. The engine
and motor/generator are connected to the yoke and sun gear, respectively.
The ring gear of the planetary gear is connected to the drive wheels through
gears of Z1 , Z2 , Z4 , Z5 , and a differential. A traction motor is connected to
the driven wheels through gears of Z3 , Z2 , Z4 , Z5 , and the differential, which
couples the output torques of the ring gear and the traction motor together.
In this configuration, one clutch and two locks are used. The clutch serves for
connecting or disconnecting the engine to or from the yoke of the planetary
gear unit. Lock 1 is used to lock or release the sun gear and the shaft of the
Te , Tm/g
Te-max
Tm/g-max =
kys
kyr
Te-max
we , wm/g
Engine connected to ring gear
Motor/generator connected to sun gear
FIGURE 9.5 Operating areas of the engine and motor/generator.
314
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Acceleration
pedal
Traction
torque
command
Brake
pedal
Engine
control
Brake
torque
command
Planetary gear unit
Ring gear
Sun gear
Lock 2
Yoke
Clutch
Lock 1
Z1
Motor/generator
Engine
Differential
Traction motor control
Vehicle
control
unit
Z2
Z4
To
wheel
Traction
motor
Traction motor
controller
PPS SOC
PPS
Z3
Z5
Generator/motor
controller
Generator/motor
control
Vehicle speed signal
FIGURE 9.6 Drive train configuration.
generator/motor to or from the stationary frame of the vehicle. Lock 2 is used
to lock or release the yoke to or from the stationary frame of the vehicle. By
controlling the clutch, locks, engine, motor/generator, and the traction motor,
many operation modes are available to be used as follows.
1. Speed-coupling mode: In this mode, the traction motor is de-energized.
There are three submodes:
1.1. Engine-alone traction: The clutch is engaged to connect the engine
to the yoke, lock 1 locks the sun gear to the vehicle stationary
frame, and the motor/generator is de-energized. Lock 2 releases
the yoke from the vehicle stationary frame. The energy flow route
is shown in Figure 9.7.
Engine
TM
PLG
GB
M/G
PLG—Planetary gear unit
M/G—Motor generator
TM—Traction motor
GB—Gearbox
PPS
FIGURE 9.7 The traction energy flow route is from the engine alone.
Design and Control Methodology of Hybrid Drive Train
315
In this case, the engine alone delivers its torque to the driven
wheels. The speed relationship between the engine and the driven
wheels is
kyr ne
,
(9.6)
ndw =
irw
where ndw and ne are the speeds of the driven wheel and the
engine, and irw is the gear ratio from the ring gear to the drive
train wheels, which is expressed as
irw =
Z5 Z2
,
Z1 Z4
(9.7)
where Z1 , Z2 , Z4 , and Z5 are the tooth numbers of the gears Z1 ,
Z2 , Z4 , and Z5 .
The torque relationship between the drive wheels and the
engine is
Tdw =
irw ηyr ηrw Te
,
kyr
(9.8)
where Tdw is the torque developed on the driven wheels by the
engine torque Te , ηyr is the efficiency from the yoke to the ring
gear, and ηrw is the efficiency from the ring gear to the driven
wheels.
1.2. Motor/generator-alone traction: In this mode, the engine is shut
down; the clutch is engaged or disengaged and lock 1 releases
the sun gear and the shaft of the motor/generator from the stationary frame; lock 2 locks the yoke to the stationary frame. In
this case, the vehicle is propelled by the motor/generator alone.
The energy flow route is shown in Figure 9.8.
The speed and torque relationship between the generator/
motor and the driven wheels is
ndw = −
Engine
TM
PLG
GB
M/G
kyr
nm/g
kys irw
PLG — planetary gear unit
M/G — motor generator
TM — Traction motor
GB — Gearbox
PPS
FIGURE 9.8 Energy flow route in the mode of motor/generator traction.
(9.9)
316
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
and
Tdw =
kys irw ηsr ηrw
Tm/g ,
kyr
(9.10)
where Tdw is the tractive torque on the driven wheels developed
by the motor/generator torque Tm/g , and ηsr is the efficiency from
the sun gear to the ring gear.
It should be noted that the motor/generator has to be operated in the third quadrant, that is, negative angular velocity and
negative torque as defined in Figure 9.1.
1.3. Engine and motor/generator with speed-coupling traction: In this
mode, the clutch is engaged. Locks 1 and 2 are released from
the stationary frame. From Equation 9.2, the angular velocities
of the driven wheels, engine, and motor/generator have the
relationship
kyr
1
ndw =
ne −
(9.11)
nm/g ,
irw
kys
and the torques have the relationship
Tdw =
irw ηyr ηrw
kys ηbsr ηrw
Te =
Tm/g ,
kyr
kyr
(9.12)
where b is an index, when the power flows from the motor/
generator to the sun gear, that is, nm/g < 0, b = 1, otherwise
b = −1. Equation 9.11 implies that, at a given vehicle speed, the
engine speed can be adjusted by the motor/generator speed.
Equation 9.12 indicates that the engine torque, motor/generator
torque, and load torque on the driven wheels always hold a fixed
relationship. This implies that a change in one torque will cause a
change in the other two torques, causing the operating points
of the engine and motor/generator to change. For a detailed
discussion, refer to the next section for drive train control.
The energy flow routes are shown in Figure 9.9.
2. Torque-coupling mode: When the traction motor is energized, its torque
can be added to the torque output of the ring gear to constitute the
torque-coupling mode. Corresponding to the three modes (1.1), (1.2),
and (1.3), when the traction motor is controlled to operate in motoring
and generating, six basic operation modes are constituted.
2.1. Engine alone in mode (1.1) plus traction motor motoring: This mode is
the same as the general parallel hybrid traction mode. The energy
flow route is shown in Figure 9.10.
2.2. Engine alone in mode (1.1) plus traction motor generating: This mode
is the same as the PPS charging from the engine mode in the
317
Design and Control Methodology of Hybrid Drive Train
(a)
(b)
Engine
M/G
PLG
TM
GB
Engine
TM
PPS
PLG
M/G
GB
PPS
PLG—Planetary gear unit
M/G— otor generator
TM—Traction motor
GB—Gearbox
FIGURE 9.9 Energy flow route in speed-coupling mode: (a) M/G motoring and (b) M/G
generating.
Engine
TM
PLG
M/G
GB
PLG—Planetary gear unit
M/G—Motor generator
TM—Traction motor
GB—Gearbox
PPS
FIGURE 9.10 Energy flow route in parallel traction mode.
general hybrid drive train. The energy flow route is shown in
Figure 9.11.
2.3. Motor/generator-alone mode (1.2) plus traction motor motoring: This
mode is similar to mode (2.1), but the engine is replaced by the
motor/generator. The energy flow route is shown in Figure 9.12.
2.4. Motor/generator alone in mode (1.2) plus traction motor generating:
This mode is similar to mode (2.2) but the engine is replaced by
the motor/generator. This mode may never be used because part
of the motor/generator energy circles from the PPS and finally
to the PPS through the motor/generator and traction motor as
shown in Figure 9.13.
Engine
TM
PLG
GB
M/G
PLG—Planetary gear unit
M/G—Motor generator
TM—Traction motor
GB—Gearbox
PPS
FIGURE 9.11 Energy flow route in parallel PPS charging.
318
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Engine
PLG
TM
GB
M/G
PLG—Planetary gear unit
M/G—Motor generator
TM—Traction motor
GB—Gearbox
PPS
FIGURE 9.12 Energy flow route in the mode of two motor traction in parallel.
Engine
PLG
TM
GB
PLG—Planetary gear unit
M/G—Motor generator
TM—Traction motor
GB—Gearbox
M/G
PPS
FIGURE 9.13 Energy flow route in the mode of motor/generator traction and PPS charging.
2.5. Speed-coupling traction in mode (1.3) plus traction motor motoring:
This mode uses the full functions of speed and torque coupling.
There are two operating states of the motor/generator: motoring
and generating as shown in Figure 9.14. The operating states of
the motor/generator in motoring (Figure 9.14a) may be used at
high vehicle speeds. In this case, the engine speed may be limited to somewhat lower than its medium speed to avoid too high
an engine speed where its operating efficiency may be low. The
motor generator contributes its speed to the drive train for supporting the high vehicle speed as shown in Figure 9.14a. Similarly,
the operating states in Figure 9.14b may be used in the case of
low vehicle speed. In this case, the engine can be operated at
(a)
(b)
Engine
PLG
TM
GB
PPS
M/G
Engine
PLG
TM
GB
M/G
PPS
PLG—Planetary gear unit
M/G—Motor generator
TM—Traction motor
GB—Gearbox
FIGURE 9.14 Energy flow route: (a) traction motor motoring and (b) traction motor generating.
319
Design and Control Methodology of Hybrid Drive Train
(a)
Engine
PLG
TM
GB
(b)
M/G
PPS
Engine
PLG
TM
GB
M/G
PPS
PLG—Planetary gear unit
M/G—Motor generator
TM—Traction motor
GB—Gearbox
FIGURE 9.15 Energy flow route: (a) M/G motoring and (b) M/G generating.
speeds somewhat lower than its medium speed to avoid too low
a speed operation, where its operating efficiency may be low. The
motor/generator absorbs part of the engine speed.
2.6. Speed-coupling traction in mode (1.3) plus traction motor generating:
Similar to mode (2.5), the engine and motor/generator operate in speed-coupling mode. But the traction motor operates in
generating mode as shown in Figure 9.15.
3. Regenerative braking: When the vehicle is experiencing braking, the
traction motor, motor/generator, or both can produce braking torque
and recapture part of the braking energy to charge the PPS. In this
case, the engine is shut down with the clutch opened. The possible
energy flow is shown in Figure 9.16.
As discussed above, there are several operating modes available for use.
In control scheme design, perhaps not all the operating modes are really
used, depending on the drive train design, driving conditions, operating
characteristics of the major components, and so on.
(a)
(b)
Engine
PLG
TM
GB
PPS
M/G
(c)
Engine
PLG
TM
GB
M/G
PPS
Engine
PLG
TM
GB
M/G
PPS
PLG—Planetary gear unit
M/G—Motor generator
TM—Traction motor
GB—Gearbox
FIGURE 9.16 Energy flow in regenerative braking: (a) traction motor alone, (b) M/G alone, and
(c) both traction motor and M/G.
320
9.2
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Drive Train Control Methodology
9.2.1
Control System
The control system of the drive train is shown in Figure 9.6. The vehicle controller unit (VCU) receives the traction or braking torque commands from the
driver through the accelerator or brake pedals, and other necessary operating
information, such as SOC of the PPS and vehicle speed. Based on the real-time
information received and the control logic preset in the VCU, the VCU generates control signals to control the engine, motor/generator, traction motor, as
well as clutch and locks, through engine throttle actuator, motor/generator
controller, traction motor controller, clutch, and lock actuators.
9.2.2
Engine Speed Control Approach
Equation 9.11 indicates that the engine speed, ne , can be adjusted by controlling the motor/generator speed, nm/g , at a given wheel speed, ndw . However,
this control activity has to be carried out by controlling the engine throttle and
the motor/generator torque as shown in Figure 9.17. The control procedure
is as follows:
Suppose the engine is operating at point a with a speed of ne,a , producing torque Te,1 , with a throttle angle of 60◦ as shown in Figure 9.17; the
motor/generator has to produce its torque Tm/g,1 = Te,1 /kys (where the losses
q—Engine throttle angle
Te,2
Te,1
Te,0
d
c
a
Te,2 = kys Tg/m,2
Te,1 = kys Tg/m,1
Te,0 = kys Tg/m,0
b
e
Engine torque
q=
q=
q=
q=
q=
ne,c
q=
20∞
ne,d
25∞
q=
30∞
ne,a ne,e ne,b
q=
q=
35∞
90
∞
70
∞
60
∞
50
∞
40
∞
Engine rpm
FIGURE 9.17 Engine speed controlled by the engine throttle and motor/generator torque.
Design and Control Methodology of Hybrid Drive Train
321
are ignored) to balance the engine torque. With a fixed motor/generator
torque, and thus a fixed engine torque, increasing the engine throttle opening
will cause the engine speed to increase, to point b with θ = 70◦ for example.
Similarly, reducing the engine throttle opening will cause the engine speed
to decrease, to point c at θ = 50◦ for example. The engine speed can also be
changed by changing the motor/generator torque as shown in Figure 9.17.
With a fixed engine throttle, reducing the motor/generator torque (and thus
the engine torque) will cause the engine speed to increase from point a to point
e, or increasing the motor/generator torque will cause the engine speed to
decrease from point a to point d. Thus, the engine speed can be potentially
controlled within its optimal speed range by instantaneously controlling the
engine throttle and/or the motor/generator torque.
9.2.3 Traction Torque Control Approach
The traction torque on the driven wheels is the sum of the torques transmitted
from the ring gear of the planetary gear units and the traction motor. The
traction torque on the driven wheels can be expressed as
Ttdw = irw ηrw Tring + imw ηmw Ttm ,
(9.13)
where Ttdw is the total tractive effort on the driven wheels, Tring is the
torque output from the ring gear of the planetary gear unit, which is generated by the engine and the motor/generator, irw and ηrw are the gear
ratio and transmission efficiency from the ring gear to the driven wheels,
respectively, Ttm is the traction motor torque, and ηmw and imw are the transmission efficiency and gear ratio from the traction motor to the driven wheels,
where imw = (Z2 Z5 )/(Z3 Z4 ) (Z2 , Z3 , Z4 , and Z5 are the tooth numbers of the
corresponding gears as shown in Figure 9.6).
The total traction torque request on the driven wheels, which is commanded by the driver through the accelerator pedal, can be met by the
torque outputs from the ring gear, Tring , and the traction motor, Ttm . As mentioned above, Tring can be obtained by controlling the engine throttle and the
motor/generator torque to operate the engine with high efficiency. The contributions of Tring and Ttm to the total depend on the control strategy of the
drive train, which will be discussed in the next section.
Figure 9.18 illustrates the simulation results of an example drive train with
full engine throttle opening and full traction motor load (maximum torque vs.
motor speed). In the simulation, the engine rpm is controlled such that, at low
vehicle speeds, the engine operates with a constant speed (1200 rpm in this
example) and the motor/generator operates with positive speeds. At medium
vehicle speeds, the motor/generator is locked to the vehicle frame and the
engine speed linearly increases with vehicle speed (pure parallel or pure
torque-coupling operation). At high vehicle speeds, the engine again operates
with a constant speed (3500 rpm in this example) and the motor/generator
322
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
(a)
Low vehicle
speed region
(constant
engine speed
and positive
M/G speed)
6000
rpm
4000
VL
Medium vehicle
speed region with
Zero M/G speed and
engine speed
proportional to
vehicle speed
2000
VH
High vehicle speed
region (constant
engine speed and
negative M/G
speed)
Engine
rpm
Motor/generator
rpm
0
Traction
motor rpm
–2000
–4000
(b)
0
20
40
60
80
100
Vehicle speed (km/h)
160
140
160
Traction motor
150
Torque (Nm)
140
VH
VL
200
120
Engine
100
50
Motor/generator
0
–50
0
20
40
60
80
100
120
Vehicle speed (km/h)
Traction torque on
wheels (Nm)
(c)
VL
1000
VH
Total
800
Produced by
traction motor
600
400
200
0
Produced by engine and
motor/generator
0
20
40
60
80
100
Vehicle speed (km/h)
120
140
160
FIGURE 9.18 Torques and speeds of the engine, traction motor, motor/generator, and driven
wheels with full engine throttle opening and full loading of the traction motor along vehicle
speed: (a) speeds, (b) torques of the power plants, and (c) torques on the driven wheels.
operates with negative speed (rotating in the opposite direction of the engine).
With the pervious engine speed control, the engine operating speeds are constrained in the medium range in which the engine efficiency may be higher.
It is noted that the motor/generator is de-energized in the medium vehicle
Design and Control Methodology of Hybrid Drive Train
323
speed range for using the high engine torque and closing the energy flow
through the motor/generator, which may cause more energy loss.
Similar to the series (electrical coupling) and parallel (mechanical coupling)
drive trains, the maximum torque on the driven wheels, corresponding to
full accelerator pedal position, full engine throttle opening, and full traction
motor load, dictates vehicle performance, such as acceleration and gradeability. On the other hand, with a partially depressed accelerator pedal (partial
load request), the engine or traction motor or both have to reduce their torques
to meet the traction torque demand. Thus, a control strategy is needed to
properly allocate the total load power to the power sources.
9.2.4
Drive Train Control Strategies
The distinct property of the above hybrid drive train is that the engine speed
and torque can be decoupled completely or partially from the driven wheels
through speed coupling and torque coupling. It also has much more flexibility
than series or parallel hybrid drive trains in the choice of the active operation
mode. Thus, this drive train has more potential for the improvement of drive
train efficiency and emissions. But it heavily relies on system control. There
are much more varieties of control strategies, due to more available operation
modes. Nevertheless, the control objectives are the same as in the series and
parallel drive trains, that is, high overall fuel utilization efficiency and low
emission under conditions of (1) always meeting the driver’s torque command (traction and braking) and (2) always maintaining the SOC of the PPS
at a reasonable level, for example, around 70% and never lower than 30%.
9.2.4.1
Engine Speed Control Strategy
The vehicle speed range is divided into three regions, low, medium, and high
as shown in Figures 9.18 and 9.19. When the vehicle speed is lower than
a given speed VL , the speed-coupling mode is used to avoid too low engine
speed. The vehicle speed VL is determined by the lowest engine speed allowed
with zero motor/generator speed (lock 1 locks the sun gear to the stationary
frame) and can be expressed as
VL =
πk yr rw ne-min
(m/s),
30irw
(9.14)
where ne-min is the minimum engine rpm allowed and rw is the wheel radius
of the vehicle in m.
In this low vehicle speed region, the motor/generator has to be operated
with a positive speed, which, from Equation 9.2, can be expressed as
30irw V
nm/g = kys ne-min −
,
(9.15)
πkyr rw
where V is the vehicle speed in m/s (V ≤ VL ).
324
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Low speed region
Medium speed region
High speed region
A
G
1
N
M
Tb-tm
Te
Te
Tt-com
4
2
VH
Vehicle speed
5
Tb-me
Tb-com
VL
Tch
Tch
K
Te
Te
F
H
Tb-tm
0
Tt-com
Ttm
Ttm
Tt-com
Tch
C
3
E
B
Te
Te
Torques on the driven wheels
D
L
1—The traction torque developed by the maximum
torque with optimal throttle
2—Traction torque developed by the maximum
traction motor torque
3—The maximum traction torque developed by both
the engine and traction motor
4—The traction torque developed by the engine with
partial throttle
5—The maximum braking torque developed by the
traction motor
Tt-com—Commanded traction torque by the driver
Tt-e—Traction torque developed by the engine torque
Tt-tm—Traction torque developed by the traction motor
Tch—Equivalent traction torque for PPS charging
Tb-com—Commanded braking torque
Tb-tm—Braking torque produced by the traction motor
Tb-me—Braking torque produced by mechanical brake
FIGURE 9.19 Schematic illustration of the MAX. SOC control strategy.
As indicated by Equation 9.3, the torque produced by the motor/generator,
applied to the sun gear of the planetary unit, has the direction opposite to
its speed. Therefore, in this case, the motor/generator absorbs part of the
engine power to charge the PPS. The power on the motor/generator shaft,
Design and Control Methodology of Hybrid Drive Train
325
with ignored losses, can be expressed as
Pm/g =
2π
2π
irw
Te V.
Tm/g nm/g =
Te ne-min −
60
60
kyr rw
(9.16)
The first term on the right-hand side is the power that the engine produces,
and the second term is the power that is delivered to the driven wheels.
When the vehicle speed is higher than VL , but lower than a given speed
VH , the motor/generator is de-energized and the sun gear (the shaft of
the motor/generator) is locked to the stationary frame of the vehicle. The
drive train operates in the torque-coupling mode. The engine speed is proportional to the vehicle speed. Speed VH is determined by the maximum
engine speed allowed ne-max , beyond which the engine operating efficiency
may be reduced. When the vehicle speed is higher than VH , the engine
speed is kept constant at ne-max , and the motor/generator starts working
again with a negative speed to compensate for the engine speed. VH can be
expressed as
πk yr rw ne-max
VH =
(m/s),
(9.17)
30irw
where ne-max is the maximum engine rpm allowed.
In this medium-speed region, all the engine power is delivered to the driven
wheels.
When the vehicle speed is higher than VH , for limiting the engine speed
below the maximum engine speed allowed ne-max , the motor/generator has
to be operated in the direction opposite to the engine speed, which can be
expressed as
30kys irw V
nm/g = kys ne-max −
,
(9.18)
πkyr rw
where V ≥ VH .
The motor/generator is in motoring. The motoring power can be
expressed as
Pm/g =
2π
irw
2π
Te V −
Tm/g nm/g =
Te ne-max .
60
kyr rw
60
(9.19)
The first term on the right-hand side is the total power delivered to the
driven wheels and the second term is the power that the engine produces.
The motor/generator accepts power from the PPS.
9.2.4.2 Traction Torque Control Strategy
Similar to the torque (power) control in the parallel hybrid drive train,
Figure 9.19 conceptually shows the allocation of the total traction torque
326
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
commanded by the driver to the engine (motor/generator) and the traction
motor, or the total braking torque commanded to the traction motor and the
mechanical braking system.
9.2.4.2.1 In Low Vehicle Speed Region
As mentioned above, when the vehicle speed is lower than VL , the engine is
operated at a specified speed, ne-min . The engine torque labeled 1 in Figure 9.19
is produced with an engine throttle at which the engine has maximum fuel
utilization efficiency at this speed. This engine throttle is possibly near its full
opening point.
Point A represents a traction torque commanded by the driver, which is
larger than the torque that the engine can produce with the optimal engine
throttle as shown in Figure 9.16. In this case, the engine alone cannot handle
this commanded torque, and needs the help of the traction motor. In this
case, the engine should be controlled with its optimal throttle as shown by
point B in Figure 9.19. However, the torque that the traction motor can produce depends on the energy level of the PPS. When the SOC of the PPS is
lower than a specified value SOCL (30% for example), the PPS should not be
further discharged. In this case, the maximum power of the traction motor
is the power generated by the motor/generator described by Equation 9.16.
Neglecting the losses, the traction motor torque can be expressed as
Tmt =
60 Pm/g
=
2π ntm
ne-min
irw
−
ntm
kyr imw
Te =
2πrw ne-min
irw
−
60imw V
imw kyr
Te ,
(9.20)
where imw is the gear ratio from the traction motor to the driven wheels
described by imw = (Z2 Z5 )/(Z3 Z4 ) as shown in Figure 9.6.
In this case, the planetary gear unit, the motor/generator, and the traction
motor together function as an EVT, because no energy goes into or comes out
of the PPS.
When the SOC is higher than the bottom line, SOCL , that is, the PPS has
sufficient energy to support the traction motor, the traction motor should
be controlled to produce its torque, Ttm , to meet the commanded traction
torque as shown in Figure 9.19. In this case, the PPS supplies its power to the
traction motor.
When the commanded torque, Tt-com , is smaller than the engine torque
produced with optimal throttle as shown by point B in Figure 9.19, there are
several options in choosing the engine and traction motor operations: (1) With
the SOC of the PPS lower than SOCL , the engine may be operated with the
speed of ne-min and optimal throttle (point B in Figure 9.19). The PPS is charged
by the motor/generator with the power of Pm/g (refer to Equation 9.16) and
the traction motor torque, Tch , as shown in Figure 9.19. (2) With the SOC of the
PPS in between the SOCL and SOCH (SOCL < SOC < SOCH ), the engine and
Design and Control Methodology of Hybrid Drive Train
327
the motor/generator may be controlled so that the engine operates with the
speed of ne-min and produces the torque that meets the commanded traction
torque. The traction motor is idling (de-energized). The PPS is charged only
by the motor/generator. (3) With the SOC of the PPS higher than SOCH , the
engine is shut down and the traction motor alone produces its torque to meet
the traction torque demand.
9.2.4.2.2 In Medium Vehicle Speed Region
When the vehicle speed is in a range higher than VL but lower than VH
as shown in Figures 9.18 and 9.19, only the torque-coupling (traditionally
parallel) mode is employed, that is, lock 1 locks the sun gear of the planetary
gear unit (the shaft of the motor/generator) to the stationary vehicle frame.
In this mode, engine speed is proportional to vehicle speed. The engine and
traction motor control strategy, based on the commanded traction torque and
SOC of the PPS, is exactly the same as that discussed in Chapter 8.
9.2.4.2.3
In High Vehicle Speed Region
When the vehicle speed is higher than VH , as shown in Figures 9.18 and
9.19, the engine speed is controlled at its top ne-max. In this case, the
motor/generator works in the motoring mode, taking energy from the PPS
and delivering it to the drive train. The motoring power is described by Equation 9.19. The torques of the engine and traction motor are controlled based
on the commanded traction torque and energy level of the PPS.
When the commanded traction torque, Tt-com (point G in Figure 9.19), is
larger than the torque that the engine can produce with its optimal throttle at
the speed of ne-max , and the SOC of the PPS is lower than SOCL , that is, the
PPS cannot be discharged any more to support the motoring operation of the
motor/generator and traction motor, the engine has to be forced to operate
with a speed higher than the specified speed, ne-max , to develop larger power.
In this case, there are two options: one is to use the engine-alone mode with
only torque coupling, which is the same as in the medium vehicle speed range;
the other is to control the engine to operate with a speed somewhat higher
than the speed that corresponds to the vehicle speed in the torque-coupling
mode. That is,
ne >
30irw V
.
πk yr rw
(9.21)
The term on the right-hand side is the engine speed that corresponds to
the vehicle speed V in the torque-coupling mode. In this way, the motor/
generator can be operated in its generating mode as discussed above. The generating power from the motor/generator can feed the traction motor to
generate additional traction torque. This operating mode is the EVT mode
as discussed above.
If the SOC of the PPS is at its medium and high levels, that is, SOC >
SOCL , the engine is controlled at its specified speed, ne-max , with the optimal
328
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
engine throttle (point H in Figure 9.19). The traction motor produces its torque,
together with the engine torque, to meet the commanded traction torque.
In case the commanded traction torque is smaller than the engine torque
with the optimal throttle as shown by point K in Figure 9.19, and the SOC of
the PPS is below SOCL , the engine is operated at point K and the traction motor
works in its generating mode to charge the PPS. If the SOC is in the medium
region (SOCL < SOC < SOCH ), the traction motor may be de-energized and
the engine alone propels the vehicle (point K). If the SOC of the PPS is at
a high level (SOC > SOCH ), the engine may be shut down and the traction
motor alone propels the vehicle.
9.2.4.3
Regenerative Braking Control
Similar to the parallel drive train control, when the commanded braking
torque is larger than the maximum torque that the motor can produce in generating mode, both regenerative braking by the traction motor and mechanical
braking are applied. Otherwise only regenerative braking is applied. For a
detailed discussion, refer to Chapter 13.
It should be noted that the control strategies discussed above are only for
guidance in a real control strategy design. More careful and insightful studies are necessary based on the special design constraints, design objectives,
component characteristics, operation environments, and so on. More complicated and subtle approaches may be employed, such as fuzzy logic, dynamic
programming, and so on. Further, computer simulations are very useful in
designing a good control strategy.
9.3
Drive Train Parameters Design
The design principles of drive train parameters, such as engine power, motor
power, and the power and energy capacity of the PPS, are very similar to those
in series and parallel drive trains discussed in Chapters 7 and 8. Therefore,
they are not discussed further in this chapter. However, the torque and power
capacity design of the motor/generator may need further discussion.
From Equations 9.3 and 9.4 and Figure 9.5, it can be seen that the
motor/regenerator torque is required to balance the engine torque at nearly
full throttle, in the speed regions of lower than the minimum speed, ne-min ,
and higher than the maximum speed, ne-max. Thus, the torque capacity of the
motor/generator is determined by the maximum engine torque in the lowspeed and high-speed regions. However, for purposes of safety, the torque
capacity of the motor/generator would be designed to be able to balance
the maximum torque of the engine in its entire speed range. Figure 9.5
Design and Control Methodology of Hybrid Drive Train
329
also indicates that this maximum motor torque should be available in its
entire speed range, rather than a special point. Thus, the ideal torque–speed
characteristic is a constant torque in its entire speed range, which can be
expressed as
Tm/g-max =
Te-max
.
kys
(9.22)
Here, Te-max is the maximum engine torque with fully open throttle.
From Equation 9.16, it is obvious that the generating power of the
motor/generator is maximized at zero vehicle speed, that is, all the power
produced by the engine goes to the motor/generator. Thus, the maximum
generating power of the motor/generator can be determined by
Pm/g-max =
2π
Te-max ne-min .
60
(9.23)
Similarly, the maximum motoring power of the motor/generator occurs at
the maximum vehicle speed, Vmax , as indicated by Equation 9.19, which can
be expressed as
Pm/g-ma =
irw
2π
Te-max Vmax −
Te-max ne-max .
kyr rw
60
(9.24)
9.4 Simulation of an Example Vehicle
Based on the design and control principles discussed in previous sections,
a 1500-kg passenger car has been simulated in FTP75 urban and highway
driving cycles. The parameters of the vehicle simulated are listed in Table 9.1.
Figure 9.20 shows the simulation results of vehicle speed, engine power,
motor/generator power, traction motor power, and SOC of the PPS in an
FTP urban driving cycle. It can be seen that the motor/generator always
TABLE 9.1
Vehicle Parameters
Vehicle mass
Engine power
Traction motor power
Generator motor power
Tire rolling resistance coefficient
Aerodynamic drag coefficient
Front area
1500 kg
28 kW
40 kW
15 kW
0.01
0.3
2.2 m2
330
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
100
Vehicle speed (km/h)
50
0
20
Engine power (kW)
10
0
10
Generator/motor power (kW)
0
–10
20
Traction motor power (kW)
0
–20
1
Battery SOC
0.8
0.6
0
200
400
600
800
Time (s)
1000
1200
1400
FIGURE 9.20 Vehicle speed, engine power, generator/motor power, traction motor power, and
battery SOC in an FTP75 urban driving cycle.
works in the generating mode (negative power) because of the low vehicle
speeds. Through the regenerative braking and charging from the engine by
the motor/generator, the PPS SOC can be easily maintained at a high level,
which ensures that the PPS is always able to supply sufficient power to the
drive train for acceleration.
Figure 9.21 shows the engine operating points on the engine fuel consumption map. This figure indicates that the engine, most of the time, operates
in its high-efficiency area. Engine-alone traction with light load and high
SOC of the PPS causes some engine operating points away from its highefficiency area. The fuel consumption of the vehicle in an FTP urban driving
cycle obtained from the simulation is 5.88 L per 100 km or 40.2 mpg.
Figures 9.22 and 9.23 show the simulation results while driving in an
FTP75 highway cycle. It can be seen that the generator/motor power is zero,
except in a short time period of cycle start. This means that the drive train,
most of the time, worked with a pure torque coupling (the sun gear and its
motor/generator are locked to the vehicle frame). Simulation indicates that
the fuel consumption of the vehicle in an FTP75 highway driving cycle is
4.96 L/100 km or 47.7 mpg.
331
Design and Control Methodology of Hybrid Drive Train
Engine power (kW)
80
3.1 6.2 9.3 12.4 15.6 18.7 21.8 24.9 28
70
Operating
points
Brake specific fuel
consumption, g/kWh
(efficiency)
250 (34%)
Engine torque (Nm)
60
50
%)
260 (33
%)
270 (32 %)
31
280 (
28%)
310 (
40
)
350
30
(24%
400
20
17%)
500 (
14%)
600 ( 2%)
700 (10 (11%)
80
%)
1000 (8.6
10
0
%)
(21
0
1000
2000
3000
Engine speed (rpm)
4000
5000
FIGURE 9.21 Engine operating points on its fuel consumption map in an FTP75 urban driving
cycle.
100
50
Vehicle speed (km/h)
0
20
10
Engine power (kW)
0
10
Generator/motor power (kW)
0
–10
20
Traction motor power (kW)
0
–20
1
0.8
0.6
Battery SOC
0
100
200
300
400
500
Time (s)
600
700
800
FIGURE 9.22 Vehicle speed, engine power generator/motor power, traction motor power, and
battery SOC in an FTP75 highway driving cycle.
332
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Engine power (kW)
3.1 6.2 9.3 12.4 15.6 18.7 21.8 24.9 28
Operating
points
80
70
250 (34%)
60
Engine torque (Nm)
Brake specific fuel
consumption, g/kWg
(efficiency)
260 (33%)
%)
270 (32 (31%)
280
50
310
40
)
(28%
350
%)
(24
30
400
20
500
10
0
0
1000
2000
3000
Engine speed (rpm)
%)
(21
4000
)
(17%
14%)
600 (
%)
700 (12 %)
800 (11 6%)
00
10 (8.
5000
FIGURE 9.23 Engine operating points on its fuel consumption map in an FTP75 highway
driving cycle.
References
1. Y. Gao and M. Ehsani, “A torque and speed coupling hybrid drive train—
architecture, control, and simulation,” IEEE Transactions on Power Electronics,
21 (3), 741–748, May 2006.
2. C. C. Chan and K. T. Chau, Modern Electric Vehicle Technology, Oxford University
Press, New York, 2001.
3. I. Husani, Electric and Hybrid Vehicles—Design and Fundamentals, CRC Press LLC,
New York, 2003.
4. M. Ehsani, Y. Gao, and K. Butler, “Application of electrically peaking hybrid
(ELPH) propulsion system to a full size passenger car with simulated design
verification,” IEEE Transaction On Vehicular Technology, 48 (6), November 1999.
5. Nedunadi, M. Walls, and D. Dardalis, “A parallel hybrid drive train,” SAE, SP1466, Paper No. 1999-01-2928.
6. K. Yamaguchi, S. Moroto, K. Kobayashi, M. Kawamto, and Y, Miyaishi, “Development of a new hybrid system-duel system,” SAE, SP-1156, Paper No. 960231,
1996.
10
Design and Control Principles of Plug-In
Hybrid Electric Vehicles
As discussed in the previous chapters, in the PPS charge sustained hybrid
drive train, the net energy consumption in PPS in a complete driving cycle is
zero, that is, the energy level in the PPS at the beginning of the driving cycle
is equal to the energy level at the end of the driving cycle. All the propulsion energy comes from the primary energy source: gasoline or diesel for IC
engines; hydrogen or hydrogen-based fuel for fuel cells. During operation,
the energy in the PPS fluctuates in a narrow window. The PPS size is determined by power rather than energy capacity. The energy-to-power ratio is in
the range of 0.05–0.1 kWh/kW. That is to say, with a given power capacity, the
energy storage in PPS is considered to be sufficient if it can sustain 0.05–0.1 h
with the given power. Thus, the PPS is more an energy buffer than energy
storage. This is also the origin of the name PPS (peaking power source). At
present and in the immediate future, ultracapacitors and high-power batteries or their combination are the most promising candidates as the PPS of the
PPS charge sustained HEVs (for details, refer to Chapter 12 of Peaking Power
Source and Energy Storage).
With the development and maturing of advanced battery technologies, the
energy storage capacity of batteries has significantly improved. Obviously,
using high-energy batteries only as a PPS is a waste.
The plug-in hybrid electric drive train is designed to fully or partially use
the energy of the energy storage to displace part of the primary energy source,
such as gasoline, diesel, hydrogen, and so on.
All the configurations discussed in Chapter 5 can be employed in plug-in
hybrid electric drive trains. Most of the differences from the PPS sustained
hybrid drive train are in the drive train control strategy, energy storage design,
and perhaps, slightly different electric motor power design. This chapter will
concentrate on these three topics.
10.1
Statistics of Daily Driving Distance
Charging the energy in the energy storage device from the utility grid, to
displace part of the petroleum fuel, is the major feature of the plug-in hybrid
333
334
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
electric vehicles (PHEVs). The amount of petroleum fuel displaced by the
utility electricity depends mainly on the amount of electrical energy per
recharge, that is, the energy capacity of the energy storage; total driving distance between recharges, that is, usual daily driving distance; and electrical
power usage profiles, that is, the driving cycle features and control strategies. To achieve optimal design, especially for the energy storage system,
understanding the daily driving distance in a typical environment is very
helpful.
Figure 10.1 is a histogram showing the daily driving distance distribution and the cumulative frequency derived from the 1995 National Personal
Transportation Survey data.1,2 The cumulative frequency or utility factor in
reference1 represents the percentages of the total driving time (days) during
which the daily driving distances are less than or equal to the said distance
on the horizontal axis. Figure 10.1 reveals the fact that about half of the daily
driving distance is less than 64 km (40 miles). If a vehicle is designed to
have 64 km (40 miles) of pure EV range, that vehicle will have half of its
total driving distance from the pure EV mode. Even if the daily traveling
distance is beyond this 60 km (40 miles) pure EV range, a large amount of
the petroleum fuel can be displaced by electricity, due to the pure EV mode
taking a large portion of the daily travel. Research also shows that even if the
pure EV range is less than 64 km (40 miles), such as 32 km (20 miles), there
is still a large amount of petroleum that can be displaced in normal daily
driving.1
90
Percentage of trips
Cumulative frequency
80
70
60
50
40
30
20
10
0
0
0
16
10
32
20
48
30
64
40
80
50
96 112 128 144 160 176 192 208 224 240 km
60 70 80 90 100 110 120 130 140 150 miles
Trip distance
FIGURE 10.1 Daily driving distance distribution and cumulative factor.
Design and Control Principles of Plug-In Hybrid Electric Vehicles
10.2
335
Energy Management Strategy
First, some definitions about PHEV are introduced:
•
Charge-Depleting (CD) Mode: An operating mode in which the SOC
of the energy storage may fluctuate, but on average decreases while
driving.
•
Charge-Sustaining (CS) Mode: An operating mode in which the SOC
of the energy storage may fluctuate but on average is maintained at
a certain level while driving.
•
All Electric Range (AER): After a full recharge, the total miles (kilometers) driven electrically (engine off) before the engine turns on for
the first time.
•
Electric Vehicle Miles (EVM) or Kilometers (EVKM): After a full
recharge, the cumulative miles or kilometers driven electrically
(engine off) before the vehicle reaches CS mode.
•
Charge-Depleting Range (CDR): After a full recharge, the total miles or
kilometers driven before the vehicle reaches CS mode. It should be
noted that EVM or EVKM dictates pure electric driving. However,
CDR may include engine propulsion, but the on-average SOC of the
energy storage decreases till the sustaining level.
•
PHExx: A PHEV with useable energy storage equivalent to xx miles
of driving energy on a reference driving cycle, where xx stands for
the mileage number. For example, PHEV20 can displace petroleum
energy equivalent to 20 miles of driving on the reference driving
cycle with off-board electricity. A similar definition can be made in
kilometers. It should be noted that PHEV20, for example, does not
imply that the vehicle will achieve 20 miles of AER, EVM, or CDR
on the reference cycle, nor any other cycle. Operating characteristics
also depend on the power ratings of components, the power train
control strategy, and the nature of the driving cycle.1
10.2.1 AER-Focused Control Strategy
The idea of this control strategy is to use the energy of the energy storage
intensively in the AER.3,4 One possibility is to allow the driver to manually
select between a CS mode and a full EV operating mode. This design could be
useful for vehicles that may be used in the region where combustion engine
use is restricted. This design provides flexibility for the driver to determine
the times that the pure EV mode is used. For example, in a trip that includes
places where pure EV operation is required, the driver can select the pure
EV operating mode just prior to entering this area in order to have sufficient
336
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
range. In other places, the vehicle may be operated in pure EV mode or CS
mode, depending on the energy status of the energy storage and the power
demand. In normal conditions, where the trip does not include mandatory
pure EV operation, the driver could select the pure EV mode at the start of
the trip in order to fully use the energy of the energy storage to displace the
petroleum fuel, until the energy of the energy storage reaches its specified
level at which the CS mode will start automatically.
This energy management approach clearly divides the whole trip into pure
EV and CS modes. Thus, the design and control techniques developed for EV
and HEV in the previous chapters can be used. When series hybrid configuration is used, the power rating designs of the motor, engine, and energy
storage are almost the same as in the CS hybrid. The motor power guarantees
the acceleration and gradeability performance, the engine/generator power
supports the vehicle driving at a constant speed on flat or mild grades, and
the energy storage power is larger (or at least not smaller) than the motor
power minus the engine/generator power. However, the energy storage has
to be designed so that its useable energy can meet the requirement of the pure
EV range. When parallel or series/parallel configuration is used, the motor
power should be designed to meet the peaking power requirements of the
reference driving cycles. Otherwise, the vehicle cannot follow the speed profile of the drive cycle, and will be somewhat sluggish, compared to the driver
expectation.
The traction power computations in typical driving cycles have been discussed in detail in previous chapters. However, for the reader’s convenience,
it is repeated below.
The traction power on the driven wheels includes the rolling resistance,
aerodynamic drag, inertial force of acceleration, and grade resistance, which
can be expressed as
Pt =
V
1
dV
Mg fr + ρa CD Af V 2 + Mδ
+ Mgi (kW),
1000
2
dt
(10.1)
where M is the vehicle mass in kg, V is the vehicle speed in m/s, g is the
gravity acceleration, 9.81 m/s2 , ρa is the air mass density, 1.205 kg/m3 , CD
is the aerodynamic drag coefficient of the vehicle, Af is the front area of the
vehicle in m2 , δ is the rotational inertia factor, dV/dt is the acceleration in
m/s2 , and i is the road grade. In standard driving cycles, flat roads are used.
Figure 10.2 is a diagram showing the vehicle speed and the traction power,
on the driven wheels, versus the traveling distance in the FTP75 urban driving
cycle. The vehicle parameters used in this computation are listed in Table 10.1.
Figure 10.2 indicates that the peaking traction power on the driven wheels is
about 25 kW. However, there are power losses in the path from the energy
storage to the driven wheels. In order to meet the power requirement, the
motor output power should be designed to account for the power losses from
the motor shaft to the driven wheels. Suppose that the efficiency from the
337
Design and Control Principles of Plug-In Hybrid Electric Vehicles
100
80
60
40
20
0
Power on the driven
wheels (kW)
30
20
10
0
–10
–20
0
0
1
1.6
2
3.2
3
4.8
4
6.4
5
8.0
6
9.6
7 miles
11.2 km
Driving distance
FIGURE 10.2 Vehicle speed and traction power in an FTP75 urban driving cycle.
motor shaft to the driven wheels is 90%; then the motor shaft power rating is
about 28 kW. It should be noted that this required motor power is also related
to the vehicle speed at which this peak power occurs. For example, the peaking
power in Figure 10.2 occurs at a vehicle speed of 50 km/h (31.25 mph). In
the motor power design, we must be sure that the motor can produce this
peak power at this vehicle speed. Similarly, the peaking power of the energy
storage should include the losses in the electric motor, the power electronics,
and the transmission. Suppose that the efficiencies of the motor and power
electronics are 0.85 and 0.95, respectively; then the power capacity of the
energy storage is about 34.7 kW in this example. Table 10.2 lists the motor
power and the energy storage power in FTP75 urban, FTP75 highway, LA92,
and US06 driving cycles.
TABLE 10.1
Vehicle Parameters Used in Power Computation
Vehicle mass (kg)
Rolling resistance coefficient
Aerodynamic drag coefficient
Front area (m2 )
Rotational inertia factor
1700
0.01
0.3
2.2
1.05
338
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
TABLE 10.2
Powers of Motor and Energy Storage in Typical Driving Cycles
Cycles Power
(kW) Item
FTP75 Urban
FTP75 Highway
LA92
US06
Motor at vehicle
speed
28 at 50 km/h
(31 mph)
32 at 72 km/h
(57 mph)
55 at 57 km/h
(36 mph)
98 at 117 km/h
(73 mph)
Energy storage
35.7
39
68.5
121
Integrating Equation 10.1, over the driving time in a driving cycle, can give
the energy consumption by the driven wheels as shown in Figure 10.3. Here,
no regenerative braking is included. When including energy losses in the
power electronics, the motor, and the transmission, the useable energy in the
energy storage, for 32 km (20 miles) and 64 km (40 miles) of pure EV driving
in typical driving cycles, is listed in Table 10.3.
In vehicle design, an appropriate reference driving cycle should be selected.
An aggressive driving cycle, such as US06, will need a large motor drive
and energy storage, but will also give good vehicle acceleration and gradeability performance. On the contrary, a mild driving cycle, such as FTP75, will
lead to a small motor drive and energy storage, but also a sluggish vehicle
performance.
The following figures show simulation results of the drive train in the reference driving cycle, FTP75 urban. The vehicle parameters listed in Table 10.1
Energy consumption on wheels (kWh)
20
18
US06
16
LA92
14
FTP75
urban
12
10
FTP75
highway
8
6
4
2
0
0
0
16
10
32
20
48
30
64
40
80
50
96 km
60 miles
FIGURE 10.3 Energy consumption by the driven wheels versus driving distance in typical
driving cycles.
339
Design and Control Principles of Plug-In Hybrid Electric Vehicles
TABLE 10.3
Energy Consumption in Typical Driving Cycles
Cycles Energy (kWh) Distance
FTP75 Urban
FTP75 Highway
LA92
US06
32 km (20 miles)
5.2
5.14
7.29
8.4
64 km (40 miles)
10.4
10.28
14.58
16.8
were used. The total energy in the energy storage, fully charged, is 10 kWh.
The simulation ran nine sequential cycles and the pure EV mode was started
at the beginning of the simulation, until the SOC reached about 30%, beyond
which the CS mode was started. The control strategy in the CS mode employed
the constrained engine on–off control strategy, discussed in Section 8.2.3. In
the simulation, 400 W of constant auxiliary power was added at the terminal
of the energy storage.
Figures 10.4 and 10.5 show the engine power and the motor power.
Figure 10.6 shows the SOC of the energy storage, and the remaining energy
in the energy storage, versus the traveling distance. The pure EV mode range
is about 32 km (20 miles). Figure 10.7 shows the engine operating points
overlapping its brake-specific fuel consumption map.
Figures 10.8 and 10.9 show the fuel and electric energy consumption scenarios, in metric and English units, respectively. It can be seen that when the
traveling distance is less than four driving cycles (42.5 km or 26.6 miles), the
vehicle can completely displace the petroleum fuel with electricity with pure
EV mode. The total electric energy consumed is about 7.1 and 15.5 kWh per
100 km, or 4.05 miles/kWh (Figure 10.9). With the increasing total traveling distance, the percentage of the fuel displacement decreases, since the CS
modes take larger percentages of the trip. For nine sequential driving cycles
(96 km or 60 miles), the fuel and electrical energy consumptions are about
3.2 L/100 km (Figure 10.8) or 74 mpg (Figure 10.9), and 7.42 kWh/100 km
(Figure 10.8) or 8.43 mile/kWh (Figure 10.9).
Engine power (kW)
40
30
20
10
0
0
0
10
20
30
6.25 12.5 18.75
40
50
60
70
25 31.25 37.5 43.75
Traveling distance
80
50
90 100 km
56.25 62.5 miles
FIGURE 10.4 Engine power versus traveling distance in FTP75 urban driving cycle with AER
mode.
340
Motor power (kW)
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
40
30
20
10
0
–10
–20
–30
–40
0
0
10
20
30
6.25 12.5 18.75
40
50
60
70
25 31.25 37.5 43.75
Traveling distance
80
50
90 100 km
56.25 62.5 miles
FIGURE 10.5 Motor power versus traveling distance in FTP75 urban driving cycle with AER
mode.
10
0.8
8
0.6
6
0.4
4
0.2
2
0
0
0
10
6.25
20
12.5
30
18.75
40
50
60
70
25 31.25 37.5 43.75
Traveling distance
80
50
90
56.25
Remaining energy in the
energy storage (kWh)
SOC
1
0
100 km
62.5 miles
FIGURE 10.6 SOC and the remaining energy in the energy storage versus traveling distance in
FTP75 urban driving cycle with AER mode.
Power (kW)
7 13 19 25 31 37 43 49 55 61 67
Maximum torque
with full throttle
bsfc (g/kWh) and
fuel efficiency
180
160
250 (32.7%)
5%)
260 (31.
3%)
270 (30.
2%)
280 (29.
Torque (Nm)
140
120
100
6.4%)
310 (2
80
350
60
400
%)
(20.5
6.4%)
500 (1
6%)
3.
600 (1
7%)
700 (11.
1000 (8.18%)
40
20
0
%)
(23.4
0
500 1000 1500 2000 2500 3000 3500 4000 4500 5000
Speed (rpm)
FIGURE 10.7 Engine operating points overlapping its fuel consumption map in FTP75 urban
driving cycle with AER mode.
Design and Control Principles of Plug-In Hybrid Electric Vehicles
341
20
kWh/100 km
15
10
Total kWh
5
Liters/100 km
0
0
0
0
1
10.6
6.64
2
21.2
13.3
Total fuel liters
3
4
5
6
7
9 Cycle number
8
31.9 42.5 53.1 63.8 74.4 85.0 95.6 km
19.9 26.6 33.2 39.9 54.7 53.1 59.8 miles
Cycle number and traveling distance
FIGURE 10.8 Fuel and electric energy consumption versus the number of FTP75 urban driving
cycle and traveling distance with AER mode in metric unit.
Simulation of the same design in the LA92 driving cycle has also been
performed. The results are illustrated in the following figures (Figures 10.10
through 10.15). Comparing the two driving cycles, the LA92 driving cycle
has higher vehicle speed and larger acceleration rate. The pure EV range is
shorter, and the fuel and electric energy consumptions are higher than in
FTP75 urban driving cycles.
10.2.2
Blended Control Strategy
Unlike the AER-focused control strategy, in which a pure EV range is
designed, the blended control strategy uses both the engine and the motor for
traction, with CD mode, until the SOC of the energy storage reaches the specified low threshold, beyond which the vehicle will operate in the CS mode.
400
20
mpg
15
300
10
200
miles/kWh
5
100
Total fuel gallons
0
0
0
0
1
10.6
6.64
2
3
4
5
6
7
8
21.2 31.9 42.5 53.1 63.8 74.4 85.0
13.3 19.9 26.6 33.2 39.9 54.7 53.1
Cycle number and traveling distance
0
9
Cycle number
95.6 km
59.8 miles
FIGURE 10.9 Fuel and electric energy consumption versus the number of FTP75 urban driving
cycle and traveling distance with AER mode in English unit.
342
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
50
Engine power (kW)
40
30
20
10
0
FIGURE 10.10
0
0
20
12.5
40
25
60
80
37.5
50
Traveling distance
100
62.5
120
75
Engine power versus traveling distance in LA92 driving cycle with AER mode.
Motor power (kW)
60
40
20
0
–20
–40
–60
0
0
20
12.5
40
25
60
37.5
80
50
100
62.5
120
75
Traveling distance
FIGURE 10.11 Motor power versus traveling distance in LA92 driving cycle with AER mode.
10
0.8
8
0.6
6
0.4
4
0.2
2
0
0
0
20
12.5
40
25
60
37.5
80
50
100
62.5
Remaining energy in the
energy storage (kWh)
SOC
1
0
120
75
Traveling distance
FIGURE 10.12 SOC and the remaining energy in the energy storage versus traveling distance
in LA92 driving cycle with AER mode.
Design and Control Principles of Plug-In Hybrid Electric Vehicles
343
Power (kW)
7 13 19 25 31 37 43 49 55 61 67
Maximum torque
with full throttle
180
bsfc (g/kWh) and
fuel efficiency
160
250 (32.7%)
140
260 (31.5%)
270 (30.3%)
280 (29.2%)
120
100
6.4%)
310 (2
80
%)
350
60
(23.4
%)
400
(20.5
6.4%)
500 (1
6%)
600 (13.
7%)
700 (11.
1000 (8.18%)
40
20
0
0
500 1000 1500 2000 2500 3000 3500 4000 4500 5000
Speed (rpm)
FIGURE 10.13 Engine operating points overlapping its fuel consumption map in LA92 driving
cycle with AER mode.
In the CD mode, both the engine and the motor may operate at the same
time. The range before entering the CS mode is longer than in the pure EV
mode. Control strategies are needed to control the engine and the motor to
meet the load demand. There are many possible control strategies. The following is the one in which the engine and the motor alternately propel the
vehicle with no battery charging from the engine. The engine is constrained
to operate in its optimal fuel economy region. The details are as follows.
25
kWh/100 km
20
15
10
Total kWh
5
Liters/100 km
Total fuel liters
0
0
0
0
1
15.7
9.8
2
31.4
19.6
3
47.1
29.4
4
62.8
39.2
5
78.5
48.9
6
94.2
58.9
7 Cycle number
110 km
68.8 miles
Cycle number and traveling distance
FIGURE 10.14 Fuel and electric energy consumption versus the number of LA92 driving cycle
and traveling distance with AER mode in metric unit.
344
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
14
140
mpg
12
120
10
100
8
80
6
Miles/kWh
4
40
Total fuel gallons
2
0
60
0
0
0
1
15.7
9.8
2
31.4
19.6
3
47.1
29.4
4
62.8
39.2
5
78.5
48.9
6
94.2
58.9
20
0
7 Cycle number
110 km
68.8 miles
Cycle number and traveling distance
FIGURE 10.15 Fuel and electric energy consumption versus the number of LA92 driving cycle
and traveling distance with AER mode in English unit.
Figure 10.16 schematically shows the engine operating area. When the
requested engine torque is larger than the upper torque boundary, the engine
is controlled to operate on this boundary and the remaining torque is supplied
by the electric motor. When the requested engine torque falls below the upper
boundary, the engine alone propels the vehicle. When the requested engine
torque is below the lower torque boundary, the engine is shut down and the
electric motor alone propels the vehicle. In this way, the engine operation is
constrained within its optimal region. Due to the absence of battery charging
Maximum torque
with full throttle
ine
ope
rati
nsu
l co
Fue
ng
are
a
tio
mp
Engine torque
Eng
Iso-fuel
consumption
curves
ase
cre
n in
Engine speed
FIGURE 10.16
Operation area of the engine in the CD mode.
Top
torque
bound
Bottom
torque
bound
Design and Control Principles of Plug-In Hybrid Electric Vehicles
345
Engine power (kW)
40
30
20
10
0
0
0
10
20
30
6.25 12.5 18.75
40
25
50
60
70
31.25 37.5 43.75
80
50
90 100 km
56.25 62.5 miles
Traveling distance
FIGURE 10.17 Engine power versus traveling distance in FTP75 urban driving cycle with
CD mode.
from the engine, the battery energy level will continuously fall to its specified
lower level. Then the drive train goes into CS mode.
The example vehicle mentioned above has been simulated with the control
strategy discussed above in nine sequences of the FTP75 urban driving cycle.
The results are shown in Figures 10.17 through 10.22.
Similarly, simulation results in seven sequences of the LA92 driving cycles
are shown in Figures 10.23 through 10.28.
It should be noted that the pure EV range is mostly determined by the
capacity of the energy storage and its SOC level, at which the CS mode started.
The range in the CD mode is also related to the drive train control strategy,
especially the specified engine operating region as shown in Figure 10.16.
When the engine lower torque boundary is moved downward, that is, the
engine operating area is enlarged, the CD mode range is increased. However,
the fuel displacement is reduced when the travel distance between chargings
Motor power (kW)
40
20
0
–20
–40
0
0
10
20
30
6.25 12.5 18.75
40
50
60
70
25 31.25 37.5 43.75
Traveling distance
80
50
90 100 km
56.25 62.5 miles
FIGURE 10.18 Motor power versus traveling distance in FTP75 urban driving cycle with CD
mode.
346
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
10
0.8
8
0.6
6
0.4
4
0.2
2
0
0
0
10
20
30
6.25 12.5 18.75
40
25
50
60
70
31.25 37.5 43.75
Remaining energy in the
energy storage (kWh)
1
0
90 100 km
56.25 62.5 miles
80
50
FIGURE 10.19 SOC and the remaining energy in the energy storage versus traveling distance
in FTP75 urban driving cycle with CD mode.
is shorter than the full CD mode range, and the SOC of the energy storage
does not hit its lower limit, leaving useable energy in the energy storage.
10.3
Energy Storage Design
Energy storage is one of the most important components in the plug-in hybrid
vehicle. It is closely related to vehicle performance, fuel consumption, fuel
Power (kW)
7 13 19 25 31 37 43 49 55 61 67
Maximum torque
with full throttle
180
bsfc (g/kWh) and
fuel efficiency
160
7%)
250 (32.
Torque (Nm)
140
120
100
6.4%)
310 (2
%)
80
350
(23.4
)
20.5%
60
400 (
6.4%)
500 (1
6%)
3.
(1
0
60
7%)
700 (11. 18%)
1000 (8.
40
20
0
1.5%)
0.3%)
270 (3
9.2%)
280 (2
260 (3
0
500 1000 1500 2000 2500 3000 3500 4000 4500 5000
Speed (rpm)
FIGURE 10.20 Engine operating points overlapping its fuel consumption map in FTP75 urban
driving cycle with CD mode.
347
Design and Control Principles of Plug-In Hybrid Electric Vehicles
15
kWh/100 km
10
Total kWh
5
Liters/100 km
0
0
0
0
el liters
Total fu
1
2
3
4
5
6
7
8
9 Cycle number
10.6 21.2 31.9 42.5 53.1 63.8 74.4 85.0 95.6 km
6.64 13.3 19.9 26.6 33.2 39.9 54.7 53.1 59.8 miles
Cycle number and traveling distance
FIGURE 10.21 Fuel and electric energy consumption versus the number of FTP75 urban driving
cycle and traveling distance with CD mode in metric unit.
15
300
mpg
10
200
miles/kWh
5
0
100
Total fuel gallons
0
1
2
3
4
5
6
7
8
9 Cycle number
10.6 21.2 31.9 42.5 53.1 63.8 74.4 85.0 95.6 km
6.64 13.3 19.9 26.6 33.2 39.9 54.7 53.1 59.8 miles
Cycle number and traveling distance
0
0
0
FIGURE 10.22 Fuel and electric energy consumption versus the number of FTP75 urban driving
cycle and traveling distance with CD mode in English unit.
Engine power (kW)
50
40
30
20
10
0
0
0
20
12.5
40
25
60
80
37.5
50
Traveling distance
100
62.5
120 km
75 miles
FIGURE 10.23 Engine power versus traveling distance in LA92 driving cycle with CD mode.
348
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Motor power (kW)
60
40
20
0
–20
–40
–60
0
0
20
12.5
40
25
60
80
37.5
50
Traveling distance
100
62.5
120 km
75 miles
FIGURE 10.24 Motor power versus traveling distance in LA92 driving cycle with CD mode.
displacement, initial cost, and operation cost. The most important parameters
in the energy storage design are the storage energy and power capacities. By
simulation, similar to the above, the useable energy in the energy storage can
be determined. The total energy capacity can be obtained by
Ec =
Eusable
,
SOCtop − SOCbottom
(10.2)
where Eusable is the useable energy, in the energy storage, consumed in the
pure EV or CD modes, SOCtop is the top SOC with fully charged energy
storage (which is usually equal to 1), and SOCbottom is the SOC of the energy
storage at which the operation mode is switched from the pure EV or CD
modes to the CS mode. In the above example, the useable energy is about
7 kWh (refer to Figures 10.6, 10.8, 10.12, 10.14, 10.19, 10.21, 10.25, and 10.27)
10
0.8
8
0.6
6
0.4
4
0.2
2
0
0
0
20
12.5
40
25
60
80
37.5
50
Traveling distance
100
62.5
Remaining energy in the
energy storage (kWh)
SOC
1
0
120 km
75 miles
FIGURE 10.25 SOC and the remaining energy in the energy storage versus traveling distance
in LA92 driving cycle with CD mode.
Design and Control Principles of Plug-In Hybrid Electric Vehicles
349
Power (kW)
7 13 19 25 31 37 43 49 55 61 67
Maximum torque
with full throttle
180
bsfc (g/kWh) and
fuel efficiency
160
250 (32.7%)
Torque (Nm)
140
260 (31.5%)
270 (30.3%)
280 (29.2%)
120
100
310 (2
80
6.4%)
%)
350
60
(23.4
%)
400
6.4%)
6%)
600 (13.
7%)
700 (11.
(8.
1000 18%)
40
500 (1
20
0
(20.5
0
500 1000 1500 2000 2500 3000 3500 4000 4500 5000
Speed (rpm)
FIGURE 10.26 Engine operating points overlapping its fuel consumption map in LA92 driving
cycle with CD mode.
and the SOC operating window is 0.7 (from 1 to 0.3). The total energy capacity
of the energy storage is about 10 kWh.
It should be noted that the depth of discharge (DOD) of batteries is closely
related to battery life. Figure 10.29 illustrates the battery cycle life with the
DOD.5,6 If one deep discharge per day is assumed, a total of 4000+ deep
charges would be required for a 10–15-year lifetime. With the characteristics
shown in Figure 10.29, a 70% DOD, for NiMH, and a 50% DOD, for Li-ion
batteries, may be the proper designs.
15
kWh/100 km
10
Total kWh
5
Liters/100 km
fuel
Total
0
0
0
0
1
15.7
9.8
liters
2
3
4
5
6
31.4
47.1
62.8
78.5
94.2
19.6
29.4
39.2
48.9
58.9
Cycle number and traveling distance
7 Cycle number
110 km
68.8 miles
FIGURE 10.27 Fuel and electric energy consumption versus the number of LA92 driving cycle
and traveling distance with CD mode in metric unit.
350
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
15
300
10
200
miles/kWh
5
100
mpg
Total fuel gallons
0
0
0
0
1
15.7
9.8
0
2
3
4
5
6
7 Cycle number
31.4 47.1 62.8 78.5 94.2 110 km
19.6 29.4 39.2 48.9 58.9 68.8 miles
Cycle number and traveling distance
FIGURE 10.28 Fuel and electric energy consumption versus the number of LA92 driving cycle
and traveling distance with CD mode in English unit.
The power requirement of the energy storage is completely determined by
the electric motor power rating, which can be expressed as
Pes ≥
Pm
,
ηm ηpe
(10.3)
where Pm is the motor power rating, measured on the motor shaft, and ηm
and ηpe are the efficiencies of the motor and the power electronics between
the energy storage and the motor. This power should be designed to work at
100
90
Typical PHEV cycles
during life of vehicle
Depth of discharge (%)
80
70
60
NiHM
50
Li-oin
40
Typical HEV cycles
operating window
30
20
10
0
10,000
20,000
Number of cycles
30,000
FIGURE 10.29 Cycle life characteristics of Varta energy storage technologies.5
40,000
351
Design and Control Principles of Plug-In Hybrid Electric Vehicles
NiMH, Cobasys
Energy/power ratio (h)
1
Li-ion, SAFT
0.8
0.6
0.4
0.2
0
FIGURE 10.30
0
0.2
0.4 0.465 0.6
0.8
1
Specific power, kW/kg
1.08
1.2
1.4
Typical energy/power ratios versus specific power.
low SOC levels, such as 30%, since the energy storage always works at this
low SOC level in the CS mode.
The energy/power ratio of an energy storage is a good measure of suitability. The size of the energy storage will be minimized when its energy/power
ratio equals the required one. The energy/power ratio is defined as
Re/p =
total energy
.
power at operating SOC
(10.4)
In the example vehicle simulated above, the total energy required is about
10 kWh and the power required is about 60 kW, which is defined at 30% of
the battery SOC, yielding an Re/p of 0.167 h at 30% battery SOC. Figure 10.30
shows a typical energy/power ratio versus the specific power of energy
storage technologies.5 If 0.2 h of energy/power ratio is used in the design,
Cobasys’ NiMH battery will yield a total weight of 129 kg (60/0.465), which
carries 12 kWh of total energy (0.2 × 60). However, when SAFT’s Li-ion battery is used, the total weight will be about 56 kg (60/1.08), carrying the same
amount of energy of 12 kWh.
References
1. T. Markel, “Plug-In HEV vehicle design options and expectations,” ZEV Technology Symposium, California Air Resources Board, Sacramento, CA, September
27, 2006.
352
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
2. A. Simpson, “Cost–benefit analysis of plug-in hybrid electric vehicle technology,”
presented at the 22nd International Battery, Hybrid and Fuel Cell Electric Vehicle
Symposium and Exhibition (EVS-22), Yokohama, Japan, October 23–28, 2006.
3. J. Gonder and T. Markel, “Energy Management Strategies for Plug-In Hybrid Electric Vehicles,” SAE Paper no. 2007-01-0290, Society of Automotive Engineers,
Warrendale, PA, 2007.
4. T. Markel and K. Wipke, “Modeling grid-connected hybrid electric vehicles using
ADVISOR,” The 16th IEEE Annual Battery Conference on Application and Advances,
January 9, 2001, Long Beach, CA, 2001.
5. T. Markel and A. Simpson, “Energy storage systems considerations for gridcharged hybrid electric vehicles,” Vehicle Power and Propulsion, 2005 IEEE Conference, September 7–9, 2005.
6. T. Markel and A. Simpson, “Plug-in hybrid electric vehicle energy storage system
design,” to be presented at Advanced Automotive Battery Conference, Baltimore, MD,
May 17–19, 2006.
11
Mild Hybrid Electric Drive Train Design
Full HEVs with parallel, series, or series–parallel configurations can
significantly reduce fuel consumption by operating the engine optimally
and using effective regenerative braking.1−3 However, a high electric power
demand requires a bulky and heavy energy storage pack. This causes difficulties for packing the drive train under the hood and reduces the loading
capacity, and also increases the energy losses in the rolling tires. Full hybrid
drive trains have structures totally different from conventional drive trains. To
turn totally from conventional drive trains to full hybrids, a huge investment
of time and money is needed. A compromise is to develop an intermediate
product that is easier to convert from the current products, and yet is more
efficient than those products. One solution is to put a small electric motor
behind the engine to constitute the so-called mild or soft hybrid electric drive
train. This small electric motor can operate as an engine starter as well as an
electrical generator. It can also add additional power to the drive train when
high power is demanded and can convert part of the braking energy into electric energy. This small motor can potentially replace the clutch or the torque
converter, which is inefficient when operating with a high slip ratio.
A mild hybrid electric drive train does not require high-power energy storage due to the small power rating of the electric motor. A42-V electrical system
may be able to meet the requirements. Other subsystems of the conventional
vehicle, such as engine, transmission (gearbox), and brake, do not require
many changes.
This chapter introduces two typical configurations of mild hybrid drive
trains. Their control and parametric design are explained along with a design
example.
11.1
Energy Consumed in Braking and Transmission
As indicated in Chapter 13, a significant amount of energy is consumed in
braking, especially when driving in urban areas. Chapter 13 also indicates
that the braking power in normal driving is not large (refer to Figure 13.6).1,4
Thus, a small motor would be able to recover most of the braking energy.
353
354
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Torque ratio, output torque/input torque,
and efficiency
2.0
1.8
Torque ratio
1.6
1.4
1.2
1.0
0.8
0.6
Efficiency
0.4
0.2
0
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
Speed ratio, output speed/input speed (rpm)
FIGURE 11.1 Characteristics of a typical dynamic hydraulic torque converter.
Efficiency (%)
Vehicle speed (km/h)
Another source of energy loss in conventional vehicles is the transmission.
Conventional vehicles are usually equipped with automatic transmissions,
especially in North America. In the automatic transmission, the dynamic
hydraulic torque converter is the basic element and has low efficiency when
operating with a low-speed ratio (high-speed slip), as shown in Figure 11.1.
When the vehicle is operating with a stop-and-go driving pattern in urban
areas, the frequent accelerating of the vehicle leads to a low-speed ratio in the
torque converter, thus resulting in low operation efficiency. Figure 11.2 shows
100
80
60
40
20
0
100
80
60
40
20
0
0
200
400
600
800
Time (s)
1000
1200
1400
FIGURE 11.2 Vehicle speed and operating efficiency of an automatic transmission while
driving in an FTP75 urban drive cycle.
Mild Hybrid Electric Drive Train Design
355
the operating efficiency of a typical automatic transmission in an FTP75 urban
drive cycle. In this drive cycle, the average efficiency is about 0.5.5,6
In addition, when driving in urban areas, the engine idling time during
standstill and braking is significant. In the FTP75 drive cycle the percentage
of engine idling time reaches 44%, and in New York City it reaches about 57%.
When the engine is idling, not only does the engine itself consume energy,
but also the energy is needed to drive the transmission. For instance, about
1.7 kW of engine power is needed to drive the automatic transmission when
the vehicle is at a standstill.
Using a small electric motor to replace the torque converter and then constitute a mild hybrid electric drive train is considered to be an effective approach
to saving the energy losses in an automatic transmission, and in braking and
engine idling operation.
11.2
11.2.1
Parallel Mild Hybrid Electric Drive Train
Configuration
A parallel connected mild hybrid electric drive train is shown in Figure 11.3.
A small electric motor, which can function as engine starter, generator, and
traction motor, is placed between the engine and the automatically shifted
multigear transmission (gearbox). The clutch is used to disconnect the gearbox from the engine when required, such as during gear shifting and low
vehicle speed. The power rating of the electric motor may be in the range of
about 10% of the engine power rating. The electric motor can be smoothly
controlled to operate at any speed and torque; thus, isolation between the
electric motor and transmission is not necessary. The operation of the drive
train and each individual component is controlled by the drive train controller
and component controllers.
11.2.2
Operating Modes and Control Strategy
The drive train has several operating modes, depending on the operation of
the engine and the electric motor.
Engine-alone traction mode—In this mode, the electric motor is de-energized,
and the vehicle is propelled by the engine alone. This mode may be used
when the SOC of the batteries is in the high region, and the engine alone can
handle the power demand.
Motor-alone traction mode—In this mode, the engine is shut down and the
clutch is disengaged (open). The vehicle is propelled by the electric motor
alone. This operating mode may be used at low vehicle speed, less than
10 km/h, for example.
Battery charge mode—In this mode, the electric motor operates as a generator
and is driven by the engine, to charge the batteries.
356
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Battery
pack
Drive train
conroller
Motor
control
signal
Motor
controller
Vehicle speed
Battery SOC
Brake
pedal
Transmission gear
Engine throttle
control signal
Accelerator
pedal
Engine
Final
drive
Clutch Motor Transmission
FIGURE 11.3 Configuration of the parallel connected mild hybrid electric drive train.
Regenerative braking mode—In this mode, the engine is shut down and the
clutch is disengaged. The electric motor is operated to produce a braking
torque to the drive train. Part of the kinetic energy of the vehicle mass is
converted into electric energy and stored in the batteries.
Hybrid traction mode—In this mode, both the engine and the electric motor
deliver traction power to the drive train.
Which of the above operating modes is used in real operation depends
on the power demand, which is commanded by the driver through the
accelerator or brake pedal, the SOC of the batteries, and vehicle speed.
The control strategy is the preset control logic in the drive train controller.
The drive train controller receives the real-time signals from the driver and
each individual component (refer to Figure 11.3) and then commands the
operation of each component, according to the preset control logic. Aproposed
control logic is illustrated in Table 11.1 and Figure 11.4.5
11.2.3
Drive Train Design
The design of the mild hybrid electric drive train is very similar to the design
of the conventional drive train, because both are very close. The following is
357
Mild Hybrid Electric Drive Train Design
TABLE 11.1
Illustration of the Control Logic
Driving Condition
Control Operation
Standstill
Low speed (<10 km/h)
Braking
High-power demand (greater than the
power that the engine can produce)
Middle and low power demand
Both engine and motor are shut down
Electric motor-alone traction
Regenerative braking
Hybrid traction
Battery charge mode or engine-alone traction mode,
depending on the battery SOC (see Figure 11.4)
Battery SOC
Battery SOC top line
Battery SOC
bottom line
Battery
charge
Engine alone
traction
Battery
charge
Engine alone
traction
Time
FIGURE 11.4 Battery charge and engine-alone traction, depending on the battery SOC.
an example of the systematic design of a 1500-kg passenger car drive train.
The major parameters of the vehicle are listed in Table 11.2.
Referring to the similar conventional drive train, the engine is designed to
have a peak power of 108 kW. The engine characteristics of performance are
shown in Figure 11.5.
TABLE 11.2
Major Parameters of the Mild Hybrid Electric Drive Train
Vehicle Mass
1500 kg
Rolling resistance coefficient
Aerodynamic drag coefficient
Front area
0.01
0.28
2.25 m2
Four-gear transmission
Gear ratio:
First gear
Second gear
Third gear
2.25
1.40
1.00
Fourth gear
Final gear ratio
0.82
3.50
358
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
360
100
300
80
240
60
180
40
420
390
360
20
330
0
0
1000
2000
3000
Engine (rpm)
4000
5000
Engine torque (Nm)
120
bsfc (g/kWh)
Engine power (kW)
(a)
300
(b)
11 21 30 40 50 59 69 79 89 98108 kW
Brake specific fuel
consumption, g/kWh
(efficiency %)
Engine torque (Nm)
300
250
250 (32.7)
260 (31.5)
270 (30.3)
280 (29.2)
200
.4)
150
310 (26
.4)
350 (23
100
.5)
400 (20
6.4)
(1
0
50
600 (13.6) 0 (11.7)
70
1000 (8.2)
50
0
0
500 1000 1500 2000 2500 3000 3500 4000 4500 5000
Engine (rpm)
FIGURE 11.5 Performance of the engine: (a) performance with full throttle and (b) fuel
consumption map.
In this design, a small motor with 7-kW rating power is used, which can
operate as an engine starter and alternator and assist regenerative braking.
Figure 11.6 shows the torque and power characteristics versus speed of this
motor.
The batteries in this design example are lead–acid batteries. Lead–acid batteries are popularly used in automobiles, due to their mature technology and
low cost. They have relatively high power density, compared with other kinds
of common batteries.7 Thus they are considered to be the proper selection for
mild HEVs, in which power density is more important than energy density.
359
14
140
12
120
10
100
8
80
Power
6
60
Torque
4
40
2
0
Motor torque (Nm)
Motor power (kW)
Mild Hybrid Electric Drive Train Design
20
0
500
1000
1500
Motor (rpm)
0
2500
2000
FIGURE 11.6 Power and torque of the electric motor versus motor speed.
A cell of a lead-acid battery has the characteristics shown in Figure 11.7.
The terminal voltage varies with discharging current and time, which in turn
represent the SOC of the battery. These characteristics may be simply modeled
as shown in Figure 11.8.
In the discharging process, the battery terminal voltage can be expressed as
Vt = Vo (SOC) − [Ri (SOC) + Rc ]I,
(11.1)
where Vo (SOC) and Ri (SOC) are the open circuit voltage and internal resistance of the battery, respectively, which are functions of battery SOC, and Rc
is the conductor resistance. The discharging power at the terminals can be
expressed as
Pt = I Vo (SOC) − [Ri (SOC) + Rc ]I 2 .
(11.2)
2.05
2.0
Cell voltage (V)
1.95
1.9
1.85
1.8
1.75
20A
1.7
50A
1.65
1.6
1.55
0
70A
100A
1
2
10A
30A
3
4
5
6
7
Discharge time (h)
FIGURE 11.7 Discharge characteristics of the lead–acid battery.
8
9
10
360
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Rc
I
Ri(SOC)
Vt
Rload
V0(SOC)
FIGURE 11.8 Battery model.
The maximum power that the load can obtain at the terminals can be
expressed as
Pt max =
Vo2 (SOC)
.
4[Ri (SOC) + Rc ]
(11.3)
This maximum power is obtained when the discharging current is
I=
Vo
.
2 [Ri (SOC) + Rc ]
(11.4)
Figure 11.9a shows the terminal voltages and currents of 36 V and 12 V batteries with a current capacity of 100 Ah versus load power (discharge power).
It indicates that for the 36-V battery, the maximum power that the battery
can supply is about 8.5 kW. But for 12 V, it is less than 3 kW. Figure 11.9b
shows that the 36-V battery has a discharge efficiency of over 70% at a power
of less than 7 kW. For 12 V voltage, it is less than 2.5 kW. Thus, for the mild
hybrid electric drive train proposed in this chapter, a 42-V electric system
(36-V battery) can support the operation of the electric motor (rated power
of 7 kW).
11.2.4
Performance
Because there are few differences from the conventional drive train (engine,
transmission, etc.), the mild hybrid electric drive train is expected to have
similar acceleration and gradeability performance. Figure 11.10 shows the
performance of an example 1500-kg mild hybrid passenger car.
361
Mild Hybrid Electric Drive Train Design
20
40
35
Termial voltage
(36-V battery)
15
Terminal voltage (V)
30
25
20
10
Power
(36-V battery)
15
Terminal voltage
(12-V battery)
10
5
0
Output power (kW)
(a)
5
Power
(12-V battery)
0
100
200
400
300
500
600
Discharge current (A)
(b)
100
90
36-V battery
Efficiency (%)
80
70
60
(12-V battery)
50
40
30
0
1
2
3
4
5
6
Discharge power (kW)
7
8
9
FIGURE 11.9 Battery performance with 36 and 12 V rated voltages: (a) battery power and
terminal voltage versus discharge current and (b) battery discharge efficiency.
362
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
60
600
50
500
40
400
Distance
30
300
20
200
10
100
0
0
20
40
60
80
100 120
Vehicle speed (km/h)
140
Acceleration distance (m)
Acceleration time (s)
(a)
0
160
(b)
8
Hybrid traction (Engine + Motor)
a=25°
Engine alone traction
(46.6%)
a =20°
(36.4%) 1st gear
a=15°
(26.8%)
2nd gear a =10°
(17.6%)
3rd gear a=5° 4th gear
(8.75%)
Effort and resistance (kN)
7
6
5
4
3
2
Rolling resistance + aerodynamic drag
1
0
0
20
40
a =0°
(0%)
60 80 100 120 140 160 180 200
Vehicle speed (km/h)
FIGURE 11.10 Performance of the hybrid electric drive train: (a) acceleration and (b) tractive
effort versus vehicle speed.
Figure 11.11 shows the simulation results of a 1500-kg hybrid passenger car
in an FTP75 urban cycle. Figure 11.11b indicates that a mild hybrid electric
drive train with a small motor cannot significantly improve engine operating
efficiency, because most of the time the engine still operates in a low load
region. However, because of the elimination of engine idling and of the inefficient torque converter and utilization of regenerative braking, fuel economy
in urban driving is significantly improved. The simulation shows that for the
1500-kg passenger car mentioned above, the fuel consumption is 7.01 L per
100 km (33.2 mpg). The simulated fuel consumption for a similar conventional
vehicle is 10.7 L per 100 km (22 mpg), whereas the Toyota Camry (1445 kg curb
363
Mild Hybrid Electric Drive Train Design
(a)
100
Vehicle speed (km/h)
50
0
40
Engine power (kW)
20
0
10
Motor power (km/h)
0
–10
0.75
Battery SOC
0.7
0.65
0
200
(b)
400
600
800
Time (s)
1000
1200
1400
11 21 30 40 50 59 69 79 89 98 108 kW
+ Operating points
bsfc, g/kWh
(Efficiency %)
Engine torque (Nm)
300
250
250 (32.7)
260 (31.5)
270 (30.3)
280 (29.3)
200
310
150
)
(26.4
350
100
)
(23.4
400
50
0
(c)
0
Motor torque (Nm)
4)
500 (16. 3.6)
(1
600 70
0 (11.7)
800 (10.2)
1000 (8.2)
500 1000 1500 2000 2500 3000 3500 4000 4500 5000
Engine (rpm)
150
+ Operating points
Efficiency
100
)
(20.5
90%
88%
50
Traction
86%
83%
80%
0
90% 88%
–50
83%
80%
Regenerative braking
and generating
–100
–150
86%
70%
70%
0
500
1000
1500
Motor (rpm)
2000
2500
FIGURE 11.11 Simulation in an FTP75 urban drive cycle: (a) vehicle speed, engine power,
motor power, and battery SOC, (b) engine fuel consumption map and operating points, and
(c) motor efficiency map and operating points.
(a)
100
50
Vehicle speed (km/h)
0
40
Engine power (kW)
20
0
10
Motor power (kW)
0
–10
0.75
Battery SOC
0.7
0.65
0
100
200
300
400
500
Time (s)
600
700
800
11 21 30 40 50 59 69 79 89 98108 kW
(b)
+ Operating points
bsfc, g/kW h
(Efficiency %)
300
250 (32.7)
260 (31.5)
270 (30.3)
280 (29.2)
Engine torque (Nm)
250
200
310
150
350
100
50
0
(c)
(26.4
0
)
(23.4
)
20.5)
400 (
4)
500 (16.
600 (13.6)
700 (11.7)
1000 (8.2)
500 1000 1500 2000 2500 3000 3500 4000 4500 5000
Engine (rpm)
150
+ Operating points
100
Traction
Motor torque (Nm)
90%
88%
50
86%
83%
80%
0
80%
90% 88%
–50
83%
Regenrative braking
and generating
–100
–150
86%
70%
70%
0
500
1000
1500
Motor (rpm)
2000
2500
FIGURE 11.12 Simulation in an FTP75 highway drive cycle: (a) vehicle speed, engine power,
motor power, and battery SOC, (b) engine fuel consumption map and operating points, and
(c) motor efficiency map and operating points.
365
Mild Hybrid Electric Drive Train Design
weight, four-cyclinder, 2.4 L, 157 hp, or 117 kW maximum engine power,
automatic transmission) has a fuel economy of about 10.3 L/100 km (23 mpg).8
With mild hybrid technology, fuel consumption can be reduced by more than
30%. Figure 11.11c shows the motor efficiency map and operating points. They
indicate that the electric motor operates as a generator more than a traction
motor, to support the electric load of auxiliaries and maintain the battery SOC
balanced.
Figure 11.12 shows the simulation results of the same vehicle on an FTP75
highway drive cycle. Compared to urban driving, the speeds of both engine
and motor are higher, due to higher vehicle speed. The fuel consumption
is 7.63 L per 100 km (31 mpg) (Toyota Camry: 7.38 L/100 km or 32 mpg).8
The fuel economy has not improved compared to conventional vehicles. The
reason is that the highway vehicle has less energy losses in engine idling,
braking, and transmission than during urban driving, and thus not much
room exists for fuel economy improvement using mild hybrid technology.
11.3
11.3.1
Series–Parallel Mild Hybrid Electric Drive Train
Configuration of the Drive Train with a Planetary Gear Unit
Figure 11.13 shows the configuration of a series–parallel (speed coupling and
torque coupling) mild hybrid electric drive train, which uses a planetary gear
Planetary gear unit
Battery
Ring gear
Lock 2
Clutch 2
Clutch 1
Wheel
Motor
controller
Engine
Motor rotor
Motor stator
Lock 1
Sun gear
Yoke
Planetary gears
Transmission
(Gear box)
Gear
Wheel
FIGURE 11.13 Series–parallel mild hybrid electric drive train with a planetary gear unit.
366
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
unit to connect the engine, motor, and transmission (gearbox) together. This
configuration is very similar to that shown in Figure 9.6 in Chapter 9, but
by removing the traction motor and placing a multigear transmission. The
engine is connected to the yoke of the planetary gear unit through clutch 1,
which is used to couple or decouple the engine from the yoke. Lock 2 is used
to lock the yoke of the planetary gear unit to the vehicle frame. The electric
motor is connected to the sun gear. Clutch 2 is used to couple or decouple the
sun gear (electric motor) to or from the yoke. Lock 1 is used to lock the sun
gear and the rotor of the electric motor to the vehicle frame. The transmission
(gearbox) is driven by the ring gear of the planetary gear unit through a gear.
The operating characteristics of the planetary gear unit have been discussed
in detail in Chapter 9. They are repeated here for the reader’s convenience.
The speeds, in rpm, of the sun gear, ns , ring gear, nr , and the yoke, ny , have
the relationship
ny =
ig
1
ns +
nr ,
1 + ig
1 + ig
(11.5)
where ig is the gear ratio defined as Rr /Rs as shown in Figure 11.14. The speeds
ns , nr , and ny are defined as positive in the direction shown in Figure 11.14.
Defining kys = (1 + ig ) and kyr = (1 + ig )/ig , Equation 11.5 can be further
expressed as
ny =
1
1
ns +
nr .
kys
k yr
(11.6)
Ignoring the energy losses in the steady-state operation, the torques acting on
the sun gear, ring gear, and yoke have the relationship
Ty = −kys Ts = −kyr Tr .
r
(11.7)
nr , Tr
y
Rs
Ry
Rr
ns , Ts
s
ny
p
FIGURE 11.14 Planetary gear unit used as a speed coupling.
Ty
367
Mild Hybrid Electric Drive Train Design
Element fixed
Speed
Torque
Sun gear
ny =
1
nr
kyr
Ty = –kyrTr
Ring gear
ny =
1
ns
kys
Ty = –kysTs
Yoke
ns =–
kys
kys
nr
Ts =
kyr
kys
Tr
FIGURE 11.15 Speed and torque relationships while one element is fixed.
Equation 11.7 indicates that the torques acting on the sun gear, Ts , and ring
gear, Tr , always have the same sign. In other words, they have to always be
in the same direction. However, the torque acting on the yoke, Ty , is always
in the opposite direction of Ts and Tr . Equation 11.7 also indicates that with
ig > 1, which is the general case since Rr > Rs , Ts is the smallest, Ty is the
largest, and Tr is in between. This means that the torque acting on the yoke
is balanced by torques acting on the sun gear and ring gear.
When one element among the sun gear, ring gear, and yoke is locked to
the vehicle stationary frame, that is, one degree of freedom of the unit is
constrained, the unit becomes a single-gear transmission (one input and one
output). The speed and torque relationships, with different elements fixed,
are shown in Figure 11.15.
11.3.2
Operating Modes and Control
As suggested by the configurations of the drive trains, there are two distinct basic operating modes: speed coupling and torque coupling between
the engine and gearbox, depending on the engagement or disengagement
states of the clutches and the lock.
11.3.2.1
Speed-Coupling Operating Mode
When the vehicle is starting from zero speed, and because the engine cannot
run at zero speed and transmission has only a finite gear ratio, slip must
exist between the input shaft and output shaft of the transmission. The slip
usually occurs in a clutch in manual transmissions or in a hydrodynamic
torque converter in an automatic transmission. Thus, a certain amount of
energy is lost in this slip. However, in the case of the drive train shown in
Figure 11.13, this slip is performed between the engine and the electric motor
(yoke and sun gear). In this case, clutch 1 connects the engine shaft to the
yoke, clutch 2 releases the sun gear from the yoke, and locks 1 and 2 release
the sun gear (motor) and yoke (engine) from the vehicle frame at a given
368
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
engine (yoke) speed and ring gear speed, which is proportional to the vehicle
speed, according to Equation 11.6, the motor speed is
ns = kys
nr
ny −
kyr
.
(11.8)
When the first term on the right-hand side of Equation 11.8 is larger than
the second term, that is, at low vehicle speed, the motor velocity is positive.
However, from Equation 11.7, it is known that the motor torque must be
negative, as expressed by
Ts = −
Ty
.
kys
(11.9)
Thus, the motor power is negative and it operates as a generator. The
generating power can be expressed as
Pm/g =
2π
2π
T s ns =
(−ny Ty + nt Tr ) = −Pe + Pt ,
60
60
(11.10)
where Pm/g is the generating power of the motor (negative), Pe is the engine
power, and Pt is the power going to the transmission for propelling the vehicle.
The vehicle speed (proportional to nr ) at which the value that makes ns
equal to zero is defined as synchronous speed. With further increase of vehicle
speed, ns becomes negative and the electric motor goes into the motoring state.
In the speed-coupling operating mode, the engine speed is decoupled from
the vehicle speed and can be controlled by the motor torque and engine
throttle as discussed before.
11.3.2.2 Torque-Coupling Operating Mode
When clutch 1 is engaged, lock 2 releases the yoke and clutch 2 engages
the sun gear (motor) and yoke (engine). The engine and motor speeds are
forced to be the same. The ring gear speed and yoke (engine) speed have the
relationship (from Equation 11.6)
nr = kyr
1
1−
kys
ny .
(11.11)
From the definitions of kyr and kys by Equations 11.4 and 11.5, Equation 11.11
can be rewritten as
nr = ny .
(11.12)
Equation 11.12 implies that the gear ratio from the engine and motor to the
ring gear is 1.
369
Mild Hybrid Electric Drive Train Design
The engine torque and motor torque are added together by the planetary
gear unit and then delivered to the transmission from the ring gear, which
can be expressed as
Tr = Te + Tm/g ,
(11.13)
where Te and Tm are the engine torque and motor torque, respectively.
11.3.2.3
Engine-Alone Traction Mode
The engine-alone traction mode can be realized in two ways. One is with
the same operation of clutch 1, clutch 2, and lock 2 as in the torque-coupling
mode (the previous section). But here the motor is de-energized. In this case,
the torque delivered to the transmission is expressed as
Tr = Te .
(11.14)
The other is to lock the sun gear (motor shaft) to the vehicle frame by lock 1
and both clutch 2 and lock 2 are released. From Equation 11.6, the ring gear
speed and the yoke (engine) speed have the relationship
nr = kyr ny .
(11.15)
The torque delivered to the transmission can be expressed as
Tr =
Te
.
kyr
(11.16)
It is seen from the above discussion that the planetary gear unit plays the role
of a two-gear transmission. The lower gear ratio is 1/kyr < 1 and the higher
gear ratio is 1.
11.3.2.4
Motor-Alone Traction Mode
In this mode, the engine is shut down and clutch 1 disengages the engine from
the yoke. The motor alone is used to propel the vehicle. There are two ways
of doing this. One method is by coupling the sun gear to the yoke by clutch 2.
In this way, the motor delivers its speed and torque to the ring gear as
nm = nr
(11.17)
Tm = Tr .
(11.18)
and
The gear ratio from the motor to the ring gear is 1.
370
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
The other method is by releasing the sun gear from the yoke by clutch 2 and
locking the yoke to the vehicle frame by lock 2. In this manner, the speed and
torque of the motor are related to the speed and torque of the ring gear by
kys
nr
kyr
(11.19)
kyr
Tr .
kys
(11.20)
nm/g = −
and
Tm/g =
Equation 11.19 indicates that the motor turns in a direction opposite to that
of the ring gear.
11.3.2.5
Regenerative Braking Mode
During braking, part of the braking energy can be recovered by the
motor/generator. The operation of the drive train is the same as in electric
traction mode, but the motor produces its torque in the direction opposite to
that for traction.
11.3.2.6
Engine Starting
The engine can be started by the electric motor by engaging the sun gear to
the yoke by clutch 2. The motor directly delivers its torque to the engine to
start it.
11.3.3
Control Strategy
When the vehicle speed is lower than the synchronous speed, the speedcoupling operation mode is used. As explained in Section 11.3.2.1, the electric
motor operates with a positive speed and negative power. One part of the
engine power is used to charge the batteries and the other part is used to
propel the vehicle.
When the vehicle speed is higher than the synchronous speed, the torquecoupling operation mode is used, and the drive train control strategy in this
mode is as follows.
1. When the traction power demand is greater than the power that the
engine can develop with full throttle, a hybrid traction mode is used.
In this case, the engine is operated with full throttle and the electric
motor supplies extra power to meet the traction power demand.
2. When the traction power demand is less than the power that the engine
can develop with full throttle, the operations of the engine and electric motor are determined by the SOC of the batteries, as shown in
371
Mild Hybrid Electric Drive Train Design
Battery SOC
Battery SOC top line
Battery SOC
bottom line
Battery
charge
Engine alone
traction
Battery
charge
Engine alone
traction
Time
FIGURE 11.16 Battery charge and engine-alone traction, depending on battery SOC.
Figure 11.16. In the battery charging mode the battery charging power
may be determined by the maximum power of the electric motor or
by the maximum engine power and demanded traction power.
11.3.4
Drive Train with a Floating-Stator Motor
An alternative mild hybrid electric drive train, which has characteristics similar to the drive train discussed above, is shown in Figure 11.17.6 This drive
train uses an electric motor, which has a floating stator, to replace the planetary
gear unit and the electric motor.
As mentioned in Chapter 5, the angular velocity of the rotor is the summation of the angular velocities of the stator and the relative angular velocity
Batteries
Motor
controller
Engine
Clutch 1
Clutch 2
Transmission
(Gear box)
Motor rotor
Motor stator
Lock
FIGURE 11.17 Series–parallel mild hybrid electric drive train with a floating-stator motor.
372
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
ws’ Ts
wr’ Tr
FIGURE 11.18 An electric motor with a floating stator.
between the stator and rotor, that is,
ωr = ωs + ωrr .
(11.21)
Due to the action and reaction effect, the torque acting on the stator and rotor
is always equal to the electromagnetic torque produced in the air gap (refer
to Figure 11.18), which is, in a general sense, the electric motor torque. This
relationship is described as
Tr = Ts = Tm ,
(11.22)
where Tm is the electromagnetic torque in the air gap.
Comparing Equations 11.21 and 11.22 with Equations 11.6 and 11.7, it is
known that both the planetary gear unit and the floating stator motor have
the same operating characteristics. Therefore the mild hybrid electric drive
trains as shown in Figures 11.13 and 11.17 have the same operating principles
and use the same control strategy. However, the design of the drive train with
a planetary gear unit is more flexible since the gear ratio, ig , is selectable.
References
1. Y. Gao, L. Chen, and M. Ehsani, “Investigation of the effectiveness of regenerative
braking for EV and HEV,” Society of Automotive Engineers (SAE) Journal, SP-1466,
Paper No. 1999-01-2901, 1999.
2. Y. Gao, K. M. Rahman, and M. Ehsani, “The energy flow management and battery
energy capacity determination for the drive train of electrically peaking hybrid,”
Society of Automotive Engineers (SAE) Journal, SP-1284, Paper No. 972647, 1997.
3. Y. Gao, K. M. Rahman, and M. Ehsani, “Parametric design of the drive train of
an electrically peaking hybrid (ELPH) vehicle,” Society of Automotive Engineers
(SAE) Journal, SP-1243, Paper No. 970294, 1997.
Mild Hybrid Electric Drive Train Design
373
4. H. Gao, Y. Gao, and M. Ehsani “Design issues of the switched reluctance motor
drive for propulsion and regenerative braking in EV and HEV,” Society of Automotive Engineers (SAE) Future Transportation Technology Conference, Costa Mesa,
CA, Paper No. 2001-01-2526, August 2001.
5. Y. Gao and M. Ehsani, “A mild hybrid drive train for 42 V automotive power
system—design, control, and simulation,” Society of Automotive Engineers (SAE)
World Congress, Detroit, MI, Paper No. 2002-02-1082, 2002.
6. Y. Gao and M. Ehsani, “A mild hybrid vehicle drive train with a floating stator
motor—configuration, control strategy, design, and simulation verification,” Society of Automotive Engineers (SAE) Future Car Congress, Crystal City, VA, Paper No.
2002-01-1878, June 2002.
7. Y. Gao and M. Ehsani, “Investigation of battery technologies for the Army’s hybrid
vehicle application,” Proceedings of the IEEE 56th Vehicular Technology Conference,
Vancouver, British Columbia, Canada, September 2002.
8. Y. Gao and M. Ehsani, “Electronic braking system of EV and HEV—integration of
regenerative braking, automatic braking force control and ABS,” Society of Automotive Engineers (SAE) Future Transportation Technology Conference, Costa Mesa,
CA, Paper No. 2001-01-2478, August 2001.
12
Peaking Power Sources and Energy Storages
“Energy storages” are defined in this book as devices that store energy, deliver
energy outside (discharge), and accept energy from outside (charge). There
are several types of energy storages that have been proposed for EV and
HEV applications. These energy storages, so far, mainly include chemical
batteries, ultracapacitors or supercapacitors, and ultra-high-speed flywheels.
The fuel cell, which essentially is a type of energy converter, will be discussed
in Chapter 14.
There are a number of requirements for energy storages applied in an
automotive application, such as specific energy, specific power, efficiency,
maintenance requirement, management, cost, environmental adaptation and
friendliness, and safety. For application on an EV, specific energy is the first
consideration since it limits the vehicle range. On the other hand, for HEV
applications, specific energy becomes less important and specific power is the
first consideration, because all the energy is from the energy source (engine
or fuel cell) and sufficient power is needed to ensure vehicle performance,
particularly during acceleration, hill climbing, and regenerative braking. Of
course, other requirements should be fully considered in the vehicle drive
train development.
12.1
Electrochemical Batteries
Electrochemical batteries, more commonly referred to as “batteries,” are electrochemical devices that convert electrical energy into potential chemical
energy during charging, and convert chemical energy into electric energy
during discharging. A battery is composed of several cells stacked together.
A cell is an independent and complete unit that possesses all the electrochemical properties. Basically, a battery cell consists of three primary
elements: two electrodes (positive and negative), immersed into electrolyte
as shown in Figure 12.1.
Battery manufacturers usually specify the battery with coulometric capacity (ampere-hours), which is defined as the number of ampere-hours gained
when discharging the battery from a fully charged state until the terminal
375
376
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
e–
Discharge
Charge
–
+
P
Ion migration
Electrolyte
Positive electrode
Negative electrode
N
FIGURE 12.1 A typical electrochemical battery cell.
voltage drops to its cut-off voltage, as shown in Figure 12.2. It should be
noted that the same battery usually has a different number of ampere-hours
at different discharging current rates. Generally, the capacity will become
smaller with a large discharge current rate, as shown in Figure 12.3. Battery
manufacturers usually specify a battery with a number of ampere-hours along
with a current rate. For example, a battery labeled as 100 Ah at C/5 rate has a
100 Ah capacity at a 5-h discharge rate (discharging current = 100/5 = 20 A).
Another important parameter of a battery is the SOC. SOC is defined as
the ratio of remaining capacity to fully charged capacity. With this definition,
a fully charged battery has an SOC of 100% and a fully discharged battery
Cell voltage
Open circuit voltage
Cut-off voltage
Discharging time
FIGURE 12.2 Cut-off voltage of a typical battery.
377
Peaking Power Sources and Energy Storages
2.05
2.00
Cell voltage (V)
1.95
10
1.90
20
1.85
30
1.80
50
1.75
70
1.70
Amperes
1.65
100
1.60
1.55
0
10
20
30
40 50 60
Capacity (Ah)
70
80
90
100
FIGURE 12.3 Discharge characteristics of a lead–acid battery.
has an SOC of 0%. However, the term “fully discharged” sometimes causes
confusion, because of the different capacities at different discharge rates, and
different cut-off voltages (refer to Figure 12.3). The change in SOC in a time
interval, dt, with discharging or charging current i may be expressed as
ΔSOC =
idt
,
Q(i)
(12.1)
where Q(i) is the ampere-hour capacity of the battery at current rate i.
For discharging, i is positive, and for charging, i is negative. Thus, the SOC
of the battery can be expressed as
SOC = SOC0 −
idt
,
Q(i)
(12.2)
where SOC0 is the initial value of the SOC.
For EVs and HEVs, the energy capacity is considered to be more important
than the coulometric capacity (ampere-hours), because it is directly associated with vehicle operation. The energy delivered from the battery can be
expressed as
EC =
t
V(i, SOC)i(t)dt,
(12.3)
0
where V(i, SOC) is the voltage at the battery terminals, which is a function of
battery current and SOC.
378
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
12.1.1
Electrochemical Reactions
For simplicity, and because it is the most widespread battery technology in
today’s automotive applications, the lead–acid battery case is used as an
example to explain the operating principle theory of electrochemical batteries.
A lead–acid battery uses an aqueous solution of sulfuric acid (2H+ + SO2−
4 )
as electrolyte. The electrodes are made of porous lead (Pb, anode, electrically negative) and porous lead oxide (PbO2 , cathode, electrically positive).
The processes taking place during discharging are shown in Figure 12.4a,
where the lead is consumed and lead sulfate is formed. The chemical reaction
on the anode can be written as
−
Pb + SO2−
4 → PbSO4 + 2e .
(12.4)
This reaction releases two electrons and, thereby, gives rise to an excess negative charge on the electrode that is relieved by a flow of electrons through
the external circuit to the positive (cathode) electrode. At the positive electrode the PbO2 is also converted to PbSO4 and, at the same time, water is
formed. The reaction can be expressed as
−
PbO2 + 4H+ + SO2−
4 + 2e → PbSO4 + 2H2 O.
(12.5)
During charging, the reactions on the anode and cathode are reversed as
shown in Figure 12.4a and can be expressed as follows:
Anode:
PbSO4 + 2e− → Pb + SO2−
4
2H2O
PbO2
2e–
Pb
4H+
SO4–2
2e–
SO4–2
4H+
SO4–2
2H2O
Positive electrode, PbSO4
SO4–2
(b)
Negative electrode, PbSO4
PhSO4
PhSO4
2e–
Positive electrode, PbO2
2e–
Negative electrode, Pb
(a)
(12.6)
FIGURE 12.4 Electrochemical processes of a lead–acid battery cell: (a) discharging and (b)
charging.
379
Peaking Power Sources and Energy Storages
and
Cathode:
−
PbSO4 + 2H2 O → PbO2 + 4H+ + SO2−
4 + 2e .
(12.7)
The overall reaction in a lead–acid battery cell can be expressed as follows:
Overall:
Pb + PbO2 + 2H2 SO4
Discharge
Charge
2PbSO4 + 2H2 O.
(12.8)
The lead–acid battery has a cell voltage of about 2.03 V at standard
conditions, which is affected by the concentration of the electrolyte.
12.1.2 Thermodynamic Voltage
The thermodynamic voltage of a battery cell is closely associated with the
energy released and the number of electrons transferred in the reaction. The
energy released by a battery cell reaction is given by the change in Gibbs
free energy, ΔG, usually expressed in per mole quantities. The change in
Gibbs free energy in a chemical reaction can be expressed as
ΔG =
Gi −
Gj ,
(12.9)
Products
Reactants
where Gi and Gj are the free energy in species i of products and species j of
reactants. In a reversible process, ΔG is completely converted into electric
energy, that is,
ΔG = −nFVr ,
(12.10)
where n is the number of electrons transferred in the reaction, F = 96,495 is
the Faraday constant in coulombs per mole, and Vr is the reversible voltage
of the cell. At standard conditions (25◦ C temperature and 1 atm pressure), the
open circuit (reversible) voltage of a battery cell can be expressed as
Vr0 = −
ΔG0
,
nF
(12.11)
where ΔG0 is the change in Gibbs free energy at standard conditions.
The change of free energy, and thus the cell voltage, in a chemical reaction
is a function of the activities of the solution species. From Equation 12.10
and the dependence of ΔG on reactant activities, the Nernst relationship is
derived as
RT
(activities of products)
0
Vr = Vr −
,
(12.12)
ln (activities of reactants)
nF
where R is the universal gas constant, 8.31 J/mol K, and T is the absolute
temperature in K.
380
12.1.3
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Specific Energy
Specific energy is defined as the energy capacity per unit battery weight
(Wh/kg). The theoretical specific energy is the maximum energy that can be
generated per unit total mass of cell reactants. As discussed above, the energy
in a battery cell can be expressed by the Gibbs free energy ΔG. With respect to
the theoretical specific energy, only the effective weights (molecular weights
of reactants and products) are involved; then
Espe,theo = −
ΔG
nFVr
=
(Wh/kg),
3.6 Mi
3.6 Mi
(12.13)
where
Mi is the sum of the molecular weight of the individual species
involved in the battery reaction.
Taking the lead–acid battery as an exam
ple, Vr = 2.03 V, n = 2, and Mi = 642 g, then Espe,theo = 170 Wh/kg. From
Equation 12.13, it is clear that the “ideal” couple would be derived from a
highly electronegative element and a highly electropositive element, both of
low atomic weight. Hydrogen, lithium, or sodium would be the best choice
for the negative reactants, and the lighter halogens, oxygen, or sulfur would
be the best choice for the positive reactants. To put such couples together in
a battery requires electrode designs for effective utilization of the contained
active materials, as well as electrolytes of high conductivity compatible with
the materials in both electrodes. These constraints result in oxygen and sulfur
being used in some systems as oxides and sulfides, rather than as the elements
themselves. For operation at ambient temperature, aqueous electrolytes are
advantageous because of their high conductivities. Here, alkali-group metals
cannot be employed as electrodes since these elements react with water. It
is necessary to choose other metals, which have a reasonable degree of electropositivity, such as zinc, iron, or aluminum. When considering electrode
couples, it is preferable to exclude those elements that have a low abundance
in the earth’s crust, are expensive to produce, or are unacceptable from a
health or environmental point of view.1
Examination of possible electrode couples has resulted in the study of
more than 30 different battery systems with a view to developing a reliable,
high-performance, inexpensive high-power energy source for electric traction. The theoretical specific energies of the systems championed for EVs and
HEVs are presented in Table 12.1.1 Practical specific energies, however, are
well below the theoretical maxima. Apart from electrode kinetic and other
restrictions that serve to reduce the cell voltage and prevent full utilization
of the reactants, there is a need for construction materials, which add to the
battery weight but are not involved in the energy producing reaction.
In order to appreciate the extent to which the practical value of the specific energy is likely to differ from the theoretical values, it is instructive to
consider the situation of the well-established lead–acid battery. A breakdown
of the various components of a lead–acid battery designed to give a practical
specific energy of 45 Wh/kg is shown in Figure 12.5.1 It shows that only about
381
Peaking Power Sources and Energy Storages
TABLE 12.1
Theoretical Specific Energies of Candidate Batteries for EVs and HEVs1
Cell Reaction
Battery
(+)
(−)
Charge
⇐
Discharge
⇒
Specific Energy
(Wh/kg)
Acidic Aqueous Solution
PbO2
Pb
PbO2 + 2H2 SO4 + Pb
⇔ 2PbSO4 + 2H2 O
170
Alkaline Aqueous Solution
NiOOH
Cd
2NiOOH + 2H2 O + Cd
⇔ 2Ni(OH)2 + Cd(OH)2
217
NiOOH
Fe
2NiOOH + 2H2 O + Fe
⇔ 2Ni(OH)2 + Fe(OH)2
267
NiOOH
Zn
2NiOOH + 2H2 O + Zn
⇔ 2Ni(OH)2 + Zn(OH)2
341
NiOOH
H2
Zn
2NiOOH + H2
2MnO2 + H2 O + Zn
⇔ 2Ni(OH)2
⇔ 2MnOOH + ZnO
387
317
Al
4Al + 6H2 O + 3O2
⇔ 4Al(OH)3
2815
O2
Fe
Zn
2Fe + 2H2 O + O2
2Zn + 2H2 O + O2
⇔ 2Fe(OH)2
⇔ 2Zn(OH)2
764
888
Flow
Br2
Zn
Zn + Br2
⇔ZnBr2
436
Zn + Cl2
⇔ZnCl2
833
(VO2 )2SO4 + 2HVSO4
+ 2H2 SO4
2VOSO4 + V2 (SO4 )3 + 2H2 O
114
MnO2
O2
O2
Cl2
Zn
(VO2 )2SO4 VSO4
Molten Salt
S
Na
NiCl2
FeS2
Na
LiAl
2N3S a +
⇔Na2 S3
760
2Na + NiCl2
4LiAl + FeS2
⇔ 2NaCl
⇔ 2Li2 S + 4Al + Fe
790
650
⇔ LiyC6 + Li(1 − y)CoO2
320a
Organic Lithium
LiCoO2
Li − C Li(y + x)C6 +
Li(1 − (y − x))CoO2
a
For maximum values of x = 0.5 and y = 0.
26% of the total weight of the battery is directly involved in producing electrical energy. The remainder is made up of (1) potential cell reactants that are
not discharged at the rates required for EV operation, (2) water used as the
solvent for the electrolyte (sulfuric acid alone is not suitable), (3) lead grids
for current collection, (4) “top lead,” that is, terminals, straps, and intercell
connectors, and (5) cover, connector, and separators.
A similar ratio of practical-to-theoretical specific energy is expected for
each of the candidate systems listed in Table 12.1. Present values realized by
experimental cells and prototype batteries are listed in Table 12.2.1 In recent
years, some high-power batteries have been developed for application of
HEVs.2
382
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Carver
container
separators
Pb + PbO2 +
H2SO4
(reacted)
Top lead
Pb + PbO2 + H2SO4
(unreacted)
Current-collectors
(grids)
Water in
electrolyte
FIGURE 12.5 Weight distribution of the components of a lead–acid EV battery with a specific
energy of 45 Wh/kg at the C5/5 rate.1
12.1.4
Specific Power
Specific power is defined as the maximum power of per unit battery weight
that the battery can produce in a short period. Specific power is important
in the reduction of battery weight, especially in high-power demand applications, such as HEVs. The specific power of a chemical battery depends
mostly on the battery’s internal resistance. With the battery model as shown
in Figure 12.6, the maximum power that the battery can supply to the load is
Ppeak =
V02
,
4 (Rc + Rint )
(12.14)
where Rc is the conductor resistance (ohmic resistance) and Rint is the
internal resistance caused by chemical reaction.
Internal resistance, Rint , represents the voltage drop, ΔV, which is associated with battery current. The voltage drop ΔV, termed overpotential in
battery terminology, includes two components: one caused by reaction activity ΔVA and the other by electrolyte concentration ΔVC . General expressions
of ΔVA and ΔVC are3
ΔVA = a + b log I
and
ΔVC = −
I
RT
,
ln 1 −
nF
IL
(12.15)
(12.16)
70–85
20–30
150–240
90–120
100–130
Flow
Zinc/bromine
Vanadium redox
Molten Salt
Sodium/sulfur
Sodium/nickel chloride
Lithium/iron sulfide (FeS)
a
No self-discharge, but some energy loss by cooling.
80–130
50–60
50–60
55–75
70–95
200–300
80–120
100–220
Alkaline Aqueous Solution
Nickel/cadmium
Nickel/iron
Nickel/zinc
Ni–MH
Aluminum/air
Iron/air
Zinc/air
Organic/Lithium
Li–I
35–50
Specific Energy
(Wh/kg)
Acidic Aqueous Solution
Lead/acid
System
200–300
150–250
230
130–160
90–110
110
80–150
80–150
170–260
200–300
160
90
30–80
150–400
Peak Power
(V/kg)
Status of Battery System for Automotive Application
TABLE 12.2
>95
80
80
80
65–70
75–85
75
75
65
70
<50
60
60
>80
Energy
Efficiency (%)
1000+
1000+
800+
1200+
500–2000
–
800
1500–2000
300
750–1200+
?
500+
600+
500–1000
Cycle Life
0.7
?
0a
0a
?
–
1
3
1.6
6
?
?
?
0.6
Self-Discharge
(% per 48 h)
200
230–345
110
250–450
200–250
400–450
250–350
200–400
100–300
200–350
?
50
90–120
120–150
Cost
(US$/kWh)
Peaking Power Sources and Energy Storages
383
384
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Rc
Rint
+
Vt
V0
Rload
–
FIGURE 12.6 Battery circuit model.
where a and b are constants, R is the gas constant, 8.314 J/K mol, T is the
absolute temperature, n is the number of electrons transferred in the reaction,
F is the Faraday constant, 96,495 ampere-seconds per mole, and IL is the
limit current. Accurate determination of battery resistance or voltage drop
by analysis is difficult and is usually obtained by measurement.1 The voltage
drop increases with increase in discharging current, decreasing the stored
energy in it (refer to Figure 12.3).
Table 12.2 also shows the status of battery systems potentially available
for EV. It can be seen that although specific energies are high in advanced
batteries, specific powers have to improve. About 300 W/kg might be the
optimistic estimate. However, SAFT has reported their Li-ion high power
for HEV application with a specific energy of 85 Wh/kg and a specific power
of 1350 W/kg and their high-energy batteries for EV application with about
150 Wh/kg and 420 W/kg (at 80% state-of-discharge, 150 A current, and 30 s),
respectively.2
12.1.5
Energy Efficiency
The energy or power losses during battery discharging and charging appear
in the form of voltage loss. Thus the efficiency of the battery during discharging and charging can be defined at any operating point as the ratio of the cell
operating voltage to the thermodynamic voltage, that is,
During discharging
V
η=
(12.17)
V0
and
During charging
η=
V0
.
V
(12.18)
The terminal voltage, as a function of battery current and energy stored in
it or SOC, is lower in discharging and higher in charging than the electrical
385
Peaking Power Sources and Energy Storages
95
Discharge
Efficiency, %
90
Charge
85
Net cycle
80
75
0.2
0.3
0.4
0.5
0.6
State-of-charge (SOC)
0.7
0.8
FIGURE 12.7 Typical battery charge and discharge efficiency.
potential produced by chemical reaction. Figure 12.7 shows the efficiency of
the lead–acid battery during discharging and charging. The battery has a high
discharging efficiency with high SOC and a high charging efficiency with low
SOC. The net cycle efficiency has a maximum in the middle range of the SOC.
Therefore, the battery operation control unit of an HEV should control the
battery SOC in its middle range so as to enhance the operating efficiency and
depress the temperature rise caused by energy loss. High temperature would
damage the battery.
12.1.6
Battery Technologies
The viable EV and HEV batteries consist of the lead–acid battery, nickel-based
batteries, such as nickel/iron, nickel/cadmium, and nickel–metal hydride
(Ni–MH) batteries, and lithium-based batteries such as lithium–polymer
(Li–P) and lithium–ion (Li–I) batteries.4 It seems that cadmium-based and
lithium-based batteries would be the major candidates for EVs and HEVs.
12.1.6.1
Lead–Acid Battery
The lead–acid battery has been a successful commercial product for over a
century and is still widely used as the electrical energy storage in the automotive field and other applications. Its advantages are its low cost, mature
technology, and relatively high-power capability. These advantages are attractive for application in HEVs, where high power is the first consideration. The
materials involved (lead, lead oxide, and sulfuric acid) are rather low cost
when compared with their more advanced counterparts. Lead–acid batteries
also have several disadvantages. The energy density is low, mostly because
386
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
of the high molecular weight of lead. The temperature characteristics are
poor.3 Below 10◦ C, specific power and specific energy are greatly reduced.
This aspect severely limits the application of lead–acid batteries for the
traction of vehicles operated in cold climates.
The presence of highly corrosive sulfuric acid is a potential safety hazard
for vehicle occupants. The hydrogen released by the self-discharge reactions
is another potential danger, since this gas is extremely flammable even in
tiny concentrations. The hydrogen emission is also a problem for hermetically sealed batteries. Indeed, in order to provide a good level of protection
against acid spills, it is necessary to seal the battery, thus trapping the parasitic
gases in the casing. As a result, pressure may build up in the battery, causing
swelling and mechanical constraints on the casing and sealing. The lead in the
electrodes is an environmental problem because of its toxicity. The emission
of lead consecutive to the use of lead–acid batteries may occur during the
fabrication of the batteries, in the case of vehicle wreck (spill of electrolyte
through cracks), or during their disposal at the end of battery life.
Different lead–acid batteries with improved performance are being developed for EVs and HEVs. Improvements of the sealed lead–acid batteries
in specific energy over 40 Wh/kg, with the possibility of rapid charge,
have been attained. One of these advanced sealed lead–acid batteries is
the Electrasource’s Horizon battery. It adopts the lead-wire-woven horizontal plate and hence offers competitive advantages of high specific energy
(43 Wh/kg), high specific power (285 W/kg), long cycle life (over 600 cycles
for on-road EV application), rapid recharge capability (50% capacity in 8 min
and 100% in less than 30 min), low cost (US$2000–3000 per EV), mechanical ruggedness (robust structure of the horizontal plate), maintenancefree character (sealed battery technology), and environmental friendliness.
Other advanced lead–acid battery technologies include bipolar designs and
micro-tubular grid designs.
Advanced lead–acid batteries have been developed to remedy their disadvantages. The specific energy has been increased through the reduction of
inactive materials such as the casing, current collector, separators, and so on.
The lifetime has been increased by over 50%—at the expense of cost, however.
The safety issue has been improved, with electrochemical processes designed
to absorb the parasitic releases of hydrogen and oxygen.
12.1.6.2
Nickel-Based Batteries
Nickel is a lighter metal than lead and has very good electrochemical properties desirable for battery applications. There are four different nickel-based
battery technologies: nickel–iron, nickel–zinc, nickel–cadmium, and Ni–MH.
12.1.6.2.1 Nickel/Iron Battery
The nickel/iron system was commercialized during the early years of the
20th century. Applications included fork-lift trucks, mine locomotives, shuttle
Peaking Power Sources and Energy Storages
387
vehicles, railway locomotives, and motorized hand trucks.1 The system comprises a nickel (III) hydroxy-oxide (NIOOH) positive electrode and a metallic
iron negative electrode. The electrolyte is a concentrated solution of potassium hydroxide (typically 240 g/L) containing lithium hydroxide (50 g/L).
The cell reaction is given in Table 12.1 and its nominal open-circuit voltage
is 1.37 V.
Nickel/iron batteries suffer from gassing, corrosion, and self-discharge
problems. These problems have been partially or totally solved in prototypes
that have yet to reach the market. These batteries are complex due to the
need to maintain the water level and the safe disposal of the hydrogen and
oxygen released during the discharge process. Nickel–iron batteries also suffer from low temperatures, although less than lead–acid batteries. Finally, the
cost of nickel is significantly higher than that of lead. Their greatest advantages are high power density compared with lead–acid batteries, capable of
withstanding 2000 deep discharges.
12.1.6.2.2 Nickel/Cadmium Battery
The nickel/cadmium system uses the same positive electrodes and electrolyte
as the nickel/iron system, in combination with metallic cadmium negative
electrode. The cell reaction is given in Table 12.1 and its nominal open-circuit
voltage is 1.3 V. Historically, the development of the battery has coincided
with that of nickel/iron and they have similar performance.
Nickel/cadmium technology has gained enormous technical improvement
because of the advantages of high specific power (over 220 W/kg), long cycle
life (up to 2000 cycles), high tolerance of electric and mechanical abuse, a small
voltage drop over a wide range of discharge currents, rapid charge capability
(about 40–80% in 18 min), wide operating temperature (−40◦ C to −85◦ C),
low self-discharge rate (<0.5% per day), excellent long-term storage due to
negligible corrosion, and availability in a variety of size designs. However,
the nickel/cadmium battery has some disadvantages, including high initial
cost, relatively low cell voltage, and the carcinogenicity and environmental
hazard of cadmium.
The nickel/cadmium battery can be generally divided into two major categories, namely the vented and sealed types. The vented type consists of many
alternatives. The vented sintered plate is a more recent development, which
has a high specific energy but is more expensive. It is characterized by its flat
discharge voltage profile and superior high current rate and low-temperature
performance. A sealed nickel/cadmium battery incorporates a specific cell
design feature to prevent a build-up of pressure in the cell caused by gassing
during overcharge. As a result, the battery requires no maintenance.
The major manufacturers of the nickel/cadmium battery for EV and HEV
allocation are SAFT and VARTA. Recent EVs powered by the nickel/cadmium
battery have included the Chrysler TE Van, Citroën AX, Mazda Roadster,
Mitsubishi EV, Peugeot 106, and Renault Clio.4,5
388
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
12.1.6.2.3 Ni–MH Battery
The Ni–MH battery has been on the market since 1992. Its characteristics
are similar to those of the nickel/cadmium battery. The principal difference
between them is the use of hydrogen, absorbed in a metal hydride, for the
active negative electrode material in place of cadmium. Because of superior
specific energy to Ni–Cd and since it is free from toxicity or carcinogenicity,
such as cadmium, the Ni–MH battery is superseding the Ni–Cd battery.
The overall reaction in an Ni–MH battery is
MH + NiOOH ↔ M + Ni(OH)2 .
(12.19)
When the battery is discharged, metal hydride in the negative electrode is
oxidized to form metal alloy, and nickel oxyhydroxide in the positive electrode
is reduced to nickel hydroxide. During charging, the reverse reaction occurs.
At present, Ni–MH battery technology has a nominal voltage of 1.2 V
and attains a specific energy of 65 Wh/kg and a specific power of 200 W/kg.
A key component of the Ni–MH battery is the hydrogen storage metal alloy,
which should be formulated to obtain a material that is stable over a large
number of cycles. There are two major types of metal alloys being used. These
are the rare-earth alloys based on lanthanum nickel, known as AB5 , and alloys
consisting of titanium and zirconium, known as AB2 . The AB2 alloys have a
higher capacity than the AB5 alloys. However, the trend is to use AB5 alloys
because of better charge retention and stability characteristics.
Because the Ni–MH battery is still under continual development, its advantages based on present technology are summarized as follows: it has the
highest specific energy (70–95 Wh/kg) and highest specific power (200–
300 W/kg) of nickel-based batteries; environmental friendliness (cadmium
free); a flat discharge profile (smaller voltage drop); and rapid recharge
capability. However, this battery still suffers from its high initial cost. It may
also have a memory effect and be exothermic during charging.
The Ni–MH battery has been considered as an important near-term
choice for EV and HEV applications. A number of battery manufacturers,
such as GM Ovonic, GP, GS, Panasonic, SAFT, VARTA, and YUASA, have
actively engaged in the development of this battery technology, especially for
powering EVs and HEVs. Since 1993, Ovonic has installed its Ni–MH battery
in the Solectric GT Force EV for testing and demonstration. A 19-kWh battery
has delivered over 65 Wh kg, 134 km/h, acceleration from zero to 80 km/h
in 14 s, and a city driving range of 206 km. Toyota and Honda have used the
Ni–MH battery in their HEVs—Prius and Insight, respectively.4,5
12.1.6.3
Lithium-Based Batteries
Lithium is the lightest of all metals and presents very interesting characteristics from an electrochemical point of view. Indeed, it allows a very high
thermodynamic voltage, which results in a very high specific energy and
Peaking Power Sources and Energy Storages
389
specific power. There are two major technologies of lithium-based batteries:
Li–P and Li–I.
12.1.6.3.1 Li–P Battery
Li–P batteries use lithium metal and a transition metal intercalation oxide
(My Oz ) for the negative and positive electrodes, respectively. This My Oz possesses a layered structure into which lithium ions can be inserted, or from
which they can be removed on discharge and charge, respectively. A thin
solid polymer electrolyte (SPE) is used, which offers the merits of improved
safety and flexibility in design. The general electrochemical reactions are
xLi + My Oz ↔ Lix My Oz .
(12.20)
On discharge, lithium ions formed at the negative electrode migrate through
the SPE and are inserted into the crystal structure at the positive electrode.
On charge, the process is reversed. By using a lithium foil negative electrode
and a vanadium oxide (V6 O13 ) positive electrode, the Li/SPE/V6 O13 cell is
the most attractive one within the family of Li–Ps. It operates at a nominal
voltage of 3 V and has a specific energy of 155 Wh/kg and a specific power of
315 W/kg. The corresponding advantages are a very low self-discharge rate
(about 0.5% per month), capability of fabrication in a variety of shapes and
sizes, and safe design (reduced activity of lithium with solid electrolyte). However, it has the drawback of relatively weak low-temperature performance
due to its temperature dependence of ionic conductivity.4
12.1.6.3.2 Li–I Battery
Since the first announcement of the Li–I battery in 1991, the Li–I battery technology has seen an unprecedented rise to what is now considered to be the
most promising rechargeable battery of the future. Although still in the stage
of development, the Li–I battery has already gained acceptance for EV and
HEV applications.
The Li–I battery uses a lithiated carbon intercalation material (Lix C) for
the negative electrode instead of metallic lithium, a lithiated transition
metal intercalation oxide (Li1−x My Oz ) for the positive electrode, and a liquid organic solution or a solid polymer for the electrolyte. Lithium ions are
swinging through the electrolyte between the positive and negative electrodes during discharge and charge. The general electrochemical reaction is
described as
Lix C + Li1−x My Oz ↔ C + LiMy Oz .
(12.21)
On discharge, lithium ions are released from the negative electrode, migrate
via the electrolyte, and are taken up by the positive electrode. On charge the
process is reversed. Possible positive electrode materials include Li1−x CoO2 ,
Li1−x NiO2 , and Li1−x Mn2 O4 , which have the advantages of stability in air,
high voltage, and reversibility for the lithium intercalation reaction.
390
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
The Lix C/Li1−x NiO2 type, loosely written as C/LiNiO2 or simply called
the nickel-based Li–I battery, has a nominal voltage of 4 V, a specific energy
of 120 Wh/kg, an energy density of 200 Wh/L, and a specific power of
260 W/kg. The cobalt-based type has higher specific energy and energy
density, but with a higher cost and significant increase of the self-discharge
rate. The manganese-based type has the lowest cost and its specific energy
and energy density lie between those of the cobalt-based and nickel-based
types. It is anticipated that the development of the Li–I battery will ultimately
move to the manganese-based type because of the low cost, abundance, and
environmental friendliness of the manganese-based materials.
Many battery manufacturers, such as SAFT, GS Hitachi, Panasonic, SONY,
and VARTA, have actively engaged in the development of the Li–I battery.
Starting in 1993, SAFT focused on the nickel-based Li–I battery. Recently,
SAFT reported the development of Li–I high-power batteries for HEV applications with a specific energy of 85 Wh/kg and a specific power of 1350 W/kg.
SAFT also announced high-energy batteries for EV applications with about
150 Wh/kg and 420 W/kg (at 80% SOC, 150 A current, and 30 s), respectively.2
12.2
Ultracapacitors
Because of the frequent stop-and-go operation of EVs and HEVs, the discharging and charging profile of the energy storage is highly varied. The
average power required from the energy storage is much lower than the peak
power for acceleration and hill climbing in a relatively short duration. The
ratio of peak power to average power can reach over 10:1 (Chapter 2). In HEV
design, the peak power capacity of the energy storage is more important than
its energy capacity, and usually constrains its size reduction (refer to Chapters 8 and 9). Based on present battery technology, battery design has to carry
out the trade-off among specific energy, specific power, and cycle life. The
difficulty in simultaneously obtaining high values of specific energy, specific
power, and cycle life has led to some suggestions that the energy storage
system of EV and HEV should be a hybridization of an energy source and
a power source. The energy source, mainly batteries and fuel cells, has high
specific energy, whereas the power source has high specific power. Power
sources can be recharged from the energy source during less demanding
driving or regenerative braking. The power source that has received wide
attention is the ultracapacitor.
12.2.1
Features of Ultracapacitors
The ultracapacitor is characterized by a much higher specific power but a
much lower specific energy compared to batteries. Its specific energy is in
391
Peaking Power Sources and Energy Storages
the range of a few watt-hours per kilogram. However, its specific power can
reach up to 3 kW/kg, much higher than any type of battery. Due to the low
specific energy density and the dependence of terminal voltage on SOC, it is
difficult to use ultracapacitors alone as an energy storage for EVs and HEVs.
Nevertheless, there are a number of advantages that can result from using the
ultracapacitor as an auxiliary power source. One promising application is the
so-called battery and ultracapacitor hybrid energy storage system for EVs and
HEVs.4,6 Specific energy and specific power requirements can be decoupled,
thus affording an opportunity to design a battery that is optimized for specific
energy and cycle life with little attention being paid to specific power. Due
to the load leveling effect of the ultracapacitor, the high current discharging
from the battery and the high current charging to the battery by regenerative
braking are minimized so that available energy, endurance, and life of the
battery can be significantly increased.
12.2.2
Basic Principles of Ultracapacitors
Double-layer capacitor technology is the major approach to achieving the
ultracapacitor concept. The basic principle of a double-layer capacitor is illustrated in Figure 12.8. When two carbon rods are immersed into a thin sulfuric
acid solution, separated from each other and applied with increasing voltage from zero to 1.5 V, almost nothing happens up to 1 V; then at a little over
1.2 V, a small bubble appears on the surface of both electrodes. Bubbles at a
voltage above 1 V indicate the electrical decomposition of water. Below the
Charger
Polarizing
electrodes
Collector
+
+
+
+
+
+
+
+
+
+
+
+
Collector
Separator
– Electrolyte
–
–
–
–
–
–
–
–
–
+
–
–
–
–
+
+
+
+
+
+
+
+
+
+
+
+
+
–
–
–
–
–
–
–
–
–
–
–
Electric double layers
FIGURE 12.8 Basic principles of a typical electric double-layer capacitor.
–
392
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
decomposition voltage, while the current does not flow, an “electric double
layer” occurs at the boundary of electrode and electrolyte. The electrons are
charged across the double layer and for a capacitor.
An electrical double layer works as an insulator only below the decomposing voltage. The stored energy, Ecap , is expressed as
Ecap =
1
CV 2 ,
2
(12.22)
where C is the capacitance in faraday and V is the usable voltage in volts.
This equation indicates that the higher rated voltage V is desirable for larger
energy density capacitors. Up to now, a capacitor’s rated voltage with aqueous
electrolyte has been about 0.9 V per cell, and with nonaqueous electrolyte it
is 2.3–3.3 V for each cell.
There is great merit in using an electric double layer in place of plastic or
aluminum oxide films in a capacitor, since the double layer is very thin—as
thin as one molecule with no pin holes—and the capacity per area is quite
large, at 2.5–5 μF/cm2 .
Even if a few μF/cm2 are obtainable, the energy density of capacitors is not
large when using aluminum foil. For increasing the capacitance, electrodes
are made from specific materials that have a very large area, such as activated
carbons, which are famous for their surface areas of 1000–3000 m2 /g. Ions
are adsorbed on those surfaces, and result in 50 F/g (1000 m2 /g × 5 F/cm2 ×
10, 000 cm2 /m2 = 50 F/g). Assuming that the same weight of electrolyte is
added, 25 F/g is quite a large capacity density. Nevertheless, the energy density of these capacitors is far smaller than that of batteries; the typical specific
energy of ultracapacitors is at present about 2 Wh/kg, only 1/20 of 40 Wh/kg,
which is the available value of typical lead–acid batteries.
12.2.3
Performance of Ultracapacitors
The performance of an ultracapacitor may be represented by terminal voltages during discharge and charge with different current rates. There are three
parameters in a capacitor: the capacitance itself (its electric potential VC ), the
series resistance RS , and the dielectric leakage resistance, RL , as shown in
Figure 12.9. The terminal voltage of the ultracapacitor during discharge can
be expressed as
Vt = VC − iRS .
(12.23)
The electric potential of the capacitor can be expressed as
i + iL
dVC
=−
,
dt
C
where C is the capacitance of the ultracapacitor.
(12.24)
393
Peaking Power Sources and Energy Storages
+
i
RS
iL
+
Vt
iC
VC
–
RL
C
–
FIGURE 12.9 Ultracapacitor equivalent circuit.
On the other hand, the leakage current iL can be expressed as
iL =
VC
.
RL
(12.25)
Substituting Equation 12.25 into Equation 12.24, one can obtain
VC
i
dVC
=−
.
+
dt
CRL
C
(12.26)
The terminal voltage of the ultracapacitor cell can be represented by the
diagram shown in Figure 12.10. The analytical solution of Equation 12.26 is
VC = VC0 −
VC
–
0
t
i t/CRL
e
dt e−(t/CRL ) ,
C
1
CRL
+
∫
+
i
– 1
C
+
RS
FIGURE 12.10 Block diagram of the ultracapacitor model.
+
(12.27)
Vt
+
–
394
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Cell terminal voltage (V)
2.5
2.0
I= 50 A
1.5
100
1.0
200
300
0.5
0
400
600
0
20
40
60
80
Discharge time (s)
100
120
140
FIGURE 12.11 Discharge characteristics of the 2600F Maxwell Technologies ultracapacitor.
where i is the discharge current and is a function of time in real operation.
The discharge characteristics of a Maxwell 2600F ultracapacitor are shown
in Figure 12.11. At different discharge current rates, the voltage linearly
decreases with discharge time. At a large discharge current, the voltage
decreases much faster than at a small current rate.
A similar model can be used to describe the charging characteristics of
an ultracapacitor and readers who are interested may carry out their own
analysis.
The operation efficiency in discharging and charging can be expressed as
follows:
Discharging:
ηd =
V t It
(VC − It RS ) It
=
V C IC
VC (It + IL )
(12.28)
ηc =
VC IC
VC (It − IL )
=
,
V t It
(VC + It RS ) It
(12.29)
and
Charging:
where Vt is the terminal voltage and It is the current input to or output
from the terminal. In actual operation, the leakage current IL is usually very
small (few mA) and can be ignored. Thus, Equations 12.28 and 12.29 can be
rewritten as
Discharging:
ηd =
VC − RS It
Vt
=
Vc
Vc
(12.30)
395
Peaking Power Sources and Energy Storages
100
Discharge efficiency (%)
90
80
It = 50 A
100 A
70
60
50
200 A
300 A
40 400 A
30
FIGURE 12.12
600 A
0.5
0
1
1.5
Terminal voltage (V)
2
2.5
Discharge efficiency of the 2600F Maxwell Technologies ultracapacitor.
and
Charging:
ηc =
VC
VC
=
.
VC + R S I t
Vt
(12.31)
The above equations indicate that the energy loss in an ultracapacitor is
caused by the presence of series resistance. The efficiency decreases at high
current rate and low cell voltage, as shown in Figure 12.12. Thus, in actual
operation, the ultracapacitor should be kept at its high-voltage region, for
higher than 60% of its rated voltage.
The energy stored in an ultracapacitor can be obtained through the energy
needed to charge it to a certain voltage level, that is,
EC =
0
t
VC IC dt =
0
V
CVC dVC =
1
CVC2 ,
2
(12.32)
where VC is the cell voltage in volts. At its rated voltage, the energy stored in
the ultracapacitor reaches its maxima. Equation 12.32 indicates that increasing the rated voltage can significantly increase the stored energy since the
energy increases with the voltage squared. In real operation, it is impossible to completely utilize the stored energy because of the low power in the
low state-of-energy (SOE). Thus, an ultracapacitor is usually given a bottom
voltage, VCb , below which the ultracapacitor will stop delivering energy.
Consequently, the available or useful energy for use is less than its fully
396
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
1
64% of energy
available
0.9
0.8
0.7
SOE
0.6
0.5
0.4
0.3
0.2
40% voltage drop
0.1
0
0
0.5
1
1.5
Cell voltage (V)
2
2.5
FIGURE 12.13 SOE versus cell voltage.
charged energy, which can be expressed as
Eu =
1 2
2
,
C VCR − VCb
2
(12.33)
where VCR is the rated voltage of the ultracapacitor.
The usable energy in an ultracapacitor can also be expressed in SOE, which
is defined as the ratio of the energy in the ultracapacitor at a voltage of VC to
the energy at full charged voltage, VCR , as expressed by
SOE =
0.5 CVC2
2
0.5 CVCR
=
VC
VCR
2
.
(12.34)
For example, 60% of the rated voltage is the bottom voltage and 64%
of the total energy is available for use, as shown in Figure 12.13.
12.2.4
Ultracapacitor Technologies
According to the goals set by the U.S. Department of Energy for the inclusion of ultracapacitors in EVs and HEVs, the near-term specific energy
and specific power should be better than 5 Wh/kg and 500 W/kg, respectively, while the advanced performance values should be over 15 Wh/kg and
1600 W/kg. So far, none of the available ultracapacitors can fully satisfy these
goals. Nevertheless, some companies are actively engaged in the research
and development of ultracapacitors for EV and HEV applications. Maxwell
Technologies has claimed that its power BOOSTCAP® ultracapacitor cells
(2600 F at 2.5 V) and integrated modules (145 F at 42 V and 435 F at 14 V) are
in production. The technical specifications are listed in Table 12.3.
397
Peaking Power Sources and Energy Storages
TABLE 12.3
Technical Specifications of Maxwell Technologies Ultracapacitor Cell and
Integrated Modules7
BCA P0010
(Cell)
Capacitance (farads, −20%/+20%)
Maximum series resistance ESR at 25◦ C (mΩ)
Voltage (V): continuous (peak)
Specific power at rated voltage (W/kg)
Specific energy at rated voltage (Wh/kg)
Maximum current (A)
Dimensions (mm) (reference only)
Weight (kg)
Volume (L)
Operating temperaturea (◦ C)
Storage temperature (◦ C)
Leakage current (mA) 12 h, 25◦ C
a
BMOD0115
(Module)
BMOD0117
(Module)
2600
145
0.7
2.5 (2.8)
4300
4.3
600
60 × 172
(cylinder)
0.525
0.42
10
4
42 (50)
14 (17)
2900
1900
2.22
1.82
600
600
195 × 165 × 415 195 × 265 × 145
(box)
(box)
16
6.5
22
7.5
435
−35 to +65
−35 to +65
−35 to +65
−35 to +65
−35 to +65
−35 to +65
5
10
10
Steady-state case temperature.
12.3
Ultra-High-Speed Flywheels
The use of flywheels for storing energy in mechanical form is not a new
concept. More than 25 years ago, the Oerlikon Engineering Company in
Switzerland made the first passenger bus solely powered by a massive
flywheel. This flywheel, weighing 1500 kg and operating at 3000 rpm, was
recharged by electricity at each bus stop. The traditional flywheel is a massive steel rotor with hundreds of kilograms that spins on the order of ten
hundreds of rpm. On the contrary, the advanced flywheel is a lightweight
composite rotor with tens of kilograms and rotates on the order of ten
thousands of rpm; it is the so-called ultra-high-speed flywheel.
The concept of ultra-high-speed flywheels appears to be a feasible means
for fulfilling the stringent energy storage requirements for EV and HEV
applications, namely high specific energy, high specific power, long cycle life,
high energy efficiency, quick recharge, maintenance-free characteristics, cost
effectiveness, and environmental friendliness.
12.3.1
Operation Principles of Flywheels
A rotating flywheel stores energy in the kinetic form as
Ef =
1 2
Jf ω ,
2 f
(12.35)
398
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
where Jf is the moment of inertia of the flywheel in kg m2 /s and ωf is the
angular velocity of the flywheel in rad/s. Equation 12.35 indicates that
enhancing the angular velocity of the flywheel is the key technique to increasing its energy capacity and reducing its weight and volume. At present, a
speed of over 60,000 rpm has been achieved in some prototypes.
With current technology it is difficult to directly use the mechanical energy
stored in a flywheel to propel a vehicle, due to the need for continuous variation transmission with a wide gear ratio variation range. The commonly
used approach is to couple an electric machine to the flywheel directly or
through a transmission to constitute a so-called mechanical battery. The
electric machine, functioning as the energy input and output port, converts the mechanical energy into electric energy or vice versa, as shown in
Figure 12.14.
Equation 12.35 indicates that the energy stored in a flywheel is proportional to the moment of inertia of the flywheel and flywheel rotating speed
squared. A lightweight flywheel should be designed to achieve a large
moment of inertia per unit mass and per unit volume by properly designing its
V
I
Power
electronics
Axle
Housing
Flywheel
Rotor
Electric
machine
Stator
FIGURE 12.14 Basic structure of a typical flywheel system (mechanical battery).
Peaking Power Sources and Energy Storages
399
W(r)
R2
r
R1
FIGURE 12.15 Geometry of a typical flywheel.
geometric shape. The moment of inertia of a flywheel can be calculated by
Jf = 2πρ
R2
W(r)r3 dr,
(12.36)
R1
where ρ is the material mass density and W(r) is the width of the flywheel
corresponding to radius r, as shown in Figure 12.15. The mass of the flywheel
can be calculated by
R2
W(r)r dr.
(12.37)
Mf = 2πρ
R1
Thus, the specific moment of inertia of a flywheel, defined as the moment
of inertia per unit mass, can be expressed as
R2
Jfs =
3
R1 W(r)r dr
.
R2
R1 W(r)r dr
(12.38)
Equation 12.38 indicates that the specific moment of inertia of a flywheel
is independent of its material mass density and dependent solely on its
geometric shape W(r).
For a flywheel with equal width, the moment of inertia is
Jf = 2πρ(R24 − R14 ) = 2πρ(R22 + R12 )(R22 − R12 ).
(12.39)
The specific moment of inertia is
Jfs = R22 + R12 .
(12.40)
400
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
The volume density of the moment of inertia, defined as the moment of
inertia per unit volume, is, indeed, associated with the mass density of the
material. The volume of the flywheel can be obtained by
R2
W(r)r dr.
(12.41)
Vf = 2π
R2
The volume density of the moment of inertia can be expressed as
R
ρ R12 W(r)r3 dr
.
Jf V = R
2
R1 W(r)r dr
(12.42)
For a flywheel with equal width, the volume density of the moment of
inertia is
(12.43)
Jf V = ρ R22 + R12 .
Equations 12.42 and 12.43 indicate that heavy material can, indeed, reduce
the volume of the flywheel with a given moment of inertia.
12.3.2
Power Capacity of Flywheel Systems
The power that a flywheel delivers or obtains can be obtained by differentiating Equation 12.35 with respect to time, that is,
Pf =
dωf
dEf
= J f ωf
= ωf Tf ,
dt
dt
(12.44)
where Tf is the torque acting on the flywheel by the electric machine. When
the flywheel discharges its energy, the electric machine acts as a generator and
converts the mechanical energy of the flywheel into electric energy. On the
other hand, when the flywheel is charged, the electric machine acts as a motor
and converts electric energy into mechanical energy stored in the flywheel.
Equation 12.44 indicates that the power capacity of a flywheel system depends
completely on the power capacity of the electric machine.
An electric machine usually has the characteristics shown in Figure 12.16,
which has two distinct operating regions—constant-torque and constantpower regions. In the constant-torque region, the voltage of the electric
machine is proportional to its angular velocity, and the magnetic flux in the
air gap is constant. However, in the constant-power region, the voltage is
constant and the magnetic field is weakened with the increase of machine
angular velocity. During charging of the flywheel, that is, accelerating the
flywheel from a low speed, ω0 , to a high speed, the maximum speed, ωmax ,
for example, the torque delivered from the electric machine is
Tm = Jf
dωf
,
dt
(12.45)
401
Peaking Power Sources and Energy Storages
Torque
Constant torque,
increasing voltage
and constant flux
Voltage
Constant power,
constant voltage
and field weakening
wo
wb
wmax
Angular velocity
FIGURE 12.16 Typical torque and voltage profile versus rotational speed.
where it is assumed that the electric machine is directly connected to the
flywheel. The time, t, needed can be expressed as
t=
ωmax
ω0
Jf
dω =
Tm
ωb
ω0
Jf
dω +
pm /ωb
ωmax
ωb
Jf
dω.
pm /ω
(12.46)
With the given accelerating time, t, the maximum power of the electric
machine can be obtained from Equation 12.46 as
Pm =
Jf 2
ωb − 2ω0 ωb + ω2max .
2t
(12.47)
Equation 12.47 indicates that the power of the electric machine can be
minimized by designing its corner speed or base speed, ωb , equal to the
bottom speed of the flywheel, ω0 . This conclusion implies that the effective
operating speed range of the flywheel should coincide with the constantpower region of the electric machine. The power of the electric machine can
be minimized as
Jf 2
(12.48)
ωmax − ω20 .
Pm =
2t
Another advantage achieved by coinciding the operating speed range of
the flywheel with the constant-power speed range is that the voltage of the
electric machine is always constant (refer to Figure 12.16), therefore thus
significantly simplifying the power management system, such as DC/DC
converters and their controls.
402
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
TABLE 12.4
Composite Materials for Ultra-High-Speed Flywheel4
E-glass
Graphite epoxy
S-glass
Kevlar epoxy
12.3.3
Tensile Strength σ (MPa)
Specific Energy ρ (kg/m3 )
Ratio σ/ρ (Wh/kg)
1379
1586
2069
1930
1900
1500
1900
1400
202
294
303
383
Flywheel Technologies
Although higher rotational speed can significantly increase the stored energy
(Equation 12.35), there is a limit beyond which the tensile strength σ of the
material constituting the flywheel cannot withstand the stress resulting from
the centrifugal force. The maximum stress acting on the flywheel depends
on its geometry, specific density ρ, and rotational speed. Maximum benefit
can be obtained by adopting flywheel materials that have the maximum ratio
of σ/ρ. Note that if the speed of the flywheel is limited by material strength,
the theoretical specific energy is proportional to the ratio of σ/ρ. Table 12.4
summarizes the characteristics of some composite materials for ultra-highspeed flywheels.
A constant-stress principle may be employed for the design of ultra-highspeed flywheels. To achieve maximum energy storage, every element in the
rotor should be equally stressed to its maximum limit. This results in a shape
of gradually decreasing thickness that theoretically approaches zero as the
radius approaches infinity, as shown in Figure 12.17.4
Magnetic bearing
Vacuum
Flywheel
Rotor of the electric machine
Stator of the electric machine
Terminals
Housing
FIGURE 12.17 Basic structure of a typical flywheel system.
Magnetic bearing
Peaking Power Sources and Energy Storages
403
Due to the extremely high rotating speed and in order to reduce the aerodynamic loss and frictional loss, the housing inside which the flywheel is
spinning is always highly vacuumed, and noncontact, magnetic bearings are
employed.
The electric machine is one of the most important components in the flywheel system, since it has a critical impact on the performance of the system.
At present, PM brushless DC motors are usually accepted in the flywheel
system. Apart from possessing high power density and high efficiency, the
PM brushless DC motor has the unique advantage that no heat is generated
inside the PM rotor, which is particularly essential for the rotor to work in a
vacuum environment to minimize windage loss.
A switched reluctance machine (SRM) is also a very promising candidate
for application in a flywheel system. SRM has a very simple structure and
can operate efficiently at very high speed. In addition, SRM presents a large
extended constant-power speed region, which allows more energy in the flywheel that can be delivered (refer to Section 12.3.2). In this extended speed
region, only the machine excitation flux is varied, which is easily realized. On
the contrary, the PM brushless motor shows some difficulty in weakening the
field flux induced by the PM.
In contrast to applying the ultra-high-speed flywheel for energy storage in
stationary plants, its application to EVs and HEVs suffers from two specific
problems. First, gyroscopic forces occur whenever a vehicle departs from its
straight-line course, such as in turning and in pitching upward or downward
from road grades. These forces essentially reduce the maneuverability of the
vehicle. Second, if the flywheel is damaged, its stored energy in mechanical
form will be released in a very short period of time. The corresponding power
released will be very high, which can cause severe damage to the vehicle. For
example, if a 1-kWh flywheel breaks apart in 1–5 s, it will generate a huge
power of 720–3600 kW. So containment in the case of failure is presently the
most significant obstacle to implementing the ultra-high-speed flywheel in
EVs and HEVs.
The simplest way of alleviating gyroscopic forces is to use multiple smaller
flywheels. By operating them in pairs (one half spinning in one direction and
another half in the opposite direction), the net gyroscopic effect becomes theoretically zero. Practically, it still has some problems related to the distribution
and coordination of these flywheels. Also the overall specific energy and specific power of all flywheels may be smaller than a single one. Similarly, the
simplest way of minimizing the damage due to breakage of the ultra-highspeed flywheel is to adopt multiple small modules, but this means vehicle
performance suffers from the possible reduction of specific energy and specific power. Recently, a new failure containment has been proposed. Instead
of diminishing the thickness of the rotor’s rim to zero based on the maximum
stress principle, the rim thickness is purposely enlarged. Hence, the neck
area just before the rim (virtually a mechanical fuse) will break first at the
instant the rotor suffers from a failure. Owing to the use of this mechanical
404
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
fuse, only the mechanical energy stored in the rim needs to be released or
dissipated in the casing on failure.4
Many companies and research agencies (such as Lawrence Livermore
National Laboratory (LLNL) in the United States, Ashman Technology,
AVCON, Northrop Grumman, Power R&D, Rocketdyne/Rockwell Trinity
Flywheel US Flywheel Systems, Power Center at UT Austin, and so on)
have engaged in the development of ultra-high-speed flywheels as energy
storages of EVs and HEVs. However, technologies of ultra-high-speed flywheels are still in their infancy. Typically, the whole ultra-high-speed flywheel
system can achieve a specific energy of 10–150 Wh/kg and specific power of
2–10 kW/kg. LLNL has built a prototype (20 cm diameter and 30 cm height)
that can achieve 60,000 rpm, 1 kWh, and 100 kW.
12.4
12.4.1
Hybridization of Energy Storages
Concept of Hybrid Energy Storage
The hybridization of energy storage involves to combining two or more
energy storages together so that the advantages of each can be brought out and
the disadvantages can be compensated by others. For instance, the hybridization of a chemical battery with an ultracapacitor can overcome problems such
as the low specific power of chemical batteries and low specific energy of
ultracapacitors, thus achieving high specific energy and high specific power.
Basically, the hybridized energy storage consists of two basic energy storages, one with high specific energy and the other with high specific power.
The basic operation of this system is illustrated in Figure 12.18. In highpower demand operation, such as acceleration and hill climbing, both basic
energy storages deliver their power to the load as shown in Figure 12.18a.
On the other hand, in low-power demand operation, such as constant-speed
cruising operation, the high specific energy storage will deliver its power
to the load and charge the high specific power storage to recover its charge
lost during high-power demand operation, as shown in Figure 12.18b. In
regenerative braking operation, the peak power will be absorbed by the
high specific power storage, and only a limited part is absorbed by the
high specific energy storage. In this way, the whole system would be much
smaller in weight and size than if any one of them alone was the energy
storage.
12.4.2
Passive and Active Hybrid Energy Storage with
Battery and Ultracapacitor
Based on the available technologies of various energy storages, there are
several viable hybridization schemes for EVs and HEVs: typically, battery
405
Peaking Power Sources and Energy Storages
(a)
High power demand
High specific
energy storage
Power
converter
Load
Power
converter
Load
Power
converter
Load
High specific
power storage
(b)
Low power demand
High specific
energy storage
High specific
power storage
(c)
Negative power
High specific
energy storage
High specific
power storage
Primary power flow
Secondary power flow
FIGURE 12.18 Concept of a hybrid energy storage operation. (a) Hybrid powering, (b) power
split, and (c) hybrid charging.
and battery hybrids, and battery and ultracapacitor hybrids. The latter is
more natural because the ultracapacitor can offer a much higher power
than batteries, and it collaborates with various batteries to form the battery
and ultracapacitor hybrids. During the hybridization, the simplest way is to
directly and parallelly connect the ultracapacitors to the batteries as shown in
Figure 12.19. In this configuration, the ultracapacitors simply act as a current
filter, which can significantly level the peak current of the batteries and reduce
the battery voltage drop as shown in Figures 12.20 and 12.21. The major disadvantages of this configuration are that the power flow cannot be actively
controlled and the ultracapacitor energy cannot be fully used.
Figure 12.22 shows a configuration in which a two-quadrant DC/DC converter is placed between the batteries and ultracapacitors. This design allows
the batteries and the ultracapacitors to have a different voltage; also the power
flow between them can be actively controlled and the energy in the ultracapacitors can be fully used. In the long term, an ultra-high-speed flywheel
will replace the batteries in hybrid energy storage to obtain a high efficiency,
compact, and long-life storage system for EVs and HEVs.
406
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Batteries
Ultracapacitor
+
–
FIGURE 12.19 Direct and parallel connection of batteries and ultracapacitors.
12.4.3
Battery and Ultracapacitor Size Design
The best design of a hybrid energy storage with a battery and an ultracapacitor
is that the overall energy and power capacities just meet the energy and power
requirements of the vehicle without much margins.8 The energy and power
requirements of a vehicle to its energy storage can be represented by the
Current (A)
100
Load
current
50
0
Battery
current
–50
Ultracapacitor
current
–100
Voltage (V)
110
Battery & ultracapacitor
105
Battery
alone
100
95
90
0
5
10
15
20
25
30
35
40
45
50
Time (s)
FIGURE 12.20 Variation of battery and ultracapacitor currents and voltages with a step current
output change.
407
Peaking Power Sources and Energy Storages
300
Battery current
Ultracapacitor current
250
200
Current (A)
150
100
50
0
–50
–100
–150
–200
0
200
400
600
800
1000
1200
1400
Time (s)
FIGURE 12.21 Battery and ultracapacitor currents during operation of HEV in an FTP75 urban
drive cycle.
energy/power ratio, which is defined as
Re/p =
Er
,
Pr
(12.49)
Two
quadrant
DC/DC
converter
Batteries
Ultracapacitors
where Er and Pr are the energy and power required by the vehicle, respectively.
The energy and power requirements mostly depend on the design of the
vehicle drive train and control strategy as discussed in Chapters 7 through 9.
When Re/p is known, the battery and ultracapacitor in the hybrid storage can
be designed so that the energy/power ratio of the hybrid energy storage is
FIGURE 12.22 Actively controlled hybrid battery/ultracapacitor energy storage.
408
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
equal to Re/p , which is expressed as8
W b Eb + W c E c
= Re/p ,
W b Pb + W c P c
(12.50)
where Wb and Wc are the weights of the battery and ultracapacitor, respectively, Eb and Ec are the specific energies of the battery and ultracapacitor,
respectively, and Pb and Pc are the specific powers of the battery and
ultracapacitor, respectively.
Equation 12.50 can be further written as
Wc = kWb ,
where
k=
(12.51)
Eb − Re/p Pb
.
Re/p Pc − Ec
(12.52)
Thus, the specific energy of the hybrid energy storage is
Espe =
Wb Eb + Wc Ec
Eb + kEc
=
Wb + W c
1+k
(12.53)
and the specific power of the hybrid energy storage is
Pspe =
Wb Pb + Wc Pc
Pb + kPc
=
.
Wb + W c
1+k
(12.54)
An example is shown in the following.
Suppose that a vehicle needs a 50-kW energy storage; the desired
energy/power ratio Re/p = 0.07 h, that is, 3.5-kWh energy is required and the
battery and ultracapacitor characteristics are shown in Tables 12.5 and 12.6.
The weights needed for a single source and hybrid sources are listed in
TABLE 12.5
Major Parameters of CHPS Battery Alternative at
Standard Testing8
CHPS Battery
Alternative
Specific
Energy (Wh/kg)
Specific
Power (W/kg)
Energy/Power
(h)
Lead–acid
NiCd
Ni–MH
Li–I (CHPS)a
28
50
64
75
120
140
0.373
0.417
0.457
100
1000b
a
b
0.1
Combat Hybrid Power System sponsored by TACOM.
Power capabilities depend on pulse length and temperature.
409
Peaking Power Sources and Energy Storages
TABLE 12.6
Characteristic Data of a 42-V Ultracapacitor7
Rated capacitance (DCCa , 25◦ C)
Capacitance tolerance
Rated voltage
Surge voltage
Max. series resis., ESR (DCC, 25◦ C)
(F)
(%)
(V)
(V)
145
±20
42
50
(mΩ)
(W/kg)
(A)
( J)
(Wh/kg)
10
2900
600
128,000
2.3
Weight
Volume
Operating temperature
(mA)
(kg)
(l)
(◦ C)
30
15
22
Storage temperature
(◦ C)
Lifetime (25◦ C)
(year)
Specific power density (42 V)
Max. current
Max. stored energy
Specific energy density (42 V)
Max. leakage current (12 h, 25◦ C)
−35 to +65
−35 to +65
10, C < 20% of initial value,
ESR < 200% of initial value
500,000, C < 20% of initial value,
ESR < 200% of initial value
Cyclability (25◦ C, I = 20 A)
a
DCC: discharging at constant current.
Tables 12.7 and 12.8. Comparing the total weights in Tables 12.7 and 12.8,
it is obvious that the hybrid energy storage can save the weight significantly,
especially with a battery that has low power density.
TABLE 12.7
Characteristic Data of a 42-V Ultracapacitor7
Lead/Acid
Ni/Cd
Ni/MH
Li–I
Ultracap
75
30
667
120
50
417
140
64
357
1000
100
50
2500
2
1750
Specific power (W/kg)
Specific energy (Wh/kg)
Total weight (kg)
TABLE 12.8
Characteristic Data of a 42-V Ultracapacitor7
Specific power (W/kg)
Specific energy (Wh/kg)
Battery weight (kg)
Ultracap weight (kg)
Total weight (kg)
Lead/Acid
Ni/Cd
Ni/MH
378.5
26.5
116
16.5
132
581.4
40.7
69
16.7
86
703
49.2
54
16.9
71
Li–I
1222
85.5
35
6.05
41
410
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
References
1. D. A. J. Rand, R. Woods, and R. M. Dell, Batteries for Electric Vehicles, Society of
Automotive Engineers (SAE), Warrendale, PA, 1988.
2. Available at http://www.saftbatteries.com, SAFT, The Battery Company, 2007.
3. T. R. Crompton, Battery Reference Book, Society of Automotive Engineers (SAE),
Warrendale, PA, 1996.
4. C. C. Chan and K. T. Chau, Modern Electric Vehicle Technology, Oxford University
Press, Oxford, 2001.
5. Y. Gao and M. Ehsani, “Investigation of battery technologies for the army’s
hybrid vehicle application,” Proceedings of the IEEE 56th Vehicular Technology
Conference, pp. 1505–1509, Fall 2002.
6. Y. Gao, H. Moghbelli, M. Ehsani, G. Frazier, J. Kajs, and S. Bayne, “Investigation of
high-energy and high-power hybrid energy storage systems for military vehicle
application,” Society of Automotive Engineers (SAE) Journal, Paper No. 2003-01-2287,
Warrendale, PA, 2003.
7. Available at http://www.maxwell.com, Maxwell Technologies, 2007.
8. Y. Gao and M. Ehsnai, “Parametric design of the traction motor and energy
storage for series hybrid off-road and military vehicles,” Power Electronics, IEEE
Transactions, 21 (3), 749–755, May 2006.
13
Fundamentals of Regenerative Breaking
One of the most important features of EVs, HEVs, and fuel cell vehicles (FCVs)
is their ability to recover significant amounts of braking energy. The electric
motors in EVs, HEVs, and FCVs can be controlled to operate as generators
to convert the kinetic or potential energy of vehicle mass into electric energy
that can be stored in the energy storage and then reused.
The braking performance of a vehicle is an important factor in vehicle safety.
A successfully designed braking system for a vehicle must always meet the
distinct demand of quickly reducing vehicle speed and maintaining vehicle
direction controllable by the steering wheel. The former requires the braking
system to be able to supply sufficient braking torque on all wheels. The latter requires proper braking force distribution on all wheels, as discussed in
Chapter 2.
Generally, the braking torque required is much larger than the torque that
an electric motor can produce, especially in heavy braking. In EVs, HEVs,
and FCVs, mechanical friction braking systems have to coexist with electrical
regenerative braking. Therefore, this is a hybrid braking system. As in the
hybrid propulsion system, there are many configurations and control strategies. However, the final goal of the design and control of such systems is to
ensure the vehicle’s braking performance and its ability to recover as much
braking energy as possible.
13.1
Braking Energy Consumed in Urban Driving
A significant amount of energy is dissipated by braking.1–3 Braking a 1500kg vehicle from 100 km/h to zero speed dissipates about 0.16 kWh of energy
[(1/2)MV 2 ] in a few tens of meters. If this amount of energy is dissipated by
coasting and only by drag forces (rolling resistance and aerodynamic drag)
without braking, the vehicle will travel about 2 km, as shown in Figure 13.1.
When a vehicle is driving in a stop-and-go pattern in urban areas, a
significant amount of energy is dissipated by frequent braking. Successful
design of the hybrid braking system for recovering as much of the braking
energy as possible requires a full understanding of braking behavior and its
411
412
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
100
2.0
1.6
60
1.2
Speed
40
0.8
20
0
Distance (km)
Speed (km/h)
Distance
80
0.4
0
50
100
150
0
200
Coating time (s)
FIGURE 13.1 Coasting speed and distance.
characteristics with respect to vehicle speed, braking power, deceleration rate,
and so on during typical urban driving cycles.2,3,4 The typical urban driving
cycles that are used in this chapter are EPA FTP75, LA92, US06, New York
City, and ECE-15.
While driving on a flat road, the driving power on the vehicle wheels can
be calculated by
V
1
dV
2
Mgfr + ρa CD AV + Mδ
(kW),
Pd =
1000
2
dt
(13.1)
where M is the vehicle mass in kg, g is the gravitational acceleration,
9.81 m/s2 , fr is the tire rolling resistance coefficient, ρa is the air mass density, 1.205 kg/m3 , CD is the aerodynamic drag coefficient, A is the frontal area
of the vehicle in m2 , V is the vehicle speed in m/s, δ is the rotational inertia factor, and dV/dt is the vehicle acceleration in m/s2 (and is negative for
deceleration). For Pd > 0, the traction wheels accept power from the power
plants and push the vehicle forward. In this case, the braking power is zero.
In contrast, Pd < 0 when braking and the kinetic energy of the vehicle mass
is dissipated by the brake system. In this case, the driving power is zero.
Integrating Equation 13.1 over the driving time in a given driving cycle can
give both the traction energy and braking energy, as shown in Figure 13.2, for
a typical passenger car with the parameters listed in Table 13.1 in the FTP75
urban driving cycle. The vehicle parameters used in this chapter are shown
in Figure 13.3 and Table 13.1.
Figure 13.2 and Table 13.2 indicate that the braking energy in typical urban
areas may reach up to more than 34% of the total traction energy. In large
cities, such as New York City, it may reach up to 80%.
413
Fundamentals of Regenerative Breaking
Vehicle speed (km/h)
100
80
60
40
20
0
0
200
400
600
800
1000
1200
1400
1200
1400
Energy (kWh)
1.5
1
Traction
0. 5
Braking
0
0
200
400
600
800
1000
Driving time (s)
FIGURE 13.2 Traction and braking energy dissipation in an FTP75 urban driving cycle.
13.2
Braking Energy versus Vehicle Speed
Braking energy distribution over vehicle speed in typical urban driving cycles
is useful information for the design and control of the regenerative brake
system. In the speed range in which the braking energy is most dissipated, the
operating efficiency of the electric motor, functioning as a generator, may be of
most concern. In other speed ranges, regenerative braking may be abandoned
TABLE 13.1
Vehicle Parameters Used in this Paper
Item
Symbol Unit
kg
Value
Vehicle mass
Rolling resist. coefficient
M
fr
1500 (fully, loaded), 1250 (unloaded)
Aerodynamic drag coefficient
Front area
CD
A
Wheel base
Distance from gravity center
to front wheel center
L
La
m
2.2
2.7
m
1.134 (fully loaded), 0.95 (unloaded)
Gravity center height
hg
m
0.6 (fully loaded), 0.5 (unloaded)
0.01
0.3
m2
414
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Mj
hg
Mg
Fbf
Wf
Fbr
Lb
La
L
Wr
FIGURE 13.3 Forces acting on a vehicle in braking.
with no significant compromise on energy recovered. Figure 13.4 shows such
a diagram of braking energy distribution over vehicle speed while driving
in the FTP75 urban driving cycle for a vehicle whose parameters are listed
in Table 13.1. Figure 13.5 further shows the braking energy dissipated in a
speed range that is less than a given speed. These two figures indicate that
only 10% of the total braking energy is dissipated in the speed range below
15 km/h. Table 13.3 shows the braking energy dissipated in the speed range
below 15 km/h while driving in other typical urban driving cycles.
The braking energy dissipated in the low-speed range, such as below
15 km/h in all the typical driving cycles, is insignificant. This result indicates
that we need not attempt to obtain high operating efficiency at low speeds in
TABLE 13.2
Maximum Speed, Average Speed, Total Traction Energy, and Energies Dissipated by
the Drag and Braking per 100 km Traveling Distance in Different Drive Cycles
FTP75 Urban
LA92
US06
New York
ECE15
Max. speed (km/h)
Ave. speed (km/h)
86.4
27.9
107.2
39.4
128.5
77.4
44.6
12.2
120
49.8
Traveling distance per cycle (km)
Traction energy (kWh)
Per cycle
Per km
Braking energy (kWh)
Per cycle
Per km
Percentage of braking energy to
traction energy
10.63
15.7
12.8
1.90
7.95
1.1288
1.1062
2.3559
0.15
2.2655
0.1769
0.2960
0.1555
0.9691
0.1219
0.6254
0.0589
55.4
1.3666
0.0870
58.01
0.9229
0.0721
40.73
0.2425
0.1274
81.9
0.3303
0.0416
34.08
415
Fundamentals of Regenerative Breaking
20
Braking energy percentage
distributing on vehicle speed (%)
18
16
14
12
10
8
6
4
2
0
0
10
20
30
40
50
60
70
80
90 100
Vehicle speed (km/h)
FIGURE 13.4 Braking energy distribution over vehicle speed in an FTP75 urban driving cycle.
Braking energy percentage in the speed range
below the given speed shown in horizontal axis (%)
the design and control of regenerative braking. In fact, it is difficult to regenerate at low speeds, because of the low motor electromotive force (voltage)
generated at low motor rotational speeds.
It should be noted that the rotational speed of the vehicle driven wheels,
which is proportional to motor angular speed, is decoupled from the translatory speed of the vehicle body when the vehicle wheels are close to being
locked. Thus, the operation of the hybrid brake system must be at speeds
120
100
80
60
40
20
0
0
10
20
30 40 50 60 70
Vehicle speed (km/h)
80
90
100
FIGURE 13.5 Braking energy dissipated over the speed range below the given speed.
416
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
TABLE 13.3
Braking Energy Dissipated in the Speed Range Below 15 km/h in Typical
Urban Driving Cycles
Braking energy |V < 15 km/h (%)
FTP75 Urban
LA92
US06
New York
ECE-15
10.93
5.51
3.27
21.32
4.25
higher than a minimum threshold value. Electric regenerative braking should
be applied primarily to recapture as much braking energy as possible. At
speeds lower than this threshold, mechanical braking should be primarily
applied to ensure the vehicle’s braking performance.
13.3
Braking Energy versus Braking Power
Another important factor is braking energy versus braking power. Understanding braking energy versus braking power in a typical driving cycle
is very helpful for power capacity design of the electric motor drive and
on-board energy storage, so that they are capable of recovering most of the
braking energy without oversize design.
Figure 13.6 shows the braking simulation results for the vehicle whose
parameters are listed in Table 13.1 while driving in an FTP75 urban driving
cycle. This figure indicates that around 15% of the total braking energy is dissipated in the braking power range greater than 14.4 kW. This result implies
that a 15-kW electric motor can recover about 85% of the total braking energy
in this driving cycle. Table 13.4 shows the simulation results for other urban
driving cycles, which also indicates the braking power range in which 85% of
the total braking energy is dissipated. These data are good indicators of the
design of the power capacity of the electric motor and the on-board energy
storage from the braking point of view.
13.4
Braking Power versus Vehicle Speed
Another important consideration is the braking power characteristics versus vehicle speed in typical urban driving cycles. Understanding this is very
TABLE 13.4
The Braking Power Range in which 85% of Braking Energy is Dissipated
in Typical Urban Driving Cycles
Power range in which 85% of
total energy is consumed
FTP75 Urban
LA92
US06
New York
ECE-15
0–14.4
0–44.5
0–46.5
0–18.5
0–33.5
417
Braking energy percentage (%) at the braking power
range greater than the power on the x axis
Fundamentals of Regenerative Breaking
100
90
80
70
60
50
40
30
20
10
0
0
2
4
6
8
10
12
14
Braking power (kW)
16
18
20
FIGURE 13.6 Braking energy percentage in the range that is greater than the power shown in
the horizontal axis.
helpful for proper design and control of the speed–power profile of electric motors in order to optimally match the driving application. Figure 13.7
shows the simulation results for the vehicle mentioned above. The bars in
Figure 13.7 represent the maximum braking power in particular driving cycles
at specified vehicle speeds. The solid lines represent the supposed motor
speed–power profiles that can recover at least 85% of the braking energy as
indicated in Table 13.4.
It can be seen that the braking power versus vehicle speed profiles naturally match the power–speed characteristics of the motor, in that the power
is proportional to the speed from zero speed to base speed (constant torque)
and is constant beyond the base speed. Thus, electric motors do not need a
special design and control for regenerative braking purposes.
13.5
Braking Energy versus Vehicle Deceleration Rate
Another important consideration is the braking energy distribution over the
vehicle deceleration rate, which reflects the required braking force. Understanding this feature will also help the design and control of the hybrid
braking system of EVs, HEVs, and FCVs. Figure 13.8 shows the braking
energy consumed in the vehicle deceleration range of less than a value
shown in the horizontal axis for the vehicle mentioned above while driving
418
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
(a)
25
Braking power (kW)
20
15
10
5
0
0
10
20
30
40
50
60
70
80
90 100
Vehicle speed (km/h)
FTP75 urban
(b)
(c)
100
90
70
Braking power (kW)
80
Braking power (kW)
80
70
60
50
40
30
60
50
40
30
20
20
10
10
0
0
0
20
40
60
80
Vehicle speed (km/h)
100
120
0
20
40
60
80
100
Vehicle speed (km/h)
US06
0
20
40
60
80
100
Vehicle speed (km/h)
LA92
(d) 30
35
Braking power (kW)
Braking power (kW)
140
(e) 40
25
20
15
10
5
0
120
30
25
20
15
10
5
0
10
20
30
40
Vehicle speed (km/h)
New York city
50
60
0
120
140
ECE-15
FIGURE 13.7 Braking power versus vehicle speed in typical urban driving cycles. (a) FTP75
urban, (b) LA92, (c) US06, (d) New York city, and (e) ECE-15.
in the FTP75 urban driving cycle. It can be seen in this figure that braking
in this diving cycle is very gentle (the maximum deceleration rate is less
than 0.15g). Table 13.5 shows the maximum deceleration rates and braking
energy in a deceleration rate of less than 0.15g in other typical urban driving
cycles.
419
Fundamentals of Regenerative Breaking
110
Braking energy in the deceleration range of less
than the given values in the horizantal axis (%)
100
90
80
70
60
50
40
30
20
10
0
0
0.1
0.05
Deceleration rate (g)
0.15
FIGURE 13.8 Braking energy dissipated in various vehicle deceleration rates.
TABLE 13.5
Maximum Deceleration Rates and Braking Energy Dissipated
in the Deceleration Range of Less Than 0.15g
Maximum deceleration rate, g
% of braking energy consumed in
the deceleration range of less
than 0.15 g
13.6
FTP75 Urban
LA92
US06
New York
ECE-15
0.12
100
0.40
56
0.31
59
0.27
69
0.14
100
Braking Energy on Front and Rear Axles
The braking performance for passenger cars requires the braking force distribution on the front and rear axles to be below the I curve, but above the
braking regulation curve as shown in Figures 2.40 and 2.41 in Chapter 2. This
requirement implies that most of the braking force is applied on the front axle.
Consequently, regenerative braking on the front axle is better than that on the
rear axle. However, for other kinds of vehicles, such as trucks, the rear axle
may be better.
420
13.7
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Brake System of EV, HEV, and FCV
Regenerative braking in EVs, HEVs, and FCVs introduces slight complexity
to the braking system design. Two basic questions arise: how to distribute the
total required braking forces among regenerative braking and frictional braking to recover as much braking energy as possible and how to distribute the
total braking forces on the front and rear axles to achieve stable braking performance. Usually, regenerative braking is effective only for the driven axle (the
front axle for passenger cars). The electric motor must be controlled to produce
the proper amount of braking force for recovering braking energy as much
as possible and, at the same time, the total braking force must be sufficient
to meet vehicle deceleration commanded by the driver. This chapter introduces two configurations of hybrid braking systems and their corresponding
design and control principles. One is the parallel hybrid brake system, which
has a simple structure and control and retains all the major components of
the conventional brakes. The other is a fully controllable hybrid brake system, which can fully control the braking force for each individual wheel, thus
greatly enhancing the vehicle’s braking performance on all types of roads.
The analysis in the following sections is based on the braking performance
described in Section 2.9 of Chapter 2.
13.7.1
Parallel Hybrid Braking System
Perhaps the simplest and closest to conventional pure mechanical brakes
(hydraulic or pneumatic) is the parallel hybrid brake system,1,3 which retains
all the major components of conventional mechanical brakes and adds electric braking directly on the front axle as shown in Figure 13.9. The mechanical
brake system consists of a master cylinder and booster. It may or may not have
an ABS controller and actuator, but will have a brake caliper and brake disks.
The electric motor directly applies its braking torque to the front axle and is
controlled by the vehicle controller, based on vehicle speed and brake pedal
position signals, which represent the desired braking strength and braking
control strategy embedded in the vehicle controller. The feature of the parallel
hybrid braking system is that only the electric braking force (torque) is electronically controlled, and the mechanical braking force (torque) is controlled
by the driver through the brake pedal before the ABS starts its function. However, the mechanical braking force is controlled by the ABS when the wheels
are about to be locked, as in the conventional brake systems. The key problem
in the design and control of such a system is to properly control the electric
braking force for recovering as much braking energy as possible.
13.7.1.1
Design and Control Principles with Fixed Ratios between
Electric and Mechanical Braking Forces
Figure 13.10 shows a braking force allocation strategy, in which the mechanical brake has a fixed ratio of braking force distribution on the front and
421
Fundamentals of Regenerative Breaking
Position
sensor
Master cylinder
and booster
Brake
pedal
Brake
caliper
Brake
plate
Pressure
sensor
Speed
sensor
Axle
Vehicle
controller
ABS controller
and actuator
Energy
storage
Motor and
controller
Mechanical connections
Fluid connections
Control signals
Electrical connections
FIGURE 13.9 Schematic structure of the parallel hybrid brake system.
0.4
g
g
g
g
0.7
g
g
0.5
I
g
0.3
0.2
g
0.1
Mechanical
+
electrical
ECE
regulation
j=
g
0.1
Fbr
1.0
0.9
b
(Mechanical brake)
g
0.4
j=
0.15
0.6
j=
j=
0.2
0.8
j=
j=
0.25
j=
j=
0.3
j=
Rear braking force ratio, Fbf/Mg
j=
0.35
0.05
0
0
0.2
Fbf-mech
0.4
Fbf-regen
0.6
0.8
1
Front braking force ratio Fbf/Mg
FIGURE 13.10 Braking forces varying with deceleration rate.
422
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
rear wheels represented by the β line. Curve I is the ideal braking force
distribution of the vehicle. The ECE regulation, which stipulates the minimum rear braking force, is also plotted. The total braking force is the curve
labeled mechanical + electrical. The braking force on the front wheels consists
of mechanical braking, Fbf-mech , and electric braking, Fbf-regen , as shown in
Figure 13.10. When the wheel speed is lower than a given threshold, 15 km/h
for example, either very low vehicle speed or speed close to wheel lock-up, the
electric regenerative braking will produce no braking force and the braking
is completely done by the mechanical system.
When the wheel speed is higher than the given threshold and the desired
vehicle deceleration is less than a given value (0.15g in Figure 13.10), which
is represented by the stroke of the brake pedal, all the braking force is produced by electric regenerative braking and no mechanical braking force is
applied to the front and rear wheels. As described in Section 11.7, most of
the braking energy is in this deceleration range. Zero mechanical force may
be implemented by employing larger clearances between the brake pads and
plates or by carrying out small modifications to the conventional master cylinders. When the desired deceleration is greater than the given value (0.15g in
Figure 13.10), both mechanical and electric braking share the total front wheel
braking force as shown in Figure 13.10. The design of the electric braking
portion is associated with the power capacity of the electric motor and onboard energy storage. But the total braking force curve in Figure 13.10 must
be above the ECE regulation curve. When the desired deceleration is higher
than a given value (0.6g in Figure 13.10), the electric regenerative braking
force is gradually reduced to zero with increase of the desired deceleration
(0.9g in Figure 13.10). This design will ensure that the actual front and rear
wheel braking forces are close to the ideal braking distribution curve, resulting in short braking distances, and strong mechanical braking that may be
more relied upon for emergency cases. Figure 13.11 shows the total braking
force, regenerative braking force, and mechanical braking forces on the front
and rear wheels, respectively, along with vehicle deceleration rate.
With the design principle described above and the electric regenerative
braking control rule with respect to wheel speed, the braking energy available for recovering can be computed in various typical driving cycles by using
a computer model. The simulation results are listed in Table 13.6. The data
indicate that in normal urban driving, most of the braking energy can be
recovered.
It should be noted that the maximum braking torque required of the electric
motor is that which can produce 0.15g of deceleration. Thus, a large-sized
electric motor would not be needed.
13.7.1.2
Design and Control Principles for Maximum
Regenerative Braking
This design and control principle will follow the rule that allocates the total
braking force to the front wheels as much as possible, under the condition of
423
Fundamentals of Regenerative Breaking
Braking force ratios to the vehicle weight
0.8 0–a–b–c–d: Total braking force on front wheels
d
c
0–a–e–f–g: Regenerative braking forces on front wheels
0.7 0–h–i–c–d: Mechanical braking force on front wheels
0–h–j–k–m: Mechanical braking forces on rear wheels
0.6
b
0.5
0.4
i
0.3
0.2
0.1
0
k
e
a
j
f
h
0
0.1
0.2
0.3 0.4 0.5 0.6 0.7
Braking deceleration, g
0.8
0.9
m
g
1
FIGURE 13.11 Braking force ratios relative to the vehicle weight varying with deceleration
rate.
meeting the braking regulations (the ECE regulation used here), that is, the
total braking force distribution will follow the maximum front braking force
curve (minimum rear braking force) stipulated by the ECE regulation, which
is represented by curve 0–a–b–c as shown in Figure 13.12. More details are
described below.
When the braking strength is less than 0.2g, all the braking force is allocated
to the front wheels for regenerative braking and no mechanical braking force
is applied to the front or rear wheels. Motor torque may be controlled through
sensing the braking pedal position. In this case, the master cylinder will not
establish hydraulic pressure. When the braking strength is greater than 0.2g,
the mechanical system starts creating pressure and the mechanical braking
forces on the front and rear wheels start increasing, following the βm line. At
the same time, the electric motor adds its electric braking torque on the front
wheels to make the total braking force follow the ECE regulation line. For
example, when the braking strength required is 0.5g, the total braking force
is at point b and the mechanical braking force on the front and rear wheels is
at point d. The maximum possible braking force for regenerative braking is
the segment of d–b, labeled Fregen-max-possi. . However, for fully recovering all
TABLE 13.6
Percentage of Total Braking Energy Available
for Recovering
FTP75 Urban
LA92
US06
New York
ECE-15
89.69
82.92
86.55
76.16
95.75
424
0.4
r-lines
j=
m = 0.2
0.1
m = 0.2
j=
g
0.05
i
h
f
l
1.0
m = 1.0
m=
0.9
m = 0.9
m=
m = 0.8
m = 0.67
m = 0.7
m =0
.8
bhb-max
g
e
ECE regulation
b
d
0
m = 0.6
m = 0.4
0.1
m = 0.3
m = 0.3
bm
m =0
.7
m = 0 m = 0.67
.6
m = 0.5
m = 0.4
g
0.2
0.15
f-lines
g
1.0
g
0.3
m = 0.5
g
0.8
g
0.5
j=
0.2
g
0.7
g
0.6
j=
g
0.4
j=
0.25
g
0.9
j=
j=
j=
0.3
j=
0.35
j=
Ratio of braking force on rear wheels to vehicle weight, Fbr/Mg
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
c
a
0
0.1
Ff-mach
Ff-mach
0.2
0.3
0.4
Fregen
Fmotor-max
0.5
0.6
0.7
0.8
0.9
1
Ratio of braking force on front wheels
to vehicle weight, Fbf /Mg
Fmotor-max
Fregen-max-possi
FIGURE 13.12 Schematic illustration of braking force distribution on the front wheels
(electrical + mechanical) and the rear wheels.
this maximum possible braking power, two conditions have to be satisfied.
One condition is that the electric motor should be capable of producing this
braking force. Assuming that the maximum electric motor braking force is
limited to that shown in Figure 13.12 by segment f–e, with a braking strength
of 0.5g, the operating point of the total braking force should be at point e and
the mechanical braking force at point f. Assuming that the electric motor is
powerful enough, another condition will have to be met to recover the maximum possible braking power limited by the ECE regulation. This condition
is that the road adhesive coefficients must be larger than 0.67. Otherwise, the
front wheels will be locked. Figure 13.12 shows a case of braking strength
j = 0.5 g and μ = 0.6 and an operating point of total braking force at point
g. In this case, the operation will be different for brake systems with and
without a mechanical ABS. For the system without ABS, to meet the braking
force requirement on the rear wheels, the mechanical brake has to operate at
point h. Thus the regenerative braking force takes the segment h–g, which is
smaller than the maximum braking force that the electric motor can produce.
Fundamentals of Regenerative Breaking
425
However, for the brake system with a mechanical ABS, when the front wheels
are close to being locked, the ABS will start its function and the mechanical
braking force on the front wheels will be decoupled from the β line, instead
following the f line of μ = 0.6; this limits further increase in the mechanical
braking force on the front wheels. In this case, the electric motor can still
produce its maximum braking force for maximum braking energy recovery
as shown in Figure 13.12 by the segment i–g and the operating point of the
mechanical braking is at point i.
The above design and control principle for maximum regenerative braking is based on the idea that the maximum possible braking force on the
front wheels is limited by the ECE regulation. However, it can be seen from
Figure 13.12 that the ECE regulation produces a nonlinear braking force distribution curve. This nonlinearity may lead to a complex design and control.
A simple straight line can be used to replace the ECE regulation in the design
and control of the hybrid brake system as shown in Figure 13.12 by the line
βhb-max . The line βhb-max is generated as follows.
The minimum braking force on the rear wheels (maximum braking force
on the front wheels) when the front wheels are locked, stipulated by the ECE
regulation, is expressed as
Fbf
q + 0.07
,
≤
Wf
0.85
(13.2)
where Fbf is the total braking force on the front wheels, Wf is the vertical
loading on the front wheels, and q is the braking strength, q = j/g (refer to
Figure 13.3). As in the definition of the β line in Section 2.9, the front wheel
braking force can be expressed as
Fbf = βhb Fb ,
(13.3)
where Fb is the total braking force of the vehicle, which is related to the braking
strength, q, as
Fb = Mj = Mgq.
(13.4)
The braking strength, q, and the vertical loading on the front wheels, Wf ,
have the relationship
Mg Wf =
(13.5)
Lb + qhg ,
L
where M is the mass of the vehicle, L is the wheel base, and Lb is the length
from the vehicle center of gravity to the rear axle as shown in Figure 13.3.
Combining Equations 13.3 through 13.5, we obtain
Fbf
βbh qL
=
.
Wf
Lb + qhg
(13.6)
426
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
From Equations 13.2 and 13.6, we obtain
βbh ≤
(q + 0.07)(Lb + qhg )
.
0.85qL
(13.7)
It can be seen from Equation 13.7 that the upper limit of βhb to meet the ECE
regulation is a function of braking strength q. The q, at which the maximum
βhb is achieved, can be obtained by
dβbh 0 =0
dq q=q
with q0 =
(13.8)
0.07Lb /hg ; then we obtain
βbh- max
2 0.07Lb hg + Lb + 0.07hg
.
=
0.85L
(13.9)
Equation 13.9 indicates that this braking force distribution ratio is only determined by the vehicle parameters. With βbh-max , the braking force distribution
on the front and rear wheels can be plotted as shown in Figure 13.12. This line
can be used to replace the ECE regulation curve and the design and control
of the braking system may be simplified. The analysis for the braking process
is similar to the above and is left to the reader.
13.7.2
Fully Controllable Hybrid Brake System
In recent years, more advanced braking systems are emerging that allow controlling the braking force on each wheel independently.1,2,5 Hydraulic electric
brake systems (H-EBSs) and mechanical electric brake systems are two typical examples. Figure 13.13 schematically shows a fully controllable hybrid
brake system, which consists of a hydraulic electric brake and an electric
regenerative brake.
The mechanical brake system consists mainly of a brake pedal and its position sensor, a master cylinder, an electrically operated and controlled brake
actuator, electrically controlled three-port switches, a fluid accumulator, and a
pressure sensor. In normal operation, ports 1 and 3 of the three-port switches
are open and port 2 is closed. The mechanical braking torque applied on each
wheel is independently produced by the corresponding brake actuator, which
is commanded by the H-EBS controller. The torque command to each wheel is
generated in the H-EBSs, based on the pressure signal from the pressure sensor, the brake pedal stroke signal from the brake pedal position sensor, wheel
speed signal from the wheel speed sensor, and the embedded control rule in
the H-EBS controller. The brake fluid from the master cylinder flows into the
fluid accumulator through the three-port switches to establish pressure and
427
Fundamentals of Regenerative Breaking
Brake pedal
Pressure sensor
Speed sensor
Brake pedal
position
sensor
Master cylinder
H-EBS and ERB
controller
Brake plat
Brake caliper
Three-port switch
Electrically powered
brake actuator
Motor drive and
transmission Energy
storage
2
1
3
3
2 1
1 2
3
1
3
2
Fluid accumulator
FIGURE 13.13 Fully controllable hybrid brake system with H-EBS and electric regenerative
braking.
emulate the braking feeling of a conventional brake system. In the case of
failure in any brake actuator, the corresponding three-port switch will switch
to the mode of ports 1 and 2 open and port 3 closed for the brake fluid from
the master cylinder to directly go to the brake caliper cylinder and therefore
to maintain the braking torque.
The electric regenerative brake mainly includes an electric motor and its
controller (drive) and an on-board energy storage. An ERB controller is used
to control electric braking, based on wheel speed, brake pedal stroke, charge
condition in the energy storage, and control rules embedded in the controller.
One of the key problems in this system is how to control the mechanical
and electric braking torques to obtain acceptable braking performance and
to recover, as much as possible, the available regenerative braking energy. In
this chapter, two typical control strategies are introduced; one emphasizes
the braking performance and the other the maximum regenerative braking
energy recovery.
13.7.2.1
Control Strategy for Optimal Braking Performance
Due to the independent control of braking force on each wheel, the fully
controllable hybrid brake system can be controlled to apply braking forces
on the front and rear wheels in the way of following the ideal braking force
distribution curve. This control strategy can yield optimal brake performance.
Figure 13.14 illustrates the principle of this control strategy for a vehicle,
on which electric regenerative braking is available only on the front wheels.
When the required total braking force on the front wheels is smaller than that
produced by the electric motor, the electric motor produces the total braking
428
0.4
0.35
ECE
regulation
g
Fbr-mech
Fbr-mech
g
g
g
a
0.1
0
1.0
g
g
0.8
0.9
j=
0.7
g
0.6
0.5
g
0.3
g
0.2
0.05
0
b
j=
0.1
j=
j=
j=
0.15
I
g
0.4
0.2
j=
j=
0.25
j=
j=
0.3
j=
Ratio of braking force on rear
wheels to the vehicle total weight
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
0.1
0.2
Fbf-regen
Fbf-regen-max
0.3
0.4
Fbf-mech
0.5
0.6
0.7
0.8
0. 9
1
Ratio of braking force on front
wheels to the vehicle total weight
FIGURE 13.14 Control strategy for the best braking performance.
force and no mechanical braking force is applied. However, the mechanical
braking produces the total braking force for the rear wheels to follow the I
curve, as shown by point a in Figure 13.14. When the required total braking
force on the front wheels is greater than that produced by the electric motor,
both electric braking and mechanical braking have to be applied. For more
braking energy recapture, the electric motor should be controlled to produce
its maximum braking force that is limited by the electric motor or energy
storage. The remaining is applied by the mechanical brake as shown by point
b in Figure 13.14.
It should be noted that at low front wheel speeds caused by actual low
vehicle speed or close to locked wheels, it is hard to produce braking torque
by the electric motor due to its low electromotive force (voltage) generated in
the stator windings of the electric motor. Therefore, in this case, the mechanical
brake has to produce the total braking force as required.
As seen in Figure 13.14, a significant amount of braking energy is dissipated
by the rear brakes, especially in weak braking (small deceleration). For example, at j = 0.3 g, around 33% of the total braking energy is dissipated in the rear
brakes. At j = 0.1 g, this percentage reaches 37.8%. Unfortunately, this is just
the case for most of the urban driving cycles. Considering no regenerative
braking at low wheel speeds (<15 km/h), the braking energy available for
recovery on the front wheels is considerably reduced. The simulation results,
shown in Table 13.7, prove this conclusion.
429
Fundamentals of Regenerative Breaking
TABLE 13.7
Scenarios of Braking Energy in Typical Urban Driving Cycles
Percentage of braking energy on
front wheels to total braking
energy
Percentage of braking energy on
rear wheels to total braking
energy
Percentage of available
regenerative braking energy on
front wheels to total braking
energy
13.7.2.2
FTP75 Urban
LA92
US06
New York
ECE-15
61.52
63.16
62.98
62.57
61.92
38.48
36.84
37.02
37.43
38.08
55.16
59.85
60.89
50.26
59.27
Control Strategy for Optimal Energy Recovery
The principle of this control strategy aims to allocate more braking force to
the front wheels under the condition of the front wheels never locking earlier
than the rear wheels on a road with any adhesive coefficient. Thus, more
braking energy will be available for regenerative braking.
The details of this control strategy are explained below with the help of
Figure 13.15.
When the vehicle brakes with an acceleration rate j on a road with an adhesive coefficient μ, and j/g < μ, the braking forces on the front and rear wheels
can be arbitrarily applied as long as the total braking force meets the requirements, that is, Fbf + Fbr = Mj. However, braking performance requires that
no wheel be locked and that the braking force on the rear wheels be above
the ECE regulation curve as shown in Figure 13.15. Thus, the braking forces
on the front and rear wheels are variable in a certain range, which is dependent on vehicle deceleration rate and road adhesive coefficient. Figure 13.15
shows the braking force ranges of a–b and c–d for the deceleration rates of
j/g = 0.7 and j/g = 0.6 (strong braking), respectively, on a road with adhesive
coefficient μ = 0.9 (concrete road). Obviously, for j/g = 0.7, the maximum
braking force on the front wheels is determined by point b, which is dictated by the f line (front wheel locked) of μ = 0.9. However, for j/g = 0.6,
the maximum braking force on the front wheels is determined by point d,
which is dictated by the ECE regulation. Actually, on this high adhesive road,
when j/g < 0.7, the braking force on the rear wheels can be very small and
almost all the braking force can be applied to the front wheels. However,
when the road adhesive coefficient is smaller (slippery road), the braking
force variable range is much smaller. Figure 13.15 shows a case with μ = 0.4
(wet mud road), and j/g = 0.3 and j/g = 0.2. Obviously point f determines
the maximum braking force on the front wheels for j/g = 0.3 and point h for
j/g = 0.2.
430
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
0.4
l
m = 0.3
h
0.1
0.2
Fb-regen-max
bhb-max
f
d
0.3
0.4
Fbf-mech
m = 1.0
m = 0.7
m = 1.0
m = 0.4
e
m = 0.3
m = 0.2
k
m
m = 0.4 n
m = 0.2
m = m = 1.0
0.9
m = 0.9
g
0.3
The ratio of braking force on front
wheels to the total vehicle weight
g
0.9
a
m = 0.6
j=
g
0.2
0
0
g
g
1.0
j=
g
0.4
j=
0.05
c
m = 0.7
0.2
0.1
g
0.7
g
0.6
j=
0.25
0.15
j=
j=
0.3
j=
0.35
r-lines
f-lines
0.5
b
ECE
regulation
0.6
0.7
0.8
0.9
Ratio of braking force on front
wheels to the total vehicle weight
1
FIGURE 13.15 Depicted control strategy for optimal energy recovery.
The above analysis provides only a control principle for the hybrid brake
system to obtain maximum braking energy on the front wheels to make
more braking energy recoverable. However, the power capacities of the electric motor and energy storage are usually not big enough to handle the
huge braking power when the braking is very strong. In this case, it is supposed that the electric motor provides its maximum braking torque and
the mechanical brake provides the remaining. Figure 13.15 shows an electric motor that produces its maximum braking force on the front wheels
to brake the vehicle at point n ( j/g = 0.4). It is obvious that when the
vehicle deceleration is less, j/g = 0.4, the electric motor itself can handle it
and no mechanical braking is needed. However, when the braking deceleration is bigger, j/g = 0.6 for example, the required braking force for the
front wheels is bigger than the electric motor can handle. In this case, the
mechanical brake has to apply additional braking force to make the operation at any point in the range m–d. It is obvious that the best operation
point is m.
With a simplified straight line βhb-max replacing the ECE regulation, very
similar analysis can be done for this control strategy. This work is left to the
reader.
Fundamentals of Regenerative Breaking
431
References
1. S. R. Cikanek and K. E. Bailey, Energy Recovery Comparison Between Series and
Parallel Braking System for Electric Vehicles Using Various Drive Cycles, Advanced
Automotive Technologies, American Society of Mechanical Engineers (ASME),
New York, DSC vol. 56/DE Vol. 86, pp. 17–31, 1995.
2. Y. Gao, L. Chu, and M. Ehsani, “Design and control principle of hybrid braking
system for EV, HEV and FCV,” 2007 IEEE VPPC.
3. Y. Gao, L. Chen, and M. Ehsani, “Investigation of the effectiveness of regenerative
braking for EV and HEV,” Society of Automotive Engineers (SAE) Journal, SP-1466,
Paper No. 1999-01-2901, 1999.
4. H. Gao, Y. Gao, and M. Ehsani, “Design issues of the switched reluctance motor
drive for propulsion and regenerative braking in EV and HEV,” in Proceedings of
the SAE 2001 Future Transportation Technology Conference, Costa Mesa, CA, Paper
No. 2001-01-2526, August 2001.
5. Y. Gao and M. Ehsani, “Electronic braking system of EV and HEV—integration of
regenerative braking, automatic braking force control and ABS,” in Proceedings of
the SAE 2001 Future Transportation Technology Conference, Costa Mesa, CA, Paper
No. 2001-01-2478, August 2001.
14
Fuel Cells
In recent decades, the application of fuel cells in vehicles has been the focus
of increased attention. In contrast to a chemical battery, the fuel cell generates
electric energy rather than storing it and continues to do so as long as a fuel
supply is maintained. Compared with the battery-powered EVs, the fuel-cellpowered vehicle has the advantages of a longer driving range without a long
battery charging time. Compared with the ICE vehicles, it has the advantages
of high energy efficiency and much lower emissions due to the direct conversion of free energy in the fuel into electric energy, without undergoing
combustion.
14.1
Operating Principles of Fuel Cells
A fuel cell is a galvanic cell in which the chemical energy of a fuel is converted
directly into electrical energy by means of electrochemical processes. The fuel
and oxidizing agent are continuously and separately supplied to the two
electrodes of the cell, where they undergo a reaction. Electrolyte is necessary to conduct the ions from one electrode to the other as shown in Figure
14.1. Fuel is supplied to the anode or positive electrode, where electrons are
released from the fuel under catalyst. The electrons, under the potential difference between these two electrodes, flow through the external circuit to the
cathode electrode or negative electrode, where combining positive ions and
oxygen, reaction products, or exhaust are produced.
The chemical reaction in a fuel cell is similar to that in a chemical battery.
The thermodynamic voltage of a fuel cell is closely associated with the energy
released and the number of electrons transferred in the reaction.1,2 The energy
released by the cell reaction is given by the change in Gibbs free energy, ΔG,
usually expressed in per mole quantities. The change in Gibbs free energy in
a chemical reaction can be expressed as
ΔG =
Products
Gi −
Gj ,
(14.1)
Reactants
433
434
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
e–
Load
I
Oxidant
(O2 or air )
Fuel
Positive ions
+
+
+
+
Exhaust
Anode
electrode
Electrolyte
Cathode
electrode
FIGURE 14.1 Basic operation of a fuel cell.
where Gi and Gj are the free energies in species i of products and species j
of reactants. In a reversible process, ΔG is completely converted into electric
energy, that is,
ΔG = −nFVr ,
(14.2)
where n is the number of electrons transferred in the reaction, F = 96, 495 is
the Faraday constant in coulombs per mole, and Vr is the reversible voltage
of the cell. At standard conditions (25◦ C temperature and 1 atm pressure), the
open-circuit (reversible) voltage of a cell can be expressed as
Vr0 = −
ΔG0
,
nF
(14.3)
where ΔG0 is the change in Gibbs free energy at standard conditions. ΔG is
expressed as
ΔG = ΔH − T ΔS,
(14.4)
where ΔH and ΔS are the enthalpy and entropy changes, respectively, in the
reaction at absolute temperature T. Table 14.1 shows the values of standard
enthalpy, entropy, and Gibbs free energy of some typical substances.3 Table
14.2 shows the thermodynamic data for some reactions in a fuel cell at 25◦
and 1 atm pressure.3
The “ideal” efficiency of a reversible galvanic cell is related to the enthalpy
for the cell reaction by
ηid =
ΔS
ΔG
=1−
T.
ΔH
ΔH
(14.5)
435
Fuel Cells
TABLE 14.1
Standard Enthalpy of Formation and Gibbs Free Energy for Typical Fuels
Substance
0
ΔH298
(kJ/mol)
Formula
ΔS0298
(kJ/mol K)
0
0
0
0
ΔG298
(kJ/mol)
Oxygen
Hydrogen
Carbon
Water
O (g)
H (g)
C (s)
H2 O (l)
0
0
0
0
0
0
−286.2
−0.1641
−237.3
Water
H2 O (g)
−242
−0.045
−228.7
Methane
Methanol
Ethanol
CH4 (g)
CH3 OH (l)
C2 H5 OH (l)
−74.9
−238.7
−0.081
−0.243
−50.8
−166.3
Carbon monoxide
Carbon dioxide
Ammonia
CO (g)
CO2
NH3 (g)
−277.7
−111.6
−393.8
−0.345
0.087
0.003
−174.8
−137.4
−394.6
−46.05
−0.099
−16.7
ηid will be 100% if the electrochemical reaction involves no change in the
number of gas moles, that is, when ΔS is zero. This is the case for reactions
C + O2 = CO2 . However, if the entropy change, ΔS, of a reaction is positive,
then the cell—in which this reaction proceeds isothermally and reversibly—
has at its disposal not only the chemical energy, ΔH, but also (in analogy to
a heat pump) a quantity of heat, TΔS, absorbed from the surroundings for
conversion into electrical energy (see Table 14.2) (Figure 14.2).
The change of free energy, and thus cell voltage, in a chemical reaction is
a function of the activities of the solution species. The dependence of cell
voltage on the reactant activities is expressed as
Vr =
Vr0
RT
(activities of products)
−
,
ln (activities of reactants)
nF
(14.6)
where R is the universal gas constant, 8.31 J/mol K, and T is the absolute
temperature in K. For gaseous reactants and products, Equation 14.6 can be
TABLE 14.2
Thermodynamic Data for Different Reactions at 25◦ C and 1◦ atm Pressure
H2 + 12 O2 → H2 O (l)
H2 + 12 O2 → H2 O (g)
C + 12 O2 → CO (g)
C + O2 → CO2 (g)
CO + 12 O2 → CO2 (g)
0
ΔH298
ΔS0298
(kJ/mol)
(kJ/mol K)
0
ΔG298
(kJ/mol)
n
E (V) ηid (%)
−286.2
−0.1641
−237.3
2
1.23
−242
−0.045
−228.7
2
1.19
94
−116.6
−393.8
0.087
0.003
−137.4
−394.6
2
4
0.71
1.02
124
100
−279.2
−0.087
−253.3
2
1.33
91
83
436
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Temperature (°C)
(a)
1.4
27
127
77
177
327
377
427
650
700
CO2
CO + 1/2O2
1.3
277
227
1.2
H2 + 1/2O2
1
0
V r (V)
1.1
C + O2
H2O(l)
CO2
0.9
0.8
0.7
0.6
250
C + 1/2O2
300
350
CO
400 450 500 550
Temperature (K)
600
Temperature (°C)
(b)
180
27
77
127
177
227
277
327
377
427
650
700
160
C + 1/2O2
CO
ηid, %
140
120
C + O2
100
CO2
CO2
CO + 1/2O2
80
40
250
H2O(l)
H2 + 1/2O2
60
300
350
400
450
500
550
600
Temperature (K)
FIGURE 14.2 Temperature dependence of cell voltage (a) and reversible efficiency (b).
expressed as
Vr =
Vr0
!
RT pi
−
vi ln 0 ,
nF
pi
i
(14.7)
where Vr is the voltage of the cell in which the reaction proceeds with gaseous
participants at nonstandard pressure pi , Vr0 is the corresponding cell voltage
437
Fuel Cells
with all gases at the standard pressure p0i (normally 1 atm), and vi is the number of moles of species i accounted as positive for products and negative for
reactants.
14.2
Electrode Potential and Current–Voltage Curve
Experiments have shown that the rest voltage, V, is usually lower than the
reversible voltage, Vr0 , calculated from the ΔG value. The voltage drop is
called the rest voltage drop, ΔV0 . The reason may be the existence of a significant kinetic hindrance to the electrode process, or because the process does
not take place in the manner assumed in the thermodynamic calculation of
Vr0 . This rest voltage drop depends, in general, on the electrode materials and
the kind of electrolyte.
When current is drawn from a cell, voltage drop is caused by the existence
of ohmic resistance in the electrode and electrolyte, which increases in direct
proportion to the current density, that is,
ΔVΩ = Re i,
(14.8)
where Re is the equivalent ohmic resistance per area and i is the current
density.
In a fuel cell, part of the generated energy is lost in pushing the species to
react, because extra energy is required to overcome the activation barriers.
These losses are called activation losses and are represented by an activation
voltage drop, ΔVa . This voltage drop is closely related to the materials of
electrodes and the catalysts. The Tafel equation is most commonly used to
describe this behavior, by which the voltage drop is expressed as3
RT
i
.
ΔVa =
ln
βnF
i0
(14.9)
Or more conveniently, it is written as
ΔVa = a + b ln(i),
(14.10)
where a = −(RT/βnF) ln(i0 ) and b = RT/βnF; i0 is the exchange current at
equilibrium state and b is constant depending on the process. For a more
detailed theoretical description, refer to pp. 230–236 of Messerle.3
When current flows, ions are discharged near the negative electrode,
and as a result the concentration of ions in this region tends to decrease.
If the current is to be maintained, ions must be transported to the electrode.
This takes place naturally by diffusion of ions from the bulk electrolyte and
by direct transport due to fields caused by concentration gradients. Bulk
438
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
movement of the electrolyte by convection or stirring also helps to bring
the ions up.
The voltage drop caused by the lack of ions is called concentration voltage drop, since it is associated with a decrease in the concentration of the
electrolyte in the immediate vicinity of the electrode. For small current
densities, the concentration voltage drop is generally small. However, as the
current density increases, it reaches a limit, when the maximum possible rate
of transport of ions to the electrode is approached as the concentration at the
electrode surface falls to zero.
The voltage drop caused by the concentration at the electrode where ions
are removed (cathode electrode in fuel cell) can be expressed as3
iL
RT
ln
(14.11)
ΔVc1 =
nF
iL − i
and at the electrode where ions are formed (anode electrode in fuel cell) as
iL + i
RT
,
(14.12)
ln
ΔVc2 =
nF
iL
where iL is the limiting current density.
The voltage drop caused by concentration is not only restricted to the electrolyte. When either the reactant or the product is gaseous, a change in partial
pressure in the reacting zones also represents a change in concentration. For
example, in a hydrogen–oxygen fuel cell, oxygen may be introduced into air.
When the reaction takes place, oxygen is removed near the electrode surface
in the pores of the electrode and the partial pressure of oxygen drops there
when compared with that in bulk air. The change in partial pressure causes a
voltage drop, which is determined by
RT
ps
,
(14.13)
ln
ΔVcg =
nF
p0
where ps is the partial pressure at the surface and p0 is the partial pressure in
the bulk feed. For more details, see pp. 236–238 of Messerle.3
Figure 14.3 shows the voltage–current curves of a hydrogen–oxygen fuel
cell with a temperature of 80◦ C. It can be seen that the drop caused by the
chemical reaction, including activation and concentration, is the source of the
voltage drop. This figure also indicates that improving the electrode materials and manufacturing using advanced technology, such as nanotechnology,
and advanced catalysts will significantly reduce the voltage drop and will
consequently improve the efficiency of the fuel cell.
The energy loss in a fuel cell is represented by the voltage drop. Thus the
efficiency of the fuel cell can be written as
ηfc =
V
,
Vr0
(14.14)
439
Fuel Cells
1.3
Vr
1.2
DV0
Cell voltage (V )
1.1
DVW
1.0
0.9
DVa+ DVc
0.8
0.7
0.6
0.5
0.4
0
0.2
0.4
0.6
0.8
Current density (A/cm2)
1.0
1.2
FIGURE 14.3 Current–voltage curves for a hydrogen–oxygen fuel cell at T = 80◦ C.
90
0.9
80
0.8
Efficiency (%)
70
0.7
Efficiency
60
0.6
50
0.5
40
0.4
30
0.3
Power
20
0.2
10
0.1
0
0
0.2
0.4
0.6
0.8
1
Current density (A/cm2)
1.2
Power density (W/cm2)
where Vr0 is the cell reversible voltage at standard conditions (T = 298 K and
p = 1 atm). The efficiency curve is strictly homothetic to the voltage curve. An
efficiency–current curve for a hydrogen–oxygen fuel cell (refer to Figure 14.3)
is shown in Figure 14.4. Figure 14.4 indicates that the efficiency decreases, and
power increases, with the increase of current. Therefore, operating a fuel cell
at its low current, and then at low power, results in high operating efficiency.
However, taking account of the energy consumed by its auxiliaries, such as
0
1.4
FIGURE 14.4 Operating efficiency and power density along with the current density in a
hydrogen–oxygen fuel cell.
440
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
the air circulating pump, the cooling water circulating pump, and so on, very
low power (<10% of its maximum power) results in low operating efficiency,
due to the larger percentage of power consumption in the auxiliary. This will
be discussed in more detail later.
14.3
Fuel and Oxidant Consumption
The fuel and oxidant consumptions in a fuel cell are proportional to the current
drawn from the fuel cell. The chemical reaction in a fuel cell can be generally
described by Equation 14.15, where A is the fuel, B is the oxidant, C and D are
the products:
A + xB B → xC C + xD D.
(14.15)
The mass flow of the fuel, associated with the current drawn from the fuel
cell, can be expressed as
ṁA =
WA I
(kg/s),
1000 nF
(14.16)
where WA is the molecular weight of fuel A, I is the fuel cell current, n is the
electrons transferred in the reaction of Equation 14.15, and F = 96, 495 C/mol
is the Faraday constant. The stoichiometric ratio of oxidant mass flow to fuel
mass flow can be expressed as
ṁB
xB WB
=
,
ṁA
WA
(14.17)
where WB is the molecular weight of oxidant B.
For a hydrogen–oxygen fuel cell (see Table 14.2 for the reaction), the
stoichiometric ratio of hydrogen to oxygen is
ṁH
ṁO
=
stoi
0.5 WO
0.5 × 32
=
= 7.937.
WH
2.016
(14.18)
The equivalent ratio of oxidant to fuel is defined as the ratio of the actual
oxidant/fuel ratio to the stoichiometric ratio, that is,
λ=
(ṁB /ṁA )actual
.
(ṁB /ṁA )stoi
(14.19)
When λ < 1, the reaction is fuel rich; when λ = 1, the reaction is stoichiometric; and when λ > 1, the reaction is fuel lean. In practice, fuel cells are always
operated with λ > 1, that is, excessive air over the stoichiometric value is
441
Fuel Cells
supplied in order to reduce the voltage drop caused by concentration. For
fuel cells using O2 as oxidant, air is usually used rather than pure oxygen. In
this case, the stoichiometric ratio of fuel to air can be expressed as
ṁair
(xO WO ) /0.232
=
,
WA
ṁA
(14.20)
where it is assumed that oxygen mass takes 23.2% of the air mass. For
hydrogen–air fuel cells, Equation 14.19 becomes
14.4
ṁair
ṁH
=
stoi
(0.5 × 32)/0.232
(0.5WO ) /0.232
= 34.21.
=
WH
2.016
(14.21)
Fuel Cell System Characteristics
In practice, fuel cells need auxiliaries to support their operation. The auxiliaries mainly include an air circulating pump, a coolant circulating pump, a
ventilation fan, a fuel supply pump, and electrical control devices as shown
in Figure 14.5. Among the auxiliaries, the air circulating pump is the largest
energy consumer. The power consumed by the air circulating pump (including its drive motor) may take about 10% of the total power output of the fuel
cell stack. The other auxiliaries consumed much less energy compared with
the air circulating pump.
Auxiliaries
1. Air circulating pump
2. Coolant circulating pump
3. Ventilation fan
4. Hydrogen circulating pump
5. Controller
Hydrogen
storage
H2
DC
H2
DC
Ia
+
Ifc
+
IL
Load
Fuel cell stack
–
Air
Water Waster heat
FIGURE 14.5 A hydrogen–air fuel cell system.
–
442
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
In a fuel cell, the air pressure on the electrode surface, p, is usually higher
than the atmospheric pressure, p0 , in order to reduce the voltage drop (see
Equation 14.13). According to thermodynamics, the power needed to compress air from low-pressure p0 to high-pressure p with a mass flow ṁair can
be calculated by1,2
γ
p γ−1/γ
ṁair RT
Pair−comp =
− 1 (W),
(14.22)
γ−1
p0
where γ is the ratio of specific heats of air (γ = 1.4), R is the gas constant of
air (R = 287.1 J/kg K), and T is the temperature at the inlet of the compressor
in K. When calculating the power consumed by the air circulating pump, the
energy losses in the air pump and motor drive must be taken into account.
Thus, the total power consumed is
Pair−cir =
Pair−comp
,
ηap
(14.23)
where ηap is the efficiency of the air pump plus motor drive.
Figure 14.6 shows an example of the operation characteristics of the
hydrogen–air fuel cell system, where λ = 2, p/p0 = 3, and ηap = 0.8, and the
net current and net power are the current and power that flow to the load
(see Figure 14.5). This figure indicates that the optimal operation region of
the fuel cell system is in the middle region of the current range, say, 7–50% of
the maximum current. A large current leads to low efficiency due to the large
voltage drop in the fuel cell stack and, on the other hand, a very small current
leads to low efficiency due to the increase in the percentage of the auxiliaries’
energy consumption.
1.0
Cell voltage, V
0.8
0.6
System efficiency
0.4
Net power density W/cm2
0.2
0
0
0.2
0.4
0.6
0.8
Net current density (A/cm2)
1.0
1.2
FIGURE 14.6 Cell voltage, system efficiency, and net power density varying with net current
density of a hydrogen–air fuel cell.
443
Fuel Cells
14.5
Fuel Cell Technologies
It is possible to distinguish six major types of fuel cells, depending on the
type of their electrolyte.4,5 They are proton exchange membrane (PEM) or
polymer exchange membrane fuel cells (PEMFCs), alkaline fuel cells (AFCs),
phosphoric acid fuel cells (PAFCs), molten carbonate fuel cells (MCFCs), solid
oxide fuel cells (SOFCs), and direct methanol fuel cells (DMFCs). Table 14.3
lists their normal operation temperature and the state of electrolyte.
14.5.1
Proton Exchange Membrane Fuel Cells
PEMFCs use solid polymer membranes as the electrolyte. The polymer
membrane is perfluorosulfonic acid, which is also referred to as Nafion
(Dupont® ). This polymer membrane is acidic; therefore the ions transported
are hydrogen ions (H+ ) or protons. The PEMFC is fueled with pure hydrogen
and oxygen or air as oxidant.
The polymer electrolyte membrane is coated with a carbon-supported catalyst. The catalyst is directly in contact with both the diffusion layer and
the electrolyte for a maximized interface. The catalyst constitutes the electrode. Directly on the catalyst layer is the diffusion layer. The assembly of
the electrolyte, catalyst layers, and gas diffusion layers is referred to as the
membrane–electrode assembly.
The catalyst is a critical issue in PEMFCs. In early realizations, very high
loadings of platinum were required for the fuel cell to operate properly.
Tremendous improvements in catalyst technology have made it possible to
reduce the loading from 28 mg to 0.2 mg/cm2 . Because of the low operating
temperature of the fuel cell and the acidic nature of the electrolyte, noble
metals are required for the catalyst layer. The cathode is the most critical
electrode because the catalytic reduction of oxygen is more difficult than the
catalytic oxidation of hydrogen.
TABLE 14.3
Operating Data of Various Fuel Cell Systems4,5
Operating
Temperature (◦ C)
Electrolyte
PEMFCs
AFCs
PAFCs
MCFCs
SOFCs
60–100
100
60–200
500–800
1000–1200
Solid
Liquid
Liquid
Liquid
Solid
DMFCs
100
Solid
Cell System
444
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Another critical issue in PEMFCs is water management. In order to operate
properly, the polymer membrane needs to be kept humid. Indeed, the conduction of ions in polymer membranes requires humidity. If the membrane is
too dry, there will not be enough acid ions to carry the protons. If it is too wet
(flooded), the pores of the diffusion layer will be blocked and reactant gases
will not be able to reach the catalyst.
In PEMFCs, water is formed on the cathode. It can be removed by keeping
the fuel cell at a certain temperature and flowing enough to evaporate the
water and carry it out of the fuel cell as a vapor. However, this approach is
difficult because the margin of error is narrow. Some fuel cell stacks run on a
large excess of air that would normally dry the fuel cell, and use an external
humidifier to supply water by the anode.
The last major critical issue in PEMFCs is poisoning. The platinum catalyst
is extremely active and thus provides great performance. The trade-off of
this great activity is a greater affinity for carbon monoxide (CO) and sulfur
products than oxygen. The poisons bind strongly to the catalyst and prevent
hydrogen or oxygen from reaching it. The electrode reactions cannot take
place on the poisoned sites and the fuel cell performance is diminished. If
hydrogen is fed from a reformer, the stream will contain some carbon monoxide. The carbon monoxide may also enter the fuel cell in the air stream if
the air is pumped from the atmosphere of a polluted city. The poisoning by
carbon monoxide is reversible, but comes at a cost and requires the individual
treatment of each cell.
The first PEMFCs were developed in the 1960s for the needs of the U.S.manned space program. It is nowadays the most investigated fuel cell
technology for automotive applications by manufacturers such as Ballard. It
is operated at 60–100◦ C and can offer a power density of 0.35–0.6 W/cm2 .
The PEMFC has some definite advantages in its favor for EV and HEV
applications.6 First, its low-temperature operation and hence its fast startup are desirable for an EV and HEV. Second, the power density is the highest
among all the available types of fuel cells. The higher the power density, the
smaller the size of the fuel cell that needs to be installed for the desired power
demand. Third, its solid electrolyte does not change, move, or vaporize from
the cell. Finally, since the only liquid in the cell is water, the possibility of any
corrosion is essentially delimited. However, it also has some disadvantages,
such as the expensive noble metal needed, expensive membrane, and easily
poisoned catalyst and membrane.7
14.5.2 Alkaline Fuel Cells
AFCs use an aqueous solution of potassium hydroxide (KOH) as the electrolyte to conduct ions between electrodes. Potassium hydroxide is alkaline.
Because the electrolyte is alkaline, the ion conduction mechanism is different
from PEMFCs. The ion carried by the alkaline electrolyte is a hydroxide ion
445
Fuel Cells
(OH− ). This affects several other aspects of the fuel cell. The half reactions are
as follows
Anode:
Cathode:
2H2 + 4OH− → 4H2 O + 4e− .
O2 + 4e− + 2H2 O → 4OH− .
Unlike in acidic fuel cells, water is formed on the hydrogen electrode. In
addition, water is needed at the cathode by oxygen reduction. Water management becomes an issue that is sometimes resolved by making the electrodes
waterproof and keeping the water in the electrolyte. The cathode reaction
consumes water from the electrolyte where as the anode reaction rejects its
product water. The excess water (2◦ mol per reaction) is evaporated outside
the stack.
AFCs are capable of operating over a wide range of temperatures (from
80◦ C to 230◦ C) pressures3 (from 2.2 to 45 atm). High-temperature AFCs also
make use of a highly concentrated electrolyte, so highly concentrated that the
ion transport mechanism changes from aqueous solution to molten salt.
AFCs are capable of achieving very high efficiencies because of the fast
kinetics allowed by the hydroxide electrolyte. The oxygen reaction (O2 →
OH− ) in particular is easier than the oxygen reduction in acidic fuel cells. As
a result, the activation losses are very low. The fast kinetics in AFCs allows
using silver or nickel as catalysts instead of platinum. The cost of the fuel cell
stack is thus greatly reduced.
AFC kinetics is further improved by the eventual circulation of the electrolyte. When the electrolyte is circulated, the fuel cell is said to be a “mobile
electrolyte fuel cell.” The advantages of such an architecture are as follows:
a easy thermal management because the electrolyte is used as coolant; more
homogeneous electrolyte concentration, which solves problems of concentration around the cathode; the possibility of using the electrolyte for water
management; the possibility of replacing the electrolyte if it has been too
polluted by carbon dioxide; and, finally, the possibility of removing the electrolyte from the fuel cell when it is turned off, which has the potential to
greatly lengthen the lifetime of the stack.
The use of a circulated electrolyte, however, poses some problems. The
greatest problem is the increased risk of leakage: potassium hydroxide is
highly corrosive and has a natural tendency to leak even through the tightest
seals. The construction of the circulation pump and heat exchanger and eventual evaporator is further complicated. Another problem is the risk of internal
electrolytic short-circuit between two cells if the electrolyte is circulated too
violently or if the cells are not isolated enough. A circulating electrolyte AFC
is pictured in Figure 14.7.8
The greatest problem with AFCs is the poisoning by carbon dioxide. The
alkaline electrolyte has great affinity for carbon dioxide and together they
form carbonate ions (CO2−
3 ). These ions do not participate in the fuel cell
reaction and diminish its performance. There is also a risk that the carbonate
446
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Output
H2
+
KOH
KOH
–
B1
B2
Fuel cell stack
Air
H2
F1
F2
G1
G2
C1
H2
F3
C2
H2 O
E
Air
H2O
D
H2
KOH Air
Electrolyte (KOH)
FIGURE 14.7 Circulating electrolyte and supplies of hydrogen and air in an AFC; B1 , B2 , heater
exchangers; C1 , C2 , condensers; D, pampers; E, motor; F1 , F2 , F3, controls; and G1 , G2 , outlets.8
will precipitate and obstruct the electrodes. This last issue may be taken care
of by circulating the electrolyte. The solution, which adds to the cost and
complexity, is to use a carbon dioxide scrubber that will remove the gas from
the air stream.
The advantages of AFCs are that they require cheap catalyst, cheap electrolyte, and high-efficiency and low-temperature operation. However, they
also have some disadvantages such as impaired durability due to corrosive
electrolyte, water produced on the fuel electrode, and poisoning by carbon
dioxide.
14.5.3
Phosphoric Acid Fuel Cells8
PAFCs rely on an acidic electrolyte, like PEMFCs, to conduct hydrogen ions.
The anode and cathode reactions are the same as the PEMFC reactions.
Phosphoric acid (H3 PO4 ) is a viscous liquid that is contained by capillarity
in the fuel cell in a porous silicon carbide matrix.
447
Fuel Cells
PAFC was the first fuel cell technology to be marketed. Many hospitals,
hotels, and military bases make use of a PAFC to cover part or the totality
of their electricity and heat requirements. Very little work has been done to
apply this technology to vehicles, probably because of temperature problems.
The phosphoric acid electrolyte temperature must be kept above 42◦ C,
which is its freezing point. Freezing and rethawing the acid unacceptably
stresses the stack. Keeping the stack above this temperature requires extra
hardware, which adds to the cost, complexity, weight, and volume. Most of
these issues are minor in the case of a stationary application, but are incompatible with a vehicular application. Another problem arising from the high
operating temperature (above 150◦ C) is the energy consumption associated
with warming up the stack. Every time the fuel cell is started, some energy
(i.e., fuel) must be spent to heat it up to operating temperature, and every
time the fuel cell is turned off, the heat (i.e., energy) is wasted. The loss is
significant for short travel times, which usually occurs during city driving.
However, this issue seems to be minor in the case of mass transportation such
as buses.
The advantages of PAFCs are that they require a cheap electrolyte, a low
operating temperature, and a reasonable start-up time. The disadvantages
are the expensive catalyst (platinum), corrosion by acidic electrolyte, CO
poisoning, and low efficiency.
14.5.4
Molten Carbonate Fuel Cells
MCFCs are high-temperature fuel cells (500–800◦ C). They rely on a molten
carbonate salt to conduct ions, usually lithium–potassium carbonate or
lithium–sodium carbonate. The ions conducted are carbonate ions (CO2−
3 ).
The ion conduction mechanism is that of a molten salt like in PAFCs or highly
concentrated AFCs.
The electrode reactions are different from other fuel cells:
Anode:
Cathode:
−
H2 + CO2−
3 → H2 O + CO2 + 2e .
1
O2 + CO2 + 2e− → CO2−
3 .
2
The major difference from other fuel cells is the necessity of providing carbon dioxide at the cathode. It is not necessary to have an external source
since it can be recycled from the anode. MCFCs are never used with pure
hydrogen but rather with hydrocarbons. Indeed, the major advantage of
high-temperature fuel cells is their capability to, almost, directly process
hydrocarbon fuels because the high temperature allows their decomposition
to hydrogen on the electrodes. This would be a tremendous advantage for
automotive applications because of the present availability of hydrocarbon
fuels. In addition, the high temperatures enhance the kinetics to the point that
cheap catalysts may be used.
448
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
MCFCs, however, pose many problems because of the nature of the electrolyte and the operating temperature required. The carbonate is an alkali
and is extremely corrosive, especially at high temperature. Not only is this
unsafe, but there is also the problem of corrosion on the electrodes. It is unsafe
to have a large device at 500–800◦ C under the hood of a vehicle. While it is
true that temperatures in internal combustion engines do reach above 1000◦ C,
these temperatures are restricted to the gases themselves and most parts of
the engine are kept cool (around 100◦ C) by the cooling system. The fuel consumption associated with heating up the fuel cell is also a problem, worsened
by the very high operating temperature and the latent heat necessary to melt
the electrolyte. These problems are likely to confine MCFCs to stationary or
steady power applications such as ships.
The major advantages of MCFCs are that they are fueled with hydrocarbon fuels, require a low-cost catalyst, have improved efficiency due to fast
kinetics, and have a lower sensitivity to poisoning. The major disadvantages
are slow start-up and reduced material choice due to high temperature, a
complex fuel cell system due to CO2 cycling, corrosive electrolyte, and slow
power response.
14.5.5
Solid Oxide Fuel Cells
SOFCs conduct ions in a ceramic membrane at high temperature (1000–
1200◦ C). Usually, the ceramic is a yttrium-stabilized zirconia (YSZ) that will
conduct oxygen ions (O2− ), but other ceramics conduct hydrogen ions. The
conduction mechanism is similar to that observed in semiconductors, often
called solid-state devices. The name of the fuel cell is derived from this
similarity. The half reactions are as follows:
Anode:
Cathode:
H2 + O2− → H2 O + 2e− .
1
O2 + 2e− → O2− .
2
Water is produced at the fuel electrode. The greatest advantage of SOFCs
is this static electrolyte. There is no moving part, except perhaps in the ancillaries. The very high operating temperature allows the use of hydrocarbon
fuels as in MCFCs. It should also be noted that SOFCs are not poisoned by
carbon monoxide and that they process it about as efficiently as hydrogen.
The anode reaction is then
CO + O2− → CO2 + 2e− .
SOFCs also benefit from reduced activation losses due to their high operating temperature. The losses are dominated by the ohmic component. SOFCs
may be of two kinds: planar and tubular. The planar type is a bipolar
449
Fuel Cells
Interconnection
Electrolyte
Fuel
flow
Air
electrode
Air
flow
Fuel electrode
FIGURE 14.8 Tubular SOFC.
stack similar to other fuel cell technologies. A tubular SOFC is described
in Figure 14.8. The major advantages of tubular technologies include easier sealing and reduced constraints on the ceramics. Disadvantages include
lower efficiency and power density.
Like MCFCs, the disadvantages of SOFCs are mostly associated with their
high operating temperature (safety and fuel economy). Supplementary problems arise because the ceramic electrolyte and electrodes are extremely brittle.
This is a major disadvantage for vehicular applications where vibrations are
a common occurrence. Thermal cycling further stresses the ceramics and is a
major concern for planar fuel cells.
14.5.6
Direct Methanol Fuel Cells
Instead of using hydrogen, methanol can be directly used as the fuel for a
fuel cell; this is the so-called DMFC. There are some definite motivations for
applying DMFCs to vehicles. First, methanol is a liquid fuel that can be easily
stored, distributed, and marketed for vehicle application; hence the current
infrastructure of fuel supply can be used without much further investment.
Second, methanol is the simplest organic fuel that can be most economically
and efficiently produced on a large scale from relatively abundant fossil fuel,
namely coal and natural gas. Furthermore, methanol can be produced from
agriculture products, such as sugar cane.9
In the DMFCs, both the anode and cathode adopt platinum or platinum
alloys as electro catalyst. The electrolyte can be trifluoromethane sulfonic
acid or PEM. The chemical reaction in a DMFC is given below:
Anode:
Cathode:
Overall:
CH3 OH + H2 O → CO2 + 6H+ + 6e− .
3
O2 + 6H+ + 6e− → 3H2 O.
2
3
CH3 OH + O2 → CO2 + 2H2 O.
2
450
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
The DMFC is relatively immature among the aforementioned fuel cells. At
the present status of DMFC technology, it generally operates at 50–100◦ C.
Compared with direct hydrogen fuel cells, DMFCs have low power density,
slow power response, and low efficiency.9–11
14.6
Fuel Supply
Fuel supply to the on-board fuel cells is the major challenge for fuel cell
vehicle applications. As mentioned before, hydrogen is the ideal fuel for fuelcell-powered vehicles.4,5 Hence, hydrogen production and storage on-board
are the major concern. Generally, there are two ways of supplying hydrogen
to fuel cells. One is to produce hydrogen in ground stations and store pure
hydrogen on-board. The other is to produce hydrogen on-board from an easycarrying hydrogen carrier and directly feed the fuel cells.
14.6.1
Hydrogen Storage
So far, there are three methods for storage of hydrogen on-board: compressed
hydrogen in a container at ambient temperature, cryogenic liquid hydrogen
at low temperature, and the metal hydride method. All these methods have
their advantages and disadvantages.
14.6.1.1
Compressed Hydrogen
Pure hydrogen may be stored on-board the vehicle under pressure in a tank.
The ideal gas equation can be used to calculate the mass of hydrogen stored
in a container with volume V and pressure p, that is,
mH =
pV
WH ,
RT
(14.24)
where p and V are the pressure and volume of the container, R is the gas
constant (8.31 J/mol K), T is the absolute temperature, and WH is the molecular weight of hydrogen (2.016 g/mol). The energy stored in hydrogen can be
calculated as
EH = mH HV,
(14.25)
where HV is the heating value of hydrogen. The heating value is either the
high heating value (HHVH = 144 mJ/kg) or the low heating value (LHVH =
120 mJ/kg), depending on the condensation energy of produced water. For
a convenient comparison with internal combustion engines, LHVH is most
often used.
451
Fuel Cells
2.0
0
1450
2900
Pressure (psi)
4350 5800
7250
8700
10150
1.8
Energy per liter
hydrogen
(kWh)
1.6
1.4
1.2
Energy needed to
compress hydrogen
(kWh)
Equivalent liters
of gasoline
Mass of hydrogen
per liter, kg
1.0
0.8
0.6
0.4
0.2
0
0
100
200
300
400
500
Pressure (100 kPa)
600
700
FIGURE 14.9 Energy per liter of hydrogen and equivalent liters of gasoline versus pressure.
Figure 14.9 shows the mass and energy in 1 L of hydrogen and the equivalent
liters of gasoline under different pressure and at room temperature (25◦ C).
The equivalent liters of gasoline are defined as the number of liters of gasoline
in which the same amount of energy is contained as that in 1 L of hydrogen.
Figure 14.9 also indicates that at a pressure of 350 bar, the energy per liter of
hydrogen is less than 1 kWh and is equivalent to about 0.1 L of gasoline. Even
if the pressure is increased to 700 bar, which is believed to be the maximum
pressure that can be reached, the energy per liter of hydrogen is still less than
2.0 kWh and is equivalent to about 0.2 L of gasoline.
In addition, a certain amount of energy is needed to compress hydrogen
from low pressure to high pressure. The process in hydrogen compression
may be assumed to be an adiabatic process, that is, no heat exchange occurs
during the process. The energy consumed can be expressed as
Ecomp
m
γ
=
RT
γ − 1 WH
p
p0
γ−1/γ
−1 ,
(14.26)
where m is the mass of hydrogen, WH is the molecular weight of hydrogen, γ
is the ratio of the specific heat (γ = 1.4), p is the pressure of hydrogen, and p0 is
atmospheric pressure. This energy consumption is also drawn in Figure 14.9.
It shows that about 20% of hydrogen energy must be consumed to compress
it to high pressure. Taking into account the inefficiency of the compressor
and the electric motor, it is estimated that about 25% of hydrogen energy is
consumed.
To contain a gas at several hundred atmospheres requires a very strong
tank. In order to keep the weight as low as possible and the volume reasonable,
452
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
today’s hydrogen tanks for automotive applications use composite materials
such as carbon fiber. The cost of a compressed hydrogen tank is thus likely to
be high.
The hazards of compressed hydrogen on-board a vehicle must be taken
into account. Besides the risk of leakages through cracks in the tank walls,
seals, and so on, there is the problem of permeation of hydrogen through the
material of the wall. The hydrogen molecule is so small that it can diffuse
through some materials.
In addition, a compressed gas tank is a potential bomb in case of a crash.
The dangers are even greater in the case of hydrogen, which has a very wide
explosive range in air from 4% to 77%12 and is capable of mixing very quickly
with air. This is to be compared with gasoline, which has an explosive range
from only 1% to 6% and is a liquid. It should be noted that hydrogen has
a high autoignition temperature of 571◦ C, whereas gasoline autoignites at
around 220◦ C but must be vaporized first.
So far, the technology of compressed hydrogen storage on-board is still a
huge challenge for vehicle application.
14.6.1.2
Cryogenic Liquid Hydrogen
Another alternative solution to storing hydrogen on-board a vehicle is to liquefy the gas at cryogenic temperatures (−259.2◦ C). The thus stored hydrogen
is commonly referred to as “LH2 .” LH2 storage is affected by the same density problems that affect compressed hydrogen. Indeed, the density of liquid
hydrogen is very low and 1 L of liquid hydrogen only weighs 71 × 10−3 kg.
This low density results in an energy content of about 8.52 mJ/L of liquid
hydrogen.
Containing a liquid at such a low temperature as −259.2◦ C is technically
challenging. It requires a heavily insulated tank to minimize the heat transfer
from the ambient air to the cryogenic liquid and thus prevent it from boiling.
The approach usually taken is to build a significantly insulated tank and to
make it strong enough to withstand some of the pressure resulting from the
boil-off. The excess pressure is then released to the atmosphere by means
of a safety valve. The tank insulation, strength, and safety devices also add
significantly to the weight and cost of LH2 storage.
The boil-off is a problematic phenomenon: if the vehicle is parked in a closed
area (garage, underground parking) there is the risk that hydrogen will build
up in the confined atmosphere and that the explosive mixture thus formed
will explode at the first spark (light switch, lighter, etc.). The refueling of a
tank with liquid hydrogen requires specific precautions: air must be kept out
of the circuit. The commonly used method is to fill the tank with nitrogen prior
to fueling in order to evacuate the residual gas in the tank. It is also necessary to
use specialized equipment, designed to handle the explosion hazard and the
cryogenic hazards. Indeed, a cryogenic liquid is a dangerous compound for
living beings, as it burn-freezes the skin and organs. It may well be, however,
453
Fuel Cells
that the ambient temperature would evaporate the cryogenic hydrogen fast
enough to limit or eliminate this risk.
14.6.1.3
Metal Hydrides
Some metals are capable of combining with hydrogen to form stable compounds that can later be decomposed under certain pressure and temperature
conditions. These metals may be iron, titanium, manganese, nickel, lithium,
and some alloys of these metals. Metal hydrides are stable under normal temperature and pressure conditions and are capable of releasing hydrogen only
when required.
The hydrogen storage metals and metal alloys are Mg, Mg2 Ni, FeTi, and
LaNi5 . These metals and metal alloys absorb hydrogen to form Mg–H2 ,
Mg2 Ni–H4 , FeTi–H2 , and LaNi5 –H6 . Theoretically, metal and metal alloys
store hydrogen at a higher density than pure hydrogen, as shown in Table 14.4.
In practice, the hydrogen storage capacity depends heavily on the surface area
of the material, on which the hydrogen molecules are absorbed. A large surface area per unit weight of material can be obtained by fine porous modules
made of finely ground powder of the metals or metal alloys. Figure 14.10
shows the practical mass and volume needed to store 6 kg of hydrogen (22 L
of gasoline equivalent). This figure indicates that Mg–H2 is the promising
technology.
Alkaline metal hydrides are possible alternatives to metal hydride absorption. These hydrides react violently with water to release hydrogen and a
hydroxide. The example of sodium hydride is shown below:
NaH + H2 O → NaOH + H2 .
The major disadvantage is the necessity of carrying a highly reactive hydride
and a corrosive solution of hydroxide in the same vehicle. The storage density
is good enough in comparison to many other hydrogen storage techniques,
TABLE 14.4
Theoretical Hydrogen Storage Densities in Compressed, Liquid, and Metal
Hydride Approaches12
H-Atoms per cm3 (×1022 )
% of Weight that is Hydrogen
H2 gas, 200 bar (2900 psi)
H2 liquid, 20◦ K (−253◦ C)
H2 liquid, 4.2◦ K (−269◦ C)
Mg–H2
0.99
100
4.2
5.3
100
100
6.5
7.6
Mg2 Ni–H2
5.9
6.0
3.6
1.89
5.5
1.37
Material
FeTi–H2
LaNi–H6
454
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
H2 gas
Liquid H2
Mg–H2
Mg2Ni–H4
FeTi–H2
LaNi5–H6
395 kg weight
375 liters volume
140 kg
86 liters
175 kg
73 liters
315 kg
83 liters
435 kg
80 liters
440 kg
64 liters
FIGURE 14.10 Current mass and volume needed to store 6 kg of hydrogen (22 L of gasoline
equivalent) in various hydrogen storage devices.14
but falls short in comparison to gasoline. The manufacture of these hydrides
and their recycling are also challenging.
Carbon nanotubes, discovered in 1991, would be a prospective method for
hydrogen storage systems, because of their potential high hydrogen absorbing capability and light weight. However, carbon nanotube technology is still
in its infancy and has a long way to go before its practical utility can be
assessed.
14.6.2
Hydrogen Production
At present, hydrogen is mostly produced from hydrocarbon fuels through
reforming. Reforming is a chemical reaction that extracts hydrogen from
hydrocarbons. During this reaction, the energy content of the fuel is transferred from the carbon–hydrogen bonds to the hydrogen gas. Hydrocarbons
such as gasoline, methane, or methanol are the most likely candidates because
of their ease of reforming.
There are three major methods of reforming: steam reforming (SR),
autothermal reforming (ATR), and partial oxidation (POX) reforming. SR
may be used with methanol, methane, or gasoline, whereas ATR and POX
reforming are the most commonly used for processing gasoline.
14.6.2.1
Steam Reforming
SR is a chemical process in which hydrogen is produced through the chemical
reaction between hydrocarbon fuels and water steam at high temperature.
455
Fuel Cells
The following chemical equations describe the reforming, using methane
(CH4 ), methanol (CH3 OH), and gasoline (iso-octane C8 H18 ) as the fuels:
CH4
◦
ΔH −79.4
+
ΔH
ΔH
◦
258 kJ/mol CH4 → 4H2
2 × (−286.2)
+
CH3 OH
◦
+
2H2 O
−238.7
H2 O
+
+
CO2
−393.8
0
131 kJ/mol CH3 OH → 3H2
−286.2
+
CO2
−393.8
0
C8 H18 + 16H2 O + 1652.9 kJ/mol C8 H18 → 25H2 + 8CO2
−224.1
16 × (−286.2)
8 × (−393.8)
0
The above reactions are highly endothermic and need to be powered by some
burning of fuels. Also, these reactions yield some carbon monoxide (CO) in
the products, which is a poison to some electrolytes such as PEMFCs, AFCs,
and PAFCs. The carbon monoxide can be further converted into hydrogen
and carbon dioxide by means of a water–gas shift reaction:
+
CO
H2 O
ΔH ◦ − 111.6
+
4 kJ/mol CO → H2
−286.2
+
CO2
−393.8
0
In SR, it is particularly preferred to use methanol as the fuel, since there is no
theoretical need for a water–gas shift reaction and since the processing temperature is low (250◦ C). The hydrogen yield is also particularly high. Among
its disadvantages, the most significant are the poisoning of the reformer catalysts by impurities in methanol and the need for external heat input to the
endothermic reaction. The heat requirements slow the reaction down and
impose a slow start-up time between 30 and 45 min.13 The methanol steam
reformer also has slow output dynamics. Although feasible, the SR of gasoline
is not commonly used.
14.6.2.2
POX Reforming
POX reforming combines fuel with oxygen to produce hydrogen and carbon
monoxide. This approach generally uses air as the oxidant, and results in a
reformate that is diluted with nitrogen. Then the carbon monoxide further
reacts with water steam to yield hydrogen and carbon dioxide (CO2 ), as mentioned above. POX reforming usually uses gasoline (iso-octane) as its fuel.
The reaction is expressed as
C8 H18 + 4O2 + 16N2 → 8CO
ΔH ◦ − 224.1
8CO
0
+
ΔH ◦ 8 × (−111.6)
0
8H2 O
+
8 × (−111.6)
+
8 × (−286.2)
9H2 + 16N2 + 668.7 kJ/mol C8 H18
0
0
32 kJ/mol 8CO → 8H2
0
+
8CO2
8 × (−393.8)
456
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Exhaust
Air
Air
Air
Air
Fuel
Fuel
processing
assembly
(FPA)
Fuel
Steam
Tail gas
combustor
(TGC)
Fuel cell
stack
POx
Water
recycle
Water
recycle
FIGURE 14.11 Fuel processing diagram.13
POX reforming is highly exothermic, which thus has the advantages of very
fast response to transients and being capable of very fast start-ups. POX
reformers are also fuel flexible, and are capable of treating a wide variety of
fuels. The disadvantages include a high operating temperature (800–1000◦ C)
and a difficult construction due to heat-integration problems between the
different steps of the reaction.13 In addition, it can be seen from the above
chemical equation that the heat produced from the first reaction is much
more than that absorbed in the second reaction, and hence POX reforming is
somewhat less efficient than the SR of methanol.
Figure 14.11 shows a fuel processing system developed by Epyx Cooperation.13
14.6.2.3 Autothermal Reforming
ATR combines fuel with both water and steam so that the exothermic heat
from the POX reaction is balanced by the endothermic heat of the SR reaction.
The chemical equation in this reaction is
C8 H18
ΔH
◦
−224.1
+
nO2
+
(8 − 2n)H2 O
(8 − 2n) × (−286.2)
→
8CO
+
(17 − 2n)H2
8 × (−111.6)
Zero heat produced in the above equation yields n = 2.83. The CO produced
in the above reaction can further react with water steam to produce hydrogen
by the water-shift reaction mentioned above.
ATR yields a more concentrated hydrogen stream than POX reforming, but
less than SR. The heat integration is easier than for POX reforming but a
catalyst is required. ATR is potentially more efficient than POX reforming.
457
Fuel Cells
14.6.3 Ammonia as Hydrogen Carrier
Ammonia is a non-carbon-based chemical that presents interesting characteristics as a hydrogen source. The extraction of hydrogen from ammonia, called
“cracking,” is shown below:
2NH3 → N2 + 3H2
The above reaction is easily achieved by heating ammonia, either alone
or over a catalyst bed, which has the advantage of lowering the reaction
temperature. The energy requirement for this reaction is minimal because it
is reversible. Ammonia presents great advantages in terms of storage as it is
easily liquefied at low pressure (about 10 atm) or at mildly low temperatures
(−33◦ C). Other advantages include a very high autoignition temperature
(651◦ C) and a limited explosive range in air (15–28%).
Despite its many advantages, ammonia has a major disadvantage: it is
toxic. Ammonia is an alkali that has an extreme affinity for water; hence it
strongly attacks the eyes and lungs and causes severe burns. This causticity
makes it challenging to conceive of ammonia as a fuel-for fuel cell-powered
automobiles.
14.7
Non-Hydrogen Fuel Cells
Some fuel cell technologies can directly process fuels other than hydrogen.4,5
Some likely couples are listed below:
•
Direct methanol PEMFCs
• Ammonia AFCs
•
Direct hydrocarbon MCFCs or SOFCs
Like their hydrogen counterparts, direct methanol PEMFCs are actively
studied and present many advantages such as the absence of a reformer, the
handling of a liquid fuel, and the absence of high temperatures in the system.
The major disadvantages are the necessity of diluting the methanol in liquid
water to feed the fuel electrode and a strong crossover of methanol—due to
its absorption in the polymer membrane, but due mostly to its slow kinetics.
Ammonia AFCs14 are possible alternatives to the thermal cracking of
ammonia. Ammonia gas is directly fed to the fuel cell and is catalytically
cracked on the anode. The ammonia fuel cell reaction yields a slightly lower
thermodynamic voltage and higher activation losses than hydrogen AFCs.
The activation losses may be reduced by improving the catalyst layer. Interestingly, it would be possible to use ammonia directly with other fuel cell
458
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
technologies if it were not for the fact that the acidic nature of their electrolyte
would be destroyed by the alkaline ammonia.
MCFCs and SOFCs have the capability of directly cracking hydrocarbons
because of their high operating temperature. Therefore, they are not directly
consuming the hydrocarbons, but are internally extracting the hydrogen from
them. This option obviously has all the disadvantages of high-temperature
fuel cells as discussed in the section on fuel cell technologies.
References
1. J. Bevan Ott and J. Boerio-Goates, Chemical Thermodynamics—Advanced Applications, Academic Press, New York, ISBN 0-12-530985-6, 2000.
2. S. I. Sandler, Chemical and Engineering Thermodynamics, Third Edition, John
Wiley & Sons, New York, ISBN 0-471-18210-9, 1999.
3. H. K. Messerle, Energy Conversion Statics, Academic Press, New York, 1969.
4. P. J. Berlowitz and C. P. Darnell, “Fuel choices for fuel cell powered vehicles,”
Society of Automotive Engineers (SAE) Journal, Paper No. 2000-01-0003, Warrendale,
PA, 2002.
5. P. J. Berlowitz and C. P. Darnell “Fuel choices for fuel cell powered vehicles,”
Society of Automotive Engineers (SAE) Journal, Paper No. 2001-01-0003, Warrendale,
PA, 2002.
6. F. Michalak, J. Beretta, and J.-P. Lisse, “Second generation proton exchange membrane fuel cell working with hydrogen stored at high pressure for fuel cell electric
vehicle,” Society of Automotive Engineers (SAE) Journal, Paper No. 2002-01-0408,
Warrendale, PA, 2002.
7. J. Larminie and A. Dicks, Fuel Cell Systems Explained, John Wiley & Sons,
New York, 2000.
8. W. Vielstich, Fuel Cells—Modern Processes for Electrochemical Production of Energy,
John Wiley & Sons, New York, ISBN 0-471-90695-6, 1970.
9. R. M. Moore, “Direct methanol fuel cells for automotive power system,” in Fuel
Cell Technology for Vehicles, R. Stobart (ed.), Society of Automotive Engineers (SAE),
ISBN: 0-7680-0784-4, Warrendale, PA, 2001.
10. N. Q. Minh and T. Takahashi, Science and Technology of Ceramic Fuel Cells, Elsevier,
Amsterdam, 1995.
11. M. Baldauf and W. Preidel, “Status of the development of a direct methanol fuel
cell,” in Fuel Cell Technology for Vehicles, R. Stobart (ed.), Society of Automotive
Engineers (SAE), ISBN: 0-7680-0784-4, Warrendale, PA, 2001.
12. S. E. Gay, J. Y. Routex, M. Ehsani, and M. Holtzapple, “Investigation of hydrogen
carriers for fuel cell based transportation,” Society of Automotive Engineers (SAE)
Journal, Paper No. 2002-01-0097, Warrendale, PA, 2002.
13. “Hydrogen at GKSS: Storage Alternative,” available at http://www.gkss.de/, last
visited in May 2003.
14. C. E. Thomas, B. D. James, F. D. Lomax Jr, and I. F. Kuhn Jr, “Societal impacts of
fuel options for fuel cell vehicles,” Society of Automotive Engineers (SAE) Journal,
Paper No. 982496, Warrendale, PA, 2002.
15
Fuel Cell Hybrid Electric Drive
Train Design
Fuel cells, as discussed in Chapter 14, are considered to be one of the advanced
power sources for applications in transportation. Compared with the IC
engines, fuel cells have the advantages of high energy efficiency and much
lower emissions. This is because they directly convert the free energy in fuel
into electrical energy, without undergoing combustion. However, vehicles
powered solely by fuel cells have some disadvantages, such as a heavy and
bulky power unit caused by the low power density of the fuel cell system,
long start-up time, and slow power response. Furthermore, in propulsion
applications, the extremely large power output at sharp acceleration and the
extremely low power output at low-speed driving lead to low efficiency, as
shown in Figure 15.1.
Hybridization of the fuel cell system with a peaking power source is an effective technology to overcome the disadvantages of the fuel-cell-alone-powered
vehicles. The fuel cell HEV is totally different from the conventional IC enginepowered vehicles and IC engine-based hybrid drive trains. Therefore, a totally
new design methodology is necessary.1 In this chapter, a general systematic
design methodology and a control strategy for the fuel cell hybrid electric
drive trains are discussed. Along with the discussion, a design example for a
passenger car drive train is introduced.
15.1
Configuration
The fuel-cell-powered hybrid drive train has the construction as shown in
Figure 15.2.1,2 It mainly consists of a fuel cell system as the primary power
source, a PPS, an electric motor drive (motor and its controller), a vehicle controller, and an electronic interface between the fuel cell system and the PPS.1
According to the power or torque command received from the accelerator or
the brake pedal and other operating signals, the vehicle controller controls
the motor power (torque) output and the energy flows between the fuel cell
system, the PPS, and the drive train. For peak power demand, for instance,
459
460
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
1.0
Cell voltage, V
0.8
System efficiency
0.6
0.4
Net power density, W/cm2
0.2
0
0
0.2
0.4
Optimal
operating region
0.6
0.8
1.0
Net current density,
1.2
A/cm2
FIGURE 15.1 Typical operating characteristics of a fuel cell system.
in sharp acceleration, both the fuel cell system and the PPS supply propulsion power to the electric motor drive. In braking, the electric motor, working
as a generator, converts part of the braking energy into electric energy and
stores it in the PPS. The PPS can also restore the energy coming from the fuel
cell system when the load power is less than the rated power of the fuel cell
Accelerator
pedal
Traction
command
signal
Brake
pedal
Brake
command
signal
Speed signal
FC
system
Electric
control
signal
Fuel cell
power signal
Peaking power signal
Vehicle
Motor control signal
controller
Electronic
interface
Peaking
power
source
Wheel
Motor
controller
Elec.
motor
Transmission
Wheel
FIGURE 15.2 Configuration of a typical fuel cell hybrid drive train.
461
Fuel Cell Hybrid Electric Drive Train Design
system. Thus, with proper design and control strategy, the PPS will never
need to be charged from outside of the vehicle.
15.2
Control Strategy
The control strategy that is preset in the vehicle controller is to control the
power flow between the fuel cell system, the PPS, and the drive train. The
control strategy should ensure the following:
1. The power output of the electric motor always meets the power
demand.
2. The energy level in the PPS is always maintained within its optimal
region.
3. The fuel cell system operates within its optimal operating region.
The driver gives a traction command or brake command through the
accelerator pedal or brake pedal (refer to Figure 15.3), which is represented
Pcomm
Pfc-rated
Pfc
Pfc-min
Ppps-traction
Ppps-charging
E
Emin
Emax
Traction
power
command
Braking
power
command
Ptr
Pb
Pcomm
If Pcomm < 0
yes
brake
--Commanded power
--Rated power of the fuel cell system
--Power of the fuel cell system
--Minimum power of the fuel cell system
--Traction power drawn from the PPS
--Charging power in to the PPS
--Energy level of the PPS
--Bottom line of the energy storage in the PPS
--Top line of the energy storage in the PPS
Traction No
If Pcomm > Pfc-rated
PPS charging
or traction
yes
Hybrid traction
Ppps-traction = Pcomm – Pfc
No
If Pcomm ≤ Pfc-rated
yes
If E < Emin
PPS
charging
No
yes
If E > Emax
PPS
charging
Pfc = Pfc-rated
No
FC
traction
No
yes
Pfc = Pfc-rated
Ppps-charging = Pfc – Pcomm
Pfc = Pfc-rated
Pfc = Pcomm
Ppps-charging = Pfc-Pcomm
Ppps = 0
FIGURE 15.3 Flowchart of the control strategy.
Pfc = 0
PPS traction Ppps-traction = Pcomm
462
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
by a power command, Pcomm , that the motor is expected to produce.
Thus, in traction mode, the electric power input to the motor drive can be
expressed as
Pm-in =
Pcomm
,
ηm
(15.1)
where ηm is the efficiency of the motor drive. However, in braking, the motor
drive functions as a generator, and the electric power output from the motor
is expressed as
Pm-out = Pmb-comm ηm ,
(15.2)
where Pmb-comm is the braking power command to the motor, which may
be different from the power command, Pcomm , coming from the brake pedal,
since not all the braking power, Pcomm , may be supplied by the regenerative
braking, as discussed in Chapter 13.
According to the motor power command and other vehicle information,
such as energy level in the PPS and minimum operating power of the fuel
cell system—below which the efficiency of the fuel cell will decrease significantly (see Figure 15.1)—the fuel cell system and the PPS are controlled
to produce the corresponding power. Various operating modes of the drive
train and the corresponding power control strategy are described in detail
below.
Standstill mode: Neither the fuel cell system nor the PPS supplies power to
the drive train. The fuel cell system may operate at idle.
Braking mode: The fuel cell system operates at idle, and the PPS absorbs
the regenerative braking energy, according to the brake system operating
characteristics.
Traction mode:
1. If the commanded motor input power is greater than the rated power
of the fuel cell system, the hybrid traction mode is used, in which
the fuel cell system operates with its rated power, and the remaining
power demanded is supplied by the PPS. The rated power of the fuel
cell system may be set as the top line of the optimal operating region
of the fuel cell.
2. If the commanded motor input power is smaller than the preset minimum power of the fuel cell system, and the PPS needs charging (the
energy level is less than the minimum value), the fuel cell system operates with its rated power—part of which goes to the drive train, while
the other part goes to the PPS. Otherwise, if the PPS does not need
charging (the energy level is near its maximum value), the fuel cell
system operates at idle and the PPS alone drives the vehicle. In the
latter case, the peak power that the PPS can produce is greater than
the commanded motor input power.
463
Fuel Cell Hybrid Electric Drive Train Design
3. If the load power is greater than the preset minimum power and
less than the rated power of the fuel cell, and the PPS does not need
charging, the fuel cell system alone drives the vehicle. Otherwise, if
the PPS does need charging, the fuel cell system operates with its rated
power—part of which goes to the drive train to drive the vehicle, while
the other part is used to charge the PPS.
Figure 15.3 illustrates the flowchart diagram of this control strategy.
15.3
Parametric Design
Similar to the design of the engine-based hybrid drive train, the parametric
design of the fuel-cell-powered hybrid drive train includes the design of the
traction motor power, the fuel cell system power, and the PPS power and
energy capacity.
15.3.1
Motor Power Design
300
60
Acceleration time
50
250
Constant speed
on flat road
200
40
Constant speed
on 5% grade road
30
150
20
100
10
50
0
0
10
20
30
40
50
60
Motor power (kW)
70
80
90
0
100
FIGURE 15.4 Motor power versus acceleration time and vehicle cruising speed.
Constant cruising speed (km/h)
Acceleration time (s) from 0 to 100 km/h
(60 mph)
The motor power is required to meet the acceleration performance of the
vehicle as discussed in previous chapters. Figure 15.4 shows the motor power
for a 1500-kg passenger car, with respect to the acceleration time from zero
to 100 km/h and a constant speed on a flat road and a 5% grade road. The
parameters used in this example are vehicle mass 1500 kg, rolling resistance
464
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
200
Vehicle mass 1500 kg
Motor power 70 kW
18
180
160
Acceleration time (s)
16
Distance
14
140
12
120
10
100
80
8
Time
6
60
4
40
2
20
0
0
10
20
30
40
50
60
70
Vehicle speed (km/h)
80
90
Acceleration distance (m)
20
100
FIGURE 15.5 Acceleration time and distance versus vehicle speed of the passenger car
example.
coefficient 0.01, aerodynamic drag coefficient 0.3, and front area 2 m2 . It can
be seen that accelerating the vehicle from zero speed to 100 km/h in 12 s needs
about 70 kW of motor power. Figure15.4 also shows the required power while
driving at a constant speed on a flat road and a 5% grade road. It can be seen
that 33 kW of motor power can support the vehicle driving at about 150 and
100 km/h on a flat road and a 5% grade road, respectively. In other words,
the motor power required for acceleration performance is much greater than
that required for constant speed driving. Thus, a 70 -kW traction motor is
considered to be the proper design for this vehicle example. Figure 15.5 shows
the acceleration time and distance covered by the vehicle during acceleration
driving.
15.3.2
Power Design of the Fuel Cell System
The PPS, as discussed in previous chapters, is only used to supply peak power
in short time periods and has a limited amount of energy in it. Thus, the fuel
cell system must be able to supply sufficient power to support the vehicle
while it drives at high constant speeds on a long trip (e.g., highway driving
between cities), and to support the vehicle to overcome a mild grade at a
specified speed without the help of the PPS.
For the 1500-kg example passenger car, as indicated in Figure 15.4, 33 kW of
motor power is sufficient to meet the power demand with about 150 km/h of
constant speed on a flat road and 100 km/h on a 5% grade road. Considering
the inefficiency of the motor drive, a fuel cell system of about 40 kW power
will be needed to support long trip driving (in the fuel cell system design, the
maximum power may be designed slightly larger than that dictated by the
constant speed driving).1
465
Fuel Cell Hybrid Electric Drive Train Design
15.3.3
Design of the Power and Energy Capacity of the PPS
15.3.3.1
Power Capacity of the PPS
Based on the maximum power of the motor determined by the specified acceleration performance, and the rated power of the fuel cell system determined
by the constant speed driving, the rated power of the peaking power source
can be determined by
Pmotor
Ppps =
− Pfc ,
(15.3)
ηmotor
where Ppps is the rated power of the peaking power source, Pmotor is the
maximum motor power, ηmotor is the efficiency of the motor drive, and Pfc
is the rated power of the fuel cell system. The rated power of the PPS in the
example passenger car is about 43 kW.
15.3.3.2
Energy Capacity of the PPS
The PPS supplies its energy to the drive train when peaking power is needed,
and restores its energy storage from regenerative braking or from the fuel cell
system. The energy changes in the PPS in a driving cycle can be expressed as
E = (Ppps-charge − Ppps-discharge ) dt
(15.4)
t
where Ppps-charge and Ppps-discharge are the charge and discharge power of the
PPS, respectively. The energy changes, E, in the PPS depend on the size of the
fuel cell system, the vehicle control strategy, and the load power profile along
with time. Figure 15.6 shows the time profiles of the vehicle speed, the power
of the fuel cell system, the PPS power, and the energy change in the PPS for
100
Vehicle speed (km/h)
50
0
40
Fuel cell power (kW)
20
0
50
PPS power (kW)
0
–50
0.1
0
–0.1
D Emax
Energy changes in PPS (kWh)
0
200
400
600
800
1000
1200
1400
Time (s)
FIGURE 15.6 Vehicle speed, fuel cell power, power of the PPS, and energy changes in the PPS.
466
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
100
Vehicle speed (km/h)
50
0
0
D Emax
–1
Energy changes in the PPS (kWh)
–2
0
200
400
600
800
1000
1200
1400
Time (s)
FIGURE 15.7 Energy changes in the PPS while powered by PPS alone in an FTP75 urban drive
cycle.
a 1500-kg passenger car with a 40-kW rated power fuel cell system, driving
in an FTP75 urban driving cycle and using the control strategy mentioned
above. Figure 15.6 indicates that the maximum energy change, ΔEmax , in the
PPS is quite small (about 0.1 kWh). This result implies that the PPS does not
need much stored energy to support the vehicle driving in this driving cycle.
It should be noted that the power producing capability of the fuel cell system is limited before the fuel cell system is warmed up, and the propulsion
of the vehicle relies on the PPS. In this case, the energy in the PPS will be
delivered quickly. Figure 15.7 shows the energy changes in the PPS in an
FTP75 urban driving cycle for a 1500-kg passenger car, while PPS alone propels the vehicle. It indicates that about 1 kWh of energy in the PPS is needed
to complete the driving cycle [approximately 10.62 km (6.64 miles) in 23 min],
and about 43.5 Wh of energy from the PPS will be discharged each minute.
Assuming that 10 min are needed to warm up the fuel cell system,3 about
435 Wh of energy in the PPS will be discharged.
Based on the maximum discharged energy in the PPS discussed above, the
energy capacity of the PPS can be determined by
CE =
ΔEmax
,
Cp
(15.5)
where CE is the total energy capacity of the PPS and Cp is the percentage of the
total energy capacity that is allowed to be used, according to the characteristics
of the PPS.
15.4
Design Example
Using the design methodology developed in previous sections, a fuelcell-powered hybrid drive train for a passenger car has been designed.1 For
Vehicle mass (kg)
Rated motor power (kW)
Rated power of fuel
cell system (kW)
Maximum power of PPS (kW)
Maximum energy storage
in PPS (kWh)
Acceleration time
(0–100 km/h or 60 mph) (s)
Gradeability (at 100 km/h or
60 mph) (%)
Fuel economy
FTP75 highway driving cycle
FTP75 urban driving cycle
Constant speed, at 100 km/h
or 60 mph
12
5
12
5
4.4 L/100 km or 53.4 mpg (gas. equi.)
1.155 kg H2 /100 km or
53.7 mile/kg H2
2/9 L/100 km or 81 mpg (gas. equi.)
0.762 kgH2 /100 km or
81.4 mile/kg H2
80.4 mile/kg H2
2.65 L/100 km or 88.7 mpg (gas.
equi.) 0.695 kg H2 /100 km or
89.1 mile/kg H2
0.512 kg H2 /100 km or
124 mile/kg H2
2.93 L/100 km or 80 mpg (gas. equi.)
0.769 kg H2 /100 km or
131 mile/kg H2
1.91 L/100 km or 123 mpg (gas. equi.)
—
—
43
1.5
1.81 L/100 km or 130 mpg (gas. equi.)
0.475 kg H2 /100 km or
1500
70
83
Fuel Cell
1500
70
40
Hybrid
Simulation Results for the 1500-kg Hybrid and Fuel-Cell-Alone-Powered Passenger Cars
TABLE 15.1
Fuel Cell Hybrid Electric Drive Train Design
467
468
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
(a)
(b)
100
100
80
60
40
20
0
Vehicle speed (km/h)
50
0
40
Fuel cell power (kW)
20
0
50
Vehicle speed (km/h)
30
PPS power (kW)
Fuel cell power (kW)
0
20
–50
0.1
10
0
Energy changes in PPS (kWh)
–0.1
0
200
400
600
0
800
1000
1200
0
1400
200
400
600
0.6
1000
1200
1400
0.6
Fuel cell system efficiency
Fuel cell system efficiency
800
Time (s)
Time (s)
0.5
Operating
points
0.4
0.3
Fuel cell system
efficiency
0.2
0.1
0.5
5
10
15
20
25
Operating
points
0.3
0.2
0.1
0
0
Fuel cell system
efficiency
0.4
0
30
0
10
20
Fuel cell system power (kW)
30
40
50
60
70
80
Fuel cell system power (kW)
FIGURE 15.8 Operating simulation of the fuel cell hybrid and fuel-cell-alone-powered
passenger car in an FTP75 urban drive cycle: (a) hybrid drive train and (b) fuel-cell-alonepowered drive train.
50
Vehicle speed (km/h)
0
40
Fuel cell power (kW)
20
Vehicle speed (km/h)
100
100
Fuel cell power (kW)
(b)
(a)
30
0
50
PPS power (kW)
0
–50
0.1
Energy changes in PPS (kWh)
0
–0.1
0
100
200
300
400
500
600
700
800
80
60
40
20
0
20
10
0
0
100
200
300
Time (s)
500
600
700 800
0.6
0.6
0.5
Fuel cell system efficiency
Fuel cell system efficiency
400
Time (s)
Operating
points
0.4
0.3
Fuel cell system
efficiency
0.2
0.1
0.5
0
5
10
15
20
Fuel cell system power (kW)
25
30
Operating
points
0.3
0.2
0.1
0
0
Fuel cell system
efficiency
0.4
0
10
20
30
40
50
60
70
80
Fuel cell system power (kW)
FIGURE 15.9 Operating simulation of the fuel cell hybrid and fuel-cell-alone-powered
passenger car in an FTP75 highway drive cycle: (a) hybrid drive train and (b) fuel-cellalone-powered drive train.
Fuel Cell Hybrid Electric Drive Train Design
469
comparison, a fuel-cell-system-alone-powered passenger car with the same
size has also been simulated. The simulation results are shown in Table 15.1
and Figures 15.8 and 15.9. The design and simulation results indicate that the
hybrid vehicle has much higher fuel efficiency and the same performance,
when compared with the fuel-cell-system-alone-powered vehicle.
References
1. Y. Gao and M. Ehsani, “Systematic design of FC powered hybrid vehicle drive
trains,” Society of Automotive Engineers (SAE) Journal, Paper No. 2001-01-2532,
Warrendale, PA, 2001.
2. D. Tran, M. Cummins, E. Stamos, J. Buelow, and C. Mohrdieck, “Development of
the Jeep Commander 2 FC hybrid electric vehicle,” Society of Automotive Engineers
(SAE) Journal, Paper No. 2001-01-2508, Warrendale, PA, 2001.
3. T. Simmons, P. Erickson, M. Heckwolf, and V. Roan, “The effects of start-up and
shutdown of a FC transit bus on the drive cycle,” Society of Automotive Engineers
(SAE) Journal, Paper No. 2002-01-0101, Warrendale, PA, 2002.
16
Design of Series Hybrid Drive Train
for Off-Road Vehicles
Off-road vehicles (military, agricultural, and construction) usually operate on
unprepared ground and need to overcome very complex and difficult ground
obstacles, such as steep grade and very soft ground. Depending on the functional requirements, different criteria are used to evaluate the performance
of various types of off-road vehicles. For tractors, their main function is to
provide adequate draft to pull various types of implement and machinery:
drawbar performance is of primary interest. This may be characterized by
the ratio of drawbar pull to vehicle weight, drawbar power, and drawbar
efficiency. For off-road transport vehicles, the transport productivity and efficiency are often used as basic criteria for evaluating their performance. For
military vehicles, the maximum feasible operating speed at two specific points
in a given area may be employed as a criterion for evaluation of their agility.1
Although different criteria are used to evaluate the performance of different types of off-road vehicles, there is a common requirement for all: mobility
over unprepared terrain. Mobility, in the broad sense, is concerned with performance of the vehicle in relation to soft terrain, obstacle negotiation and
avoidance, ride quality over rough terrain, water crossing, and so on.
This chapter discusses the design principle of an off-road hybrid electric tracked vehicle, focusing on power ratings of the traction motor,
engine/generator, and energy storage for satisfying specified vehicle performance indices, such as gradeability, acceleration, and steering on various
types of ground. For drive train control, refer to Chapter 7.
16.1
Motion Resistance
In addition to aerodynamic drag, rolling distance caused by tire deformation,
and track friction as discussed in Chapter 2, the motion resistance of an offroad vehicle mostly stems from significant deformation of the ground while
a vehicle is moving on it. In off-road operations, various types of terrain
with different characteristics, ranging from desert sand through deep mud
471
472
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
to snow, may be encountered. The mechanical properties of the terrain quite
often impose severe limitations on the mobility and performance of off-road
vehicles. The study of the relationship between the performance of an offroad vehicle and its physical environment (terrain) has now become known
as “Terranmechanics.”1
Although a study of “Terranmechanics” is beyond the scope of this book,
for proper design of a vehicle power train it is necessary to briefly introduce
some terranmechanics concepts, especially the motion resistance and thrust
that a terrain can support for off-road vehicle normal operation.
This section will briefly explain the calculation method of the tracked vehicle motion resistance and thrust, accompanied by an example tracked vehicle.
For more details on wheeled vehicles, see Wong1 and Bekker.2−4
16.1.1
Motion Resistance Caused by Terrain Compaction
The motion resistance, supported by consuming vehicle energy, to compact
terrain is studied by using penetration tests as shown in Figure 16.1. A plate
is used to simulate the contact area of a track. A vertical load, P, is placed on
the plate, resulting in sinkage, z, and terrain reaction pressure, p.
The work done by the load P can be expressed as
z0
W = bl
p dz,
(16.1)
0
where b and l are the plate dimensions of the short and long sides as shown
in Figure.16.1. The relationship between pressure, p, and sinkage, z, depends
P
z0
P
b
l
FIGURE 16.1 Terrain penetration test.
473
Design of Series Hybrid Drive Train for Off-Road Vehicles
on the terrain characteristic, which is determined by experiment and has the
expression1
1/n
p
z=
,
(16.2)
kc /b + kφ
where kc , kφ ,g and n are terrain parameters and b is the dimension of the
shorter side of the plate. Parameter kc reflects the cohesion characteristic and
kφ the internal friction characteristic of the terrain. Parameter n reflects the
“hardness” of the terrain. The terrain has a linear characteristic when n = 1, a
hard characteristic when n < 1, and a soft characteristic when n > 1. Typical
terrain parameters are listed in Table 16.1.
Substituting for p from Equation16.2 into Equation 16.1 yields
kc
+ kφ
Wc = bl
b
!
z0n+1
.
n+1
(16.3)
The interaction between the track of a tracked vehicle and the terrain is similar to that between a plate and the terrain as shown in Figure 16.1. Using
Equation 16.3, the vehicle motor resistance caused by terrain compaction can
be expressed as
!
z0n+1
Wc
kc
=b
+ kφ
.
(16.4)
Rc =
l
b
n+1
Example: A tracked vehicle with 196 kN gross weight and track dimensions of
l = 3.6 m and b = 1.0 m is operating on the terrain with parameters n = 1.6,
kc = 4.37 kN/m2.6 , and kφ = 196.73 kN/m3.6 . The vehicle resistance caused
by terrain compaction can be calculated as follows:
Pressure:
p=
196/2
P
=
= 27.2 kN/m2 .
bl
3.6 × 1.0
Sinkage:
z0 =
p
kc /b + kφ
1/n
=
27.2
4.37/1.0 + 196.73
1/1.6
= 0.2864 m.
Motion resistance:
!
z0n+1
Wc
kc
= 2b
+ kφ
Rc =
l
b
n+1
4.37
= 2 × 1.0
+ 196.73
1.0
0.28642.6
2.6
!
= 5.99 kN.
Snow (Harrison)
Lean clay (WES)
Heavy clay [Waterways
Experiment Station (WES)]
Clayey soil (Thailand)
Sandy loam (Hanamoto)
Sand loam Michigan (strong
Buchele)
Dry sand [Land Locomotion
Lab. (LLL)]
Sand loam (LLL)
Terrain
TABLE 16.1
Terrain Values
1.1
0.7
0.2
0.9
0.1
0.3
0.5
0.5
0.7
0.13
0.11
0.2
0.15
1.6
1.6
15
22
11
23
26
32
38
55
25
10
22
32
n
0
Moisture
Content (%)
7
45
5
0.07
0.04
15
5.3
0.7
12
7
45
2.3
7
11
0.1
lb/in.n+1
1.81
16.43
1.52
4.37
2.49
11.42
2.79
0.77
13.91
16.03
12.70
5.27
2.56
52.53
0.95
kN/m(n+1)
kc
10
120
10
0.08
0.10
27
6.8
1.2
16
14
140
16.8
3
6
3.9
lb/in.(n+2)
kp
103.27
1724.69
119.61
196.72
245.90
808.96
141.11
51.91
692.15
1262.53
1555.95
1515.04
43.12
1127.97
1528.43
kN/m(n+2)
3
10
2
0.15
0.09
1.4
2.0
0.75
0.6
0.3
10
0.25
0.2
0.7
0.15
lb/in.2
c
20.69
68.95
13.79
1.03
0.62
9.65
13.79
5.17
4.14
2.07
68.95
1.72
1.38
4.83
1.04
kPa
6
20
11
19.7
23.2
35
22
11
13
10
31
29
38
20
28
φ (deg)
474
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
475
Design of Series Hybrid Drive Train for Off-Road Vehicles
16.1.2
Motion Resistance Caused by Terrain Bulldozing
Another motion resistance may exist, caused by bulldozing the soil in front
of the vehicle track. This motion resistance is called bulldozing resistance. In
this section, only the equations and diagrams are introduced for the purpose
of calculating the bulldozing resistance.1
In predicting the bulldozing resistance, Bekker2,3 suggested equations by
assuming that it is equivalent to the horizontal force acting on a vertical blade:
(16.5)
Rb = b cz0 Kpc + 0.5z02 γs Kpγ ,
where b is the width of the track, c is the cohesion of the terrain (refer to Table
16.1), γs is the specific weight of the terrain, z0 is sinkage, and
Kpc = (Nc − tan φ) cos2 φ
and
Kpγ =
2Nγ
+ 1 cos2 φ.
tan φ
where, Nc and Nγ are the terrain’s bearing capacity factors as shown in
Figure 16.2, and φ is the angle of internal shearing resistance of the terrain. In
soft terrain or loose soils, local failure in front of the wheel or track may be
assumed, and the bulldozing resistance may be estimated using the following
equation1 :
+ 0.5z02 γs Kpγ
,
(16.6)
Rb = b 0.67cz0 Kpc
where
= Nc − tan φ cos2 φ
Kpc
40°
35°
Value of f
30°
N 'c
Nq
N 'q
N 'g
Nc
Ng
25°
20°
15°
10°
5°
0°
60
50
40
30
20
Value of Nc and Nq
10
0
5.7 1.0
20
40
60
80
100
Value of Ng
FIGURE 16.2 Variation of terrain bearing capacity factors with angle of internal shearing
resistance.1
476
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
and
Kpγ
=
2Nγ
tan φ
!
+ 1 cos2 φ .
Nc and Nγ are bearing capacity factors for local shear failure shown in Figure
16.2, and tan φ = 32 tan φ.
Example: The same vehicle as above is operating on the same terrain as shown
in the example in Section 16.1.1, with c = 1.0 kPa, internal shear angle φ =
19.7, and γs = 2646 N/m3 .
= 10.53
From Figure 16.2, for φ = 19.7, Nc = 11.37, and Nγ = 1.98, then Kpc
and Kpγ = 16.64. By Equation 16.6, Rb = 7.79 kN is obtained.
The total motion resistance, compacting and bulldozing terrain, is Rterr =
5.99 + 7.79 = 13.78 kN. The resistance coefficient defined as the motion
resistance per unit vehicle weight is fterr = 13.78/196 = 0.07.
16.1.3
Internal Resistance of the Running Gear
For wheeled vehicles, the internal resistance of the running gear is mainly
caused by the hysteresis of tire materials as discussed in Chapter 2. For
tracked vehicles, the internal resistance of the track and the associated suspension system may be substantial. Frictional losses in track pins, driving
sprockets and sprocket hubs, and road wheel bearing constitute the major
portion of the internal resistance of the track and associated suspension
system.1
Because of the complex nature of the internal resistance in the track and
suspension system, it is difficult to establish an analytical procedure to predict the internal resistance with sufficient accuracy. As a first approximation,
the following formula, proposed by Bekker, may be used for calculating
the average value of the internal resistance, Rin , of a conventional tracked
vehicle1,4 :
(16.7)
Rin = W(222 + 3V),
where Rin is in newtons, W is the vehicle weight in tons, and V is the vehicle
speed in km/h.
For modern lightweight tracked vehicles, the internal resistance may be less
and the empirical formula is1,4
Rin = W(133 + 2.5 V).
(16.8)
16.1.4 Tractive Effort of a Terrain
The tractive effort of a track is produced by the shearing of the terrain as
in Figure 16.3. The maximum tractive effort, Ft,max , that can be developed
Design of Series Hybrid Drive Train for Off-Road Vehicles
477
W
Shearing
F
FIGURE 16.3 Shearing action of a track.
by a track is determined by the shear strength of the terrain, which can be
expressed by1
Ft,max = Ac + W tan φ,
(16.9)
where W and A are the vertical load and contact area of the track to the
ground, respectively, and c and φ are the apparent cohesion and angle of
internal shearing resistance of the terrain, respectively. Their typical values
are shown in Table 16.1.
When the vehicle motion resistance is greater than the maximum tractive
effort that the terrain can develop, complete skidding will occur if the torque
developed by the mover is large enough, and the vehicle cannot move.
For the example vehicle in Sections 16.1.2 and 16.1.3, the terrain parameters
are c = 1.0 kPa, internal shear angle φ = 19.7, W = 196 kN, and A = 2 × l ×
B = 7.2 m2 . The maximum tractive effort is
Ft,max = Ac + W tan φ = 7.2 + 196 × tan 19.7◦ = 77.4 kN.
16.1.5
Drawbar Pull
For off-road vehicles designed for traction (i.e., tractors), drawbar performance is of prime importance, as it stands for the ability of the vehicle to pull
or push various types of working machinery, including implements and construction and earthmoving equipment. Drawbar pull, Fd , is the force available
at the drawbar and
is equal to the difference between the tractive effort and
the total resistance R. That is,
Fd = Ft −
R.
(16.10)
The tractive effort, Ft , may be determined by the power plant and transmission of the vehicle on a strong terrain, for example, a paved road, or by
the maximum thrust that the terrain can support on a soft terrain as discussed in Section 16.1.4. Because this book focuses on transportation vehicles,
drawbar performance will not be discussed further. However, the principles
478
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
developed in this book can be directly applied to the analysis of vehicle
drawbar performance.
16.2
Tracked Series Hybrid Vehicle Drive Train
Architecture
A tracked series hybrid off-road vehicle drive train mainly contains subsystems of a primary power source, a secondary power source, traction motors
and their controllers, a power converter, and a vehicle controller, as shown in
Figure 16.4.
Primary power source: The primary power source, generally a diesel engine
and a generator, is used to provide power to meet the average power demand
of the load. The diesel engine drives the generator to generate electricity to
charge the secondary power source and the batteries/ultracapacitors. Or, it
directly provides power to the electric motor drives. The generator generates
AC power. The nontraction AC power devices may be directly connected to
the output of the generator through transformers if the voltages of the AC
devices are different from the output voltage of the generator. An AC/DC
converter (rectifier) is used to convert the AC power into DC power.
Mechanical connection
Electrical connection
Signal and control
Track
To auxiliary
AC applies
To auxiliary
DC applies
AC/DC
rectifier
Gear
box
DC/DC
Converter
Power
converter
Battery
V and I
Batteries
Engine control
Operation commands
(acceleration, brake and
steering)
Vehicle
controller
Traction
Traction
motor
motor
Motor
controller
Gear
box
Engine throttle and speed
Generator control
Traction
motor
Motor
controller
Power flow control
AC
generator
Diesel
engine
Driven
sprocket
Traction motor control
Traction motor control
Speed signal
Speed signal
FIGURE 16.4 Tracked series hybrid off-road vehicle architecture.
Driven
Track sprocket
Design of Series Hybrid Drive Train for Off-Road Vehicles
479
Secondary power source: The secondary power source, generally a battery/
ultracapacitor, is used to supply peaking power to the traction motor drives
to meet the peak power demand of the load. The secondary power source
can be charged from the primary power source and/or the regenerative braking in which the motor functions as a generator to convert all or part of the
braking power of the vehicle into electrical power to charge the secondary
power source. In normal operation, the total output energy from the secondary power source should be equal to the total charging energy over the
mission period. Furthermore, the charging rate should be controlled to be in
an acceptable range.
Traction motors and controllers: The traction motors deliver their torque to the
sprockets through transmissions to propel the vehicle. The motor drives are
powered by the primary and/or secondary power sources and controlled by
their controllers to provide correct torques and speeds that meet the maneuver
requirements, according to the commands of the driver. These maneuvers
include acceleration, deceleration, forward and reverse moving, and steering.
Power converter: The power converter is an assembly of controllable power
electronics. It is used to control the power flow between the primary
power source, the secondary power source, and the motor drives. All of
the operation modes of the drive train are implemented by controlling
the power converter. The operation modes of the drive train mainly include
engine/generator-alone powered operation, batteries/ultracapacitors-alone
powered operation, both engine/generator and batteries/ultracapacitors
powered operation (hybrid traction), regenerative braking operation, and
batteries/ultracapacitors charging operation. The various architectures of the
power converter have been discussed in Chapter 7.
Vehicle controller: The vehicle controller is the highest-level, microprocessorbased system controller. The vehicle controller receives operation commands
from the driver (acceleration, deceleration, forward or reverse moving, steering, etc.) and drive train real-time operating information, such as vehicle
speed, and component real-time operating information, such as the voltages
and currents of batteries/ultracapacitors, engine throttle position, speed, etc.
Based on all the information received and the control strategy (software code
stored in the vehicle controller), the vehicle controller will generate the necessary control signals and send them to the components (engine/generator,
power converter, motor drives, etc.). The components follow the control
commands of the vehicle controller.
16.3
Parametric Design of the Drive Train
The parametric design of the drive train mainly includes traction motor
power design, engine/generator power design, and energy storage (battery/
ultracapacitor) power and energy design.
480
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
16.3.1 Traction Motor Power Design
In motor power design, the acceleration performance, maximum gradeability,
and steering are the highest considerations.
16.3.1.1 Vehicle Thrust versus Speed
A well-controlled traction motor usually has the torque and power characteristics of constant torque in the low-speed range and constant power in the
high-speed range, as shown in Figure 16.5. The corner speed is referred to
as the base speed. In traction motor design, two important parameters, maximum power and extended speed ratio x, which is defined as the ratio of
its maximum speed to its base speed, must be determined first. For a given
power rating, that is, power in the constant power range, motor torque is
expressed as
Tm =
30Pm
,
πnmb
nm ≤ nmb
=
30Pm
,
πnm
nm > nmb ,
(16.11)
where Pm is the motor power rating defined above, nmb is the motor base
speed (rpm) as shown in Figure 16.5, and nm is the motor speed (rpm) varying
from zero to its maximum.
Torque
Power
Base
speed
Motor speed
Maximum
speed
FIGURE 16.5 Typical torque and power profiles of traction motor versus motor speed.
481
Design of Series Hybrid Drive Train for Off-Road Vehicles
Thrust
Power
Vehicle
base speed
Vehicle speed
Maximum
speed
FIGURE 16.6 Typical thrust and power profiles versus vehicle speed with an electric motor as
its mover.
The motor torque and power profiles can be translated into the thrust of
the vehicle versus vehicle speed by
Ft =
Tm ηt ig
r
(16.12)
V=
πnm r
,
30ig
(16.13)
and
where ηt and ig are the transmission efficiency and gear ratio from the traction
motor to the driving sprockets, and r is the radius of driving sprockets. The
transmission may be single gear or multigear (for more details, see Chapter 4).
The typical thrust and power profiles are illustrated in Figure 16.6. The motor
power design given below is based on these profiles.
16.3.1.2
Motor Power and Acceleration Performance
The motor power required by the acceleration performance includes the
power for overcoming various mechanical resistances (losses in the locomotive mechanism as described by Equations 16.7 and 16.8, losses caused by
road deformation as discussed in Section 16.1, and aerodynamic drag) and
482
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
TABLE 16.2
Vehicle Parameter Values
M
20,000 kg
ta
Vf
8s
48 km/h (30 mph)
0.0138a
0.000918a
C
D
CD
Af
1.17
6.91 m2
a c and d correspond to the constants of 133 and 2.5
with vehicle weight in newtons and V in m/s.
the power for accelerating vehicle mass (inertial resistance). On a hard surface
road, the loss due to road deformation is negligible, and the motor power for
acceleration can be expressed as
Pacc
1
2
M 2
1
2
=
Vb + Vf + MVf c + dVf + ρa CD Af Vf3 (W),
2ta
3
2
5
(16.14)
where M is the vehicle mass in kg, ta is the time in seconds for accelerating the vehicle from zero speed to a specified final speed, Vf ,Vb is the base
speed in m/s, as shown in Figure 16.6, ρa is the air mass density in kg/m3 ,
CD is the coefficient of aerodynamic drag, Af is the front area of the vehicle
in m2 , and c and d are constants, representing the constant term and the term
proportional to the vehicle speed of the resistance of the locomotive mechanism as described in Equation 16.8. The resistance coefficient is expressed
as fr = c + dV. The vehicle parameters used in the motor power design are
shown in Table 16.2.
The motor powers required for acceleration performance (8 s from 0 to
48 km/h [30 mph]) with different extended speed ratios, x, of the thrust are
shown in Figure 16.7. It can be seen that the motor power decreases with
increasing extended speed ratio, x. However, when x ≥ 6, further decrease in
the motor power is not significant with further increase in the extended speed
ratio, x.
16.3.1.3
Motor Power and Gradeability
The motor power requirement in an uphill operation can be expressed as
Pgrade
1
2
= Mg fr + ρa CD Af V + Mg sin α V(W),
2
(16.15)
where α is the slope angle of the road and V is the vehicle speed specified by
the gradeability. When the vehicle is climbing its maximum slope (60%) in
483
Design of Series Hybrid Drive Train for Off-Road Vehicles
9
x = 2, Pm = 487 kW
x = 3, Pm = 363 kW
Acceleration time (s)
8
x = 4, Pm = 320 kW
7
x = 5, Pm = 300 kW
x = 6, Pm = 289 kW
6
x = 7, Pm = 282 kW
5
x = 8, Pm = 278 kW
4
x = 9, Pm = 275 kW
x = 10, Pm = 273 kW
3
2
1
0
0
0
10
6.25
20
12.5
30
18.6
40
25
50 km/h
31.3 mph
Vehicle speed
FIGURE 16.7 Motor power required by acceleration performance with different extended speed
ratios, x.
real operation, the ground surface is usually unprepared and thus resistance
is much larger than on prepared roads. Therefore, in the calculation of motor
power required by gradeability, additional resistance should be included to
reflect this situation. The resistance due to terrain deformation can be obtained
as discussed in Section 16.1. In the following sections, 0.06 of additional resistance coefficient is introduced (snow or sandy loam with about 21% moisture
content).5,6
Based on the specified gradeability, for example 60%, the tractive effort
profiles versus vehicle speed with different extended speed ratios and motor
powers are shown in Figure 16.8. It can be seen that with the same gradeability,
a larger extended speed ratio, x, will result in smaller motor power demand.
Figure 16.9 summarizes the motor power demand by acceleration and
gradeability performances along with the extended speed ratio, x, of the motor
drive. It can be clearly seen that the motor power required by gradeability
is larger than the motor power required by acceleration, especially with a
small x.
The above analysis indicates that one of the effective approaches for reducing the motor power rating is to increase the extended speed ratio. However,
the extended speed ratio of the motor drive is closely and naturally related to
motor type. PM motor drives have very small x, usually less than 2, due to their
rather limited field weakening capability.7 To keep the power of the motor
drive in a reasonable range and meet the gradeability requirement, they will
need a multigear transmission with three or four gears. A common induction
motor with speed adjustment control usually has an extended speed ratio of
2. Nevertheless, a properly designed induction motor, for example a spindle
motor, with field orientation control can achieve field weakening in the range
484
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Tractive effort and resistance (kN)
140
35° (70.0%)
x =2
Pm 1545 kW
x =3
Pm = 1030 kW
x=4
Pm = 773 kW
120
100
30° (57.7%)
25° (46.6%)
x =5
Pm = 618 kW
80
20° (36.4%)
15° (26.8%)
60
x =6
Pm = 515 kW
x =7
Pm = 442 kW
x=8
Pm = 386 kW
x=9
Pm = 343 kW x = 10
Pm = 309 kW
40
20
0
0
0
20
12.5
40
25
10° (17.6%)
5° (8.74%)
0° (0%)
60
37.5
100 km/h
62.5 mph
80
50
Vehicle speed
FIGURE 16.8 Tractive effort versus vehicle speed with different x and motor power.
of about 3–5 times its base speed.8−11 Even with this special design, a doublegear transmission is still needed. An SRM drive can inherently operate with
extremely long constant power range. Both 6–4 and 8–6 SRMs can reach 6–812 ;
thus a single-gear transmission would serve this requirement.
1600
Motor power (kW) required
by gradeability (60%)
1400
1200
1000
Motor power (kW) required
by acceleration (8 s, 0 to
48 km/h on hard road)
800
600
400
200
1
2
3
4
5
6
7
8
9
10
11
Extended speed ratio, x
FIGURE 16.9 Motor power required by acceleration and gradeability along with extended
speed ratios, x.
485
Design of Series Hybrid Drive Train for Off-Road Vehicles
16.3.1.4
Steering Maneuver of a Tracked Vehicle
Steering maneuver of a tracked vehicle is quite different from that of wheeled
vehicles. There are a number of possible methods that can accomplish the
steering of a tracked vehicle. These include skid-steering, steering by articulation in multibody vehicles, and curved track steering. For single-body
vehicles, skid-steering is the common method. In this book, only skid-steering
is discussed, which is closely related to the thrusts on both side tracks. For
other steering methods, readers are refered to Wong.1
In skid-steering, the thrust of one track is increased and that of the other
is reduced, so as to create a turning moment to overcome the moment of
resistance due to the skidding of the track on the ground and the rotational
inertia of the vehicle. Since the moment of the turning resistance is usually
considerable, significantly more power may be required during a turn than
in straight line motion.1
The turning behavior of a tracked vehicle using skid-steering depends on
the thrusts of the outside and inside tracks, Fto and Fti , the resultant resisting
force, Rtot , the moment of turning resistance, Mr , exerted on the track by the
ground, and vehicle parameters as shown in Figure 16.10. At low speeds on
level ground, the centrifugal force may be neglected, and the behavior of the
vehicle can be described by the following motion equations:
dV
= Fto + Fti − Rtot ,
dt
B
dωz
Iz
= (Fto − Fti ) − Mr ,
dt
2
M
(16.16)
(16.17)
Vo
Rtot/2
l
B
V
Fto
q
Vi
Mr
Fti
Rtot/2
wz
O
FIGURE 16.10 Skid steering behavior.
486
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
where Iz is the moment of inertia of the vehicle about the vertical axis passing
through its center of gravity and ωz is the turning angle velocity of the vehicle.
At low speeds and under steady-state conditions with zero linear and angular acceleration, that is, dV/dt = 0 and dωz /dt = 0, the thrusts of the outside
and inside tracks can be expressed as
Rtot
Mr
Mgfr
Mr
+
=
+
,
2
B
2
B
Rtot
Mr
Mgfr
Mr
Fti =
−
=
−
,
2
B
2
B
Fto =
(16.18)
(16.19)
where M and g are the vehicle mass and the acceleration of gravity,
respectively, and fr is the resistance coefficient due to ground deformation.
The moment of turning resistance, Mr , can be determined experimentally
or analytically. If the normal pressure is assumed to be uniformly distributed
along the track, the lateral resistance per unit length of the track, Rl , can be
expressed by
Rl =
μt Mg
,
2l
(16.20)
where μt is the coefficient of lateral resistance and l is the contact length of
the track as shown in Figure 16.11. The value of μt depends not only on the
terrain, but also on the design of the track. Over soft terrain, the vehicle sinks
Mg/2
l
dx
x
FIGURE 16.11 Moment of turning resistance of a track.
487
Design of Series Hybrid Drive Train for Off-Road Vehicles
TABLE 16.3
Values of Lateral Resistance of Tracks Over Various Surfaces
Coefficient of Lateral Resistance, μt
Track Materials
Concrete
Hard Ground (not Paved)
Grass
Steel
Rubber
0.50–0.51
0.09–0.91
0.55–0.85
0.65–0.66
0.87–1.11
0.67–1.14
Source:
J. W. Wong, Theory of Ground Vehicle, John Wiley & Sons, New York, 1978.
into the ground and the tracks together with the grousers will be sliding on
the surface as well as displacing the soil laterally during steering. The lateral
force acting on the track and grousers due to displacing the soil laterally forms
part of the lateral resistance. Table 16.3 shows the values of μt for steel and
rubber tracks over various types of ground.1
By referring to Figure 16.11, the resultant moment of the lateral resistance
about the center of the two tracks (i.e., moment of turning resistance) can be
determined by
l/2
Mgμt l/2
Mglμt
Rl x dx = 4
x dx =
.
(16.21)
Mr = 4
2l
4
0
0
Accordingly, Equations 16.18 and 16.19 can be rewritten as
Mg
lμt
fr +
,
Fto =
2
2B
Mg
lμt
Fti =
fr −
.
2
2B
(16.22)
(16.23)
From Equation 16.23, it can be seen that if lμt /2B > fr , the thrust of the inside
track, Fti , will be negative. This implies that to achieve a steady state, a conventional vehicle with a diesel engine as its mover must apply a braking force
on the inside track. In a series hybrid drive train, as shown in Figure 16.3, the
inside track motor can apply a negative torque (regenerative braking) to this
track.
Referring to the maximum terrain tractive effort and for normal operation
and steerability, the thrust of the outside track should be smaller than the
maximum terrain tractive effort, that is,
Mg tan φ
,
Fto ≤ cbl +
2
l
1 4cA
≤
+ 2 tan φ − 2fr ,
B
μr Mg
or
2
l
≤
B
μr
c
+ tan φ − fr ,
p
(16.24)
(16.25)
(16.26)
488
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Turning
left
Going
straight
Turning
right
a
FIGURE 16.12 A vehicle going straight and steering on a slope ground.
where A is the contact area of the track (A = b × l) and p is the normal pressure,
which is equal to Mg/2A.
From Equation 16.22, it can be seen that, in addition to overcoming motion
resistance, the motor drive of the outside track has to produce additional
thrust to overcome the turning resistance. The most difficult situation is steering on a slope as shown in Figure 16.12. In this case, the traction motor of the
outside track has to produce a large traction torque to overcome terrain resistance, grade resistance, and steering resistance. The total resistance of the
vehicle, in this case, can be expressed by
lμt
1
Ro = Mg fr +
+ ρa CD Af V 2 + Mg sin a,
2B
2
(16.27)
where fr is the resistance coefficient that includes the internal resistance of the
track and resistance caused by terrain deformation.
Figure 16.13 shows vehicle resistance that includes motion resistance, aerodynamic drag, grade resistance, and steering resistance as expressed in
Equation 16.27. The thrusts produced by the electric motors have the same
powers as shown in Figure 16.8 (e.g., straight line on 60% slope ground)
and different extended speed ratios, x. Since significant steering resistance
is involved, the ground grade, on which the vehicle can operate, is greatly
reduced. In motor and transmission design, this situation should be taken
into consideration.
489
Design of Series Hybrid Drive Train for Off-Road Vehicles
220
l /B = 1.5
160
35∞ (70.0%)
30∞ (57.7%)
25∞ (46.6%)
20∞ (36.4%)
15∞ (26.8%)
140
10∞ (17.6%)
Tractive effort and resistance (kN)
200
mt = 0.5
180
5∞ (8.74%)
120
x = 2, Pm = 1545 kW
80
60
40
20
0
0∞ (0%)
x = 3, Pm = 1030 kW
x = 4, Pm = 773 kW
100
x = 5, Pm = 618 kW
x = 6,
Pm = 515 kW
x = 7, Pm = 442 kW
x = 8, Pm = 386 kW
x = 9, Pm = 343 kW
x = 10, Pm = 309 kW
0
0
20
12.5
40
25
60
37.5
80
50
100 km/h
62.5 mph
Vehicle speed
FIGURE 16.13 Vehicle gradeability with steering versus vehicle speed with respect to various
x and motor powers.
16.4
Engine/Generator Power Design
The engine/generator power should be designed to meet the requirements
of constant speed operation at high speed (near its maximum speed) on
hard surface roads, and medium speed on soft surface roads for long distance trips, and also be larger than the average power at variable speed
(driving cycle) in order to prevent the energy storage from being fully discharged during the mission. The engine/generator unit also needs to produce
additional power to support the nontraction, continuous power, such as communication, lights, hotel loads, reconnaissance, and the auxiliaries (coolant
circulation, cooling fans, etc.). The peaking powers required by acceleration,
hill climbing, steering, and high-magnitude pulsed power needed by nontraction devices are provided by the PPS: batteries or a combination of batteries
and ultracapacitors.
Due to the absence of sufficient information about variable speed operation
(driving cycle) for off-road vehicles, the engine/generator power design will
be based on the constant speed operation on hard roads and soft grounds and
can be expressed by
1
Pm/g = ηt ηm Mg fr + ρa CD Af V 2 V,
2
(16.28)
490
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
800
700
Power (kW)
600
On soft ground
500
400
310 kW
300
On hard road
200
100
0
48 km/h
(30 mph)
0
0
20
12.5
40
25
60
37.5
91.4 km/h
(57 mph)
80
50
100 km/h
62.5 mph
Vehicle speed
FIGURE 16.14 Traction power on hard road and soft ground at constant speeds.
where ηt and ηm are transmission efficiency and motor drive efficiency, respectively. On hard roads, the resistance coefficient, fr , includes only the internal
resistance of run gear, which is described by c + dV as shown in Table 16.2,
and no road deformation loss. However, on soft roads, an additional resistance coefficient of 0.06 caused by ground deformation is added. Using the
parameters shown in Table 16.2 and assuming the total efficiency of converters
and electric motor drives as 0.85, the traction powers of the engine/generator
versus vehicle speeds on soft and hard surfaces are shown in Figure 16.14.
The engine/generator unit needs to produce additional power to support
nontraction continuous loads, such as communications, lights, hotel loads,
reconnaissance, and the auxiliaries (coolant circulation, cooling fans, etc.).
Due to the lack of accurate data from the nontraction loads for off-road vehicles, we take civilian vehicles as reference. The nontraction continuous power
may be estimated as 40–50 kW. Thus, the total engine/generator power is
designed around 350 kW.
It should be noted that the power of the engine/generator, designed above,
is the maximum power. In real-time operation, this power may be smaller,
depending on the real operation conditions and the overall drive train control
strategy.
16.5
Power and Energy Design of Energy Storage
Batteries or a combination of batteries and ultracapacitors are commonly
employed as the energy storage for providing peaking power to the drive
Design of Series Hybrid Drive Train for Off-Road Vehicles
491
train. Peaking power can be divided into traction power and nontraction
power. Traction power for peaking operation mainly includes the power
in acceleration, hill climbing, and steering, and nontraction power mainly
includes the power needed, for example, by high-power detection devices
and electric weapon systems in military vehicles.5,6
16.5.1
Peaking Power for Traction
In high-power traction, the drive train is powered by both the engine/
generator and batteries/ultracapacitors, and the maximum power is constrained by the designed maximum power of the traction motors as previously
specified. Thus, the traction peaking power for batteries/ultracapacitors is
Pb =
Pm,max
− Pe/g ,
ηm
(16.29)
where Pm,maxr and ηm are the maximum output power and efficiency of the
motor drives, respectively, and Pe/g is the engine/generator output power. It
should be noted that in acceleration and hill climbing operations, the engine
is not called upon to always operate at its maximum power, due to efficiency
concerns. Suppose, in the example vehicle above, the traction motor drives
have an efficiency of 0.85; the batteries/ultracapacitors power for traction is
then around 375 kW (515/0.85–230), where the engine/generator is delivering
230 kW of power to the traction motor (75% of maximum traction power of
the engine/generator [310 kW]).
16.5.2
Peaking Power for Nontraction
It is hard to accurately estimate the magnitude of the nontraction peaking
power. In military vehicles, the most significant nontraction pulsed loads
may be presented by “electric weapon” systems, for example, lasers, electrothermal chemical guns, electromagnetic armor, high-power microwaves, and
so on. The magnitude of the required pulsed power may reach 1 GW (109 W)
for a very short time period (10−3 s). It is clear that the on-board batteries/
ultracapacitors cannot supply this huge pulsed power, due to their internal
impedances. Thus, a pulse forming system that mainly consists of capacitors,
inductors, and resistors is needed. This system can be charged from the main
DC bus and can then discharge its energy to the pulsed load with a huge
power for a short time. Figure 16.15 conceptually illustrates the time profiles
of the pulsed power and the battery/ultracapacitor power, which may be
expressed as
1 tp Pb,max + Pb,min = Epulse .
2ε
(16.30)
492
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Peak power of the pulse
Epulse
Pb (Battery power)
Max. Pb
Min. Pb
Time, t
Tp
FIGURE 16.15 Conceptual illustration of the time profiles of pulsed power and battery power.
Thus,
Pb,max =
2Epulse
tp (1 + D)
,
(16.31)
where D = Pb,max /Pb,min is the charging power ratio and Tp is the period of
the pulsed power load.
Figure 16.16 illustrates the battery/ultracapacitor power for the pulsed
power load, varying with the charging power ratio D, and the period of the
pulsed power load, Tp . In this design, Tp = 4 s and D = 0.6 would be a good
estimate. Thus the maximum battery/ultracapacitor power will be around
Battery/ultracapacitor power (kW)
2000
1800
1600
D = 0.1
D = 0.2
D = 0.3
D = 0.4
D = 0.5
D = 0.6
D = 0.7
D = 0.8
D = 0.9
1400
1200
1000
800
600
400
200
0
1
2
3
4
5
6
7
8
9
10
Tp (s)
FIGURE 16.16 Peaking power of the batteries for pulsed power load varying with D and Tp .
493
Design of Series Hybrid Drive Train for Off-Road Vehicles
18 seconds discharge power (W/kg)
3000
2500
T = 40°C
2000
T = 25°C
1500
1000
T = 0°C
500
0
0
10
20
30
40
50
DoD (% )
60
70
80
90
FIGURE 16.17 Eight-second discharge power of the SAFT Li-ion batteries at different operation
temperatures and DoD.
300 kW. Adding this to the traction power, the total power requirement is
estimated around 675 kW.
It should be noted that the battery/ultracapacitor power capacity must be
maintained above the designed value for a certain period of time to support
the peaking power operation. For traction power demand, this time period is
required by acceleration, hill climbing, obstacle negotiation, and steering and
may be over 20–30 s. For nontraction peaking power demand, it is dependent
on the mission requirements.
Figure 16.17 shows the discharging power characteristics, in 18 s, of the
Li-ion batteries provided by SAFT America and tested by CHPS (Combat
Hybrid Power System sponsored by TACOM). It indicates that the battery
power is very dependent on temperature and DoD. Table 16.4 gives the
TABLE 16.4
Major Parameters of CHPS Battery Alternatives at Standard Testing
CHPS Battery Alternative
Lead acid
Ni–Cd
Ni–MH
Li-ion (high energy)
Li-ion (CHPS)
Li-ion (high power)
Specific
Energy (Wh/kg)
Specific
Power (Wh/kg)
Energy
Density (Wh/L)
28
50
64
144
100
80
75
120
140
700
1000a
1400a
73
80
135
308
214
a Power capabilities depends on pulse length and temperature.
150
494
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
specific power, specific energy, and energy density of the CHPS battery alternatives at standard testing. In this design, 1000 W/kg of specific power and
100 Wh/kg of specific energy would be a good estimate.
16.5.3
Energy Design of Batteries/Ultracapacitors
The energy requirements for batteries/ultracapacitors depend on the specific
mission requirements, for example the required time for stealth operation,
silent watch, “electric weapon” operation, and so on. However, when the
power capacity is determined, the energy capacity of the batteries can be
obtained from the energy/power ratio of the selected batteries.
As mentioned above, the battery power demand is around 675 kW and the
energy/power ratio of Li-ion battery is 0.1 h (specific energy/specific power).
Thus, 67.5 kWh of energy capacity is obtained. The battery weight is around
675 kg.
16.5.4
Combination of Batteries and Ultracapacitors
In addition to batteries, ultracapacitors are another possible PPS. Compared
with batteries, ultracapacitors have some advantages, such as 2–3 times the
specific power density of Li-ion batteries (see Tables 16.4 and 16.5), wide
temperature adaptability, high efficiency (low resistance), and fast response
to charging and discharging. Hence, it may be a good selection as a pulsed
power source. However, the ultracapacitor has the major disadvantage of a
low specific energy density of less than 5◦ W/kg. It cannot sustain its power
for more than a couple of minutes. Thus, it is difficult for ultracapacitors alone
to supply the peaking power for a vehicle.
TABLE 16.5
Technical Specifications of Maxwell MBOD 0115 Ultracapacitor Module
Capacitance
Max. series resistance ESR
Specific power density
Voltage
Max. current
Dimensions
Weight
Volume
Temperaturea
Leakage current
25 C
42 V
Continuous
Peak
(Reference only)
Operating
Storage
12 h, 25◦ C
145 Faradays (−20%/+20%)
10 mohm
2900 W/kg
42 V
50 V
600 A
195 × 265 × 415 mm
16 kg
22 L
−35◦ C to 65◦ C
−35◦ C to 65◦ C
10 mA
Source: Available at http://www.maxwell.com, Maxwell Technologies.
a Steady-state temperature.
495
Design of Series Hybrid Drive Train for Off-Road Vehicles
A good design for the PPS of a hybrid vehicle may be to combine Li-ion
batteries and ultracapacitors to constitute a hybrid energy storage, in which
the batteries supply the energy and the ultracapacitors supply the power.6
The combination of batteries and ultracapacitors meets the power and energy
requirements, that is,
Ptot = Wb Pb + Wc Pc ,
(16.32)
Etot = Wb Eb + Wc Ec ,
(16.33)
where Ptot and Etot are the total power and energy required, Wb and Wc are
the weights of the battery and the ultracapacitor, Pb and Pc are the specific
powers of the battery and the ultracapacitor, and Eb and Ec are the specific
energies of the battery and the ultracapacitor. For a given Ptot , Etot , Pb , Pc , Eb ,
and Ec , the battery and ultracapacitor weights can be obtained as
Wb =
Pc Etot − Ptot Ec
,
Pc Eb − P b Ec
(16.34)
Wc =
Ptot Eb − Pb Etot
.
Pc E b − P b Ec
(16.35)
Figure 16.18 shows the weights of batteries, ultracapacitors, and the hybrid
energy storage that is capable of supplying 675 kW of total power. It can be
1200
Ptot = 675 kW
Pb = 1 kW/kg
Pc = 2.5 kW/kg
Eb = 0.1 kWh/kg
Ec = 0.0025 kWh/kg
1000
Weight (kg)
800
Total weight
600
400
Battery weight
200
Ultracapacitor weight
0
–200
0
10
20
30
40
50
60
70
Total energy required (kWh)
80
90
100
FIGURE 16.18 Battery weight, ultracapacitor weight, and total weight of the hybrid energy
storage versus total energy.
496
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
…
Battery
Current (A)
Load
current
50
iC
iL
Ultracap
ib
100
0
Battery
current
–50
–100
0
5
10
15
Ultracapacitor
current
20
25
30
35
40
45
50
Time (s)
FIGURE 16.19 Current profiles of the battery/ultracapacitor energy storage.
seen that when the total energy requirement is less than 67.5 kWh, the hybrid
energy storage has less weight than the battery-alone energy storage. When
the total energy requirement is greater than 67.5 kWh, battery alone should
be used.
Other advantages of hybrid energy storage include leveling of the battery
peak current, as shown in Figure 16.19. This will simplify thermal management of the batteries, extend the battery life cycle, and offer fast power
response due to the very low resistance in the ultracapacitors. Other advanced
configurations of hybrid energy storage may be used for better performance
(refer to Chapter 12 for more details).
References
1. J. W. Wong, Theory of Ground Vehicle, John Wiley & Sons, New York, 1978.
2. M. G. Bekker, Theory of Land Locomotion, University of Michigan Press, Ann Arbor,
1956.
3. M. G. Bekker, Off-the-Road Locomotion, University of Michigan Press, Ann Arbor,
1960.
4. M. G Bekker, Introduction of Terrain-Vehicle Systems, University of Michigan Press,
Ann Arbor, 1969.
5. Y. Gao and M. Ehsani, “Parametric design of the traction motor and energy storage for series hybrid off-road and military vehicles,” IEEE Transactions on Power
Electronics, 21 (3), 749–755, May 2006.
6. Y. Gao and M. Ehsani, “Investigation of battery technologies for the army’s hybrid
vehicle application,” Vehicular Technology Conference, 2002. Proceedings, VTC 2002Fall, 2002 IEEE 56th, Vol. 3, pp. 1505–1509, September 24–28, 2002.
7. M. Ehsani, K. Rahman, and A. Toliyat, “Propulsion system design of electric and
hybrid vehicles,” IEEE Transactions on Industrial Electronics, 44 (1), 19–27, February
1997.
8. A. Boglietti, P. Ferraris, and M. Lazzari, “A new design criteria for spindle induction motors controlled by field orientated technique,” Electric Machine and Power
Systems, 21, 171–182, 1993.
Design of Series Hybrid Drive Train for Off-Road Vehicles
497
9. T. Kume, T. Iwakane, T. Yoshida, and I. Nagai, “A wide constant power range
vector-controlled AC motor drive using winding changeover technique,” IEEE
Transactions on Industry Applications, 27 (5), 934–939, September/October 1991.
10. M. Osama and T. A. Lipo, “A new inverter control scheme for induction motor
drives requiring speed range,” Proceedings of the IEEE-IAS Annual Meeting, Orland,
FL, pp. 350–355, 1995.
11. R. J. Kerkman, T. M. Rowan, and D. Leggate, “Indirect field-oriented control of
an induction motor in the field weakened region,” IEEE Transactions on Industry
Applications, 28 (4), 850–857, 1992.
12. K. Rahman, B. Fahimi, G. Suresh, A. Rajarathnam, and M. Ehsani, “Advanced
of switched reluctance motor applications to EV and HEV: Design and control
issues,” IEEE Transactions on Industry Applications, 36 (1), 11, January/February
2000.
13. Available at http://www.maxwell.com, Maxwell Technologies.
Appendix
Technical Overview of Toyota Prius
More and more hybrid vehicle products are being introduced to the market. Among them, Toyota Prius was the pioneer and has the largest number
of units on the road. To give the reader a case study of a commercially successful hybrid vehicle, the Toyota Prius technology is described
in this Appendix. This appendix will review the important technical features of this product, including the architecture, control, and component
characteristics. The main resource for this material is autoshop101.com
(http://www.autoshop101.com/forms/Hybrid01.pdf). However, the diagrams have been redrawn. We gratefully acknowledge the use of material
available in autoshop101.com.
A.1 Vehicle Performance
Prius is a Latin word meaning “to go before.” When the Prius was first
released, it was selected as the world’s best-engineered passenger car for
2002. The car was chosen because it is the first hybrid vehicle that seats
four to five people plus their luggage, and is one of the most economical
and environmentally friendly vehicles available. Then in 2004, the secondgeneration Prius won the prestigious Motor Trend Car of the Year award
and Best-Engineered Vehicle of 2004.
Both the Toyota Hybrid System (THS) power train in the original Prius and
the Toyota Hybrid System II (THSII) power train in the second-generation
Prius provide impressive electric power steering (EPS) fuel economy numbers
and extremely clean emissions as shown in Table A.1.
A.2 Overview of Prius Hybrid Power Train
and Control Systems
The hybrid power train of Toyota Prius uses the series–parallel architecture
as discussed in Chapters 5 and 9. Figure A.1 shows an overview of the hybrid
499
500
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
TABLE A.1
EPA Fuel Economy and Emissions
THS (2002–2003 Prius)
City
Highway
THS-II (2004 and Later)
52 mpg
45 mpg
City
Highway
SULEV
60 mpg
51 mpg
AT-PZEV
Notes: SULEV standards are about 75% more stringent than ULEV and nearly
90% than LEV for smog forming exhaust gases.
SULEV vehicles will emit less than a single pound of HCs during 100,000 miles
of driving (about the same as spilling a half quarter of gasoline).
AT-PZEV vehicles use advanced technology capable of producing zero emission during at least part of the vehicle’s driving cycle.
power train and control systems. The hybrid system components include the
following:
•
Hybrid transaxle, consisting of motor/generator 1 (MG1), motor/
generator 2 (MG2), and a planetary gear unit (refer to Figure A.3 for
more details)
•
1NZ-FXE engine
•
Inverter assembly containing an inverter, a booster converter,
a DC–DC converter, and an AC inverter
•
Hybrid vehicle electronic control unit (HV ECU), which gathers information from the sensors and sends calculated results to the engine
Shift position
sensor
Inverter
A/C
converter
Accelerator
position
sensor
Boost
converter
ECM
Skid control
ECU
SMR
Brake pedal
MG1
DC–DC
converter
HV
ECV
Speed sensor
Yew, Roll and
deceleration
Steering angle
SMR
Electric
inverter
compressor
SMR
Auxiliary
battery
Brake
ECU
MG1
Engine
Planetary
gear unit
Battery
ECU
HV
battery
CAN
Mechanical power path
Hydraulic power path
Electrical signal
FIGURE A.1 Overview of Prius power train and control systems.
Technical Overview of Toyota Prius
501
control module (ECM), inverter assembly, battery ECU, and skid
control ECU to control the hybrid system
•
Shift position sensor
• Accelerator pedal position sensor, which converts accelerator angle
into an electrical signal
•
Skid control ECU that controls regenerative braking
ECM
• High-voltage (HV) battery
•
•
Battery ECU, which monitors the charging condition of the HV
battery and controls cooling fan operation
•
Service plug, which shuts off the system
•
The system main relay (SMR) that connects and disconnects the HV
power circuit
•
Auxiliary battery, which stores 12 V DC for the vehicle’s control
systems
A.3 Major Components
A.3.1
Engine
The 1NZ-FXE engine is a 1.5-L inline four-cylinder gasoline engine with variable valve timing intelligence (VVTi) and Electric Throttle Control System
with Intelligence (ETCS-i). In the 2004 and later models, there is a special
coolant heat storage system that recovers hot coolant from the engine and
stores it in an insulated tank where it stays hot for up to three days. Later,
an electric pump precirculates the hot coolant through the engine to reduce
HC emission normally associated with a cold start.
Table A.2 shows the specifications of the 1NZ-FEX engine.
A.3.2
Hybrid Transaxle
Referring to Figure A.2, the hybrid transaxle contains
•
MG1 that generate electric power
•
MG2 that drives the vehicle
• A planetary gear unit that provides continuously variable gear ratios
and serves as a power splitting device
•
A reduction unit consisting of a silent chain, counter gears, and final
gears
•
A standard two-pinion differential
502
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
TABLE A.2
Specifications of 1NZ-FEX Engine
Model
Engine type
No. of cycles and arrangement
Value mechanism
Combustion chamber
Manifolds
Fuel system
Displacement (cm3 ) (cu. in.)
Bore × Stroke (mm) (in.)
Compression ratio
Max. output (SAE-Net)
Max. torque (SAE-Net)
Value timing
Intake
Open
Close
Exhaust
Open
Close
Firing order
Research octane number
Octane rating
Engine service massa (kg) (lb)
(References)
Oil grade
Tail emission regulation
Evaporative emission regulation
a
2004 Prius
2003 Prius
1NZ-FXE
Four-cylinder, in-line
Sixteen-value DOHC,
chain drive (with VVTi)
Pentroof type
Cross-flow
SFI
←
←
←
←
←
←
1497 (91.3)
75 × 84.7 (2.95 × 3.33)
13.0:1
57 kW at 5000 rpm
(76 hp at 5000 rpm)
111 Nm at 4200 rpm
(82 ft.lbf at 4200 rpm)
←
←
←
52 kW at 4500 rpm
(70 hp at 4500 rpm)
←
18◦ to −15◦ BTDC
72◦ to −105◦ ABDC
18◦ to −25◦ BTDC
18◦ to −15◦ ABDC
34◦ BBDC
34◦ ATDC
1-3-4-2
91 or higher
87 or higher
←
←
←
←
←
86.1 (198.8)
86.6 (190.9)
API SJ, SL, EC or ILSAC
SULEV
AT-PZEV, ORVR
API SH, SJ, EC or ILSAC
←
LEV-II, ORVR
Weight shows the figure with the oil and the engine coolant fully filled.
Table A.3 shows the main parameters of the hybrid transaxle. Table A.4
shows the specifications of MG1 and MG2.
A.3.3
HV Battery
The HV batteries are Ni–MH. Six 1.2 V cells connected together in series
constitute a battery module that has a voltage of 7.2 V.
In the 2001–2003 Prius, 38 battery modules are divided into two holders
and connected in series and have a rated voltage of 273.6 V.
503
Technical Overview of Toyota Prius
Differential
To wheel
To wheel
Final gears
Counter
gears
Chain
Transaxle
damper
Engine
flywheel
MG1
Planetary
gear unit
MG2
FIGURE A.2 Schematic illustration of the hybrid transaxle.
In the 2004 and later models, 28 battery modules are connected for a rated
voltage of 201.6 V. The cells are connected in two places to reduce the internal
resistance of the battery.
Table A.5 shows the HV battery information.
The battery ECU provides the following functions:
•
It estimates the charging/discharging amperage and the outputs
charge and discharge requests to the HV ECU so that the SOC can
be constantly maintained at a middle level.
•
It estimates the amount of heat generated during charging and
discharging, and adjusts the cooling fan to maintain HV battery
temperature.
•
It monitors the temperature and voltage of the battery, and if a malfunction is detected can restrict or stop charging and discharging
to protect the HV battery.
The SOC of the battery is controlled by the HV battery ECU. The target
SOC is 60%. When the SOC drops below the target range, the battery ECU
informs the HV ECU and then signals the engine ECM to increase its power
504
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
TABLE A.3
Main Parameters of the Transaxle
2004 Model
2003 Model
P112
P111
78
23
30
←
←
←
Chain
Number of links
Drive sprocket
Drive sprocket
72
36
35
74
39
36
Counter gear
Drive gear
Driven gear
30
35
←
←
26
75
←
←
3.8 (4.0, 3.3)
ATF WS or equivalent
4.6 (4.9, 4.0)
ATF type T-IV or equivalent
Transaxle type
Planetary gear
Number of ring gear teeth
Number of pinion gear teeth
Number of sun gear teeth
Differential gear ratio
Final gear
Drive gear
Driven gear
Fluid capacity
Liters (US qts, lmp qts)
Fluid type
TABLE A.4
Specification of the MG1 and MG2
MG1 Specification
Item
Type
Function
Maximum voltage (V)
Cooling system
MG2 specification
Item
Type
Function
Maximum voltage(V)
Maximum output kW(PS)/rpm
Maximum torque Nm (kgf.m)/rpm
Cooling system
2004 Model
2003 Model
Permanent magnet motor
Generate engine starter
AC 500
AC 273.6
Water-cooled
2004 model
2003 model
Permanent magnet motor
Generate engine starter
AC 500
50 (68)/1200–1540
400 (40.8)/0–1200
Water cooled
AC 273.6
33 (45)/1040–5600
350 (35.7)/0–400
505
Technical Overview of Toyota Prius
TABLE A.5
HV Battery Information
HV Battery Pack
Battery pack voltage
Number of Ni–MH battery modules in the pack
Number of cells
Ni–MH battery modules voltage
2004 and Later
2001–2003
206.6 V
28
168
7.2
273.6 V
38
228
←
to charge the HV battery. The normal low to high SOC deviation is 20% as
shown in Figure A.3.
The HV battery is air cooled. The battery ECU detects battery temperature
via three temperature sensors in the HV battery and one intake air temperature sensor. Based on their readings, the battery ECU adjusts the duty cycle
of the cooling fan to maintain the temperature of the HV battery within the
specified range.
Three SRMs are used to connect or disconnect power to the HV circuit based
on commands from the HV ECU. Two SRMs are placed on the positive side
and one is placed on the negative side, as shown in Figure A.4.
When the circuit is energized, SMR1 and SMR3 are turned on. The resistor
in line with SRM1 protects the circuit from excessive initial current (called
inrush current). Next, SRM2 is turned on and SRM1 is turned off.
When de-energized, SMR2 and SMR3 are turned off in that order and the
HV ECU verifies that the respective relays have been properly turned off.
A service plug is placed between the two battery holders. When the service
plug is removed, the HV circuit is shut off. The service plug assembly also
contains a safety interlock reed switch. Lifting the clip on the service plug
opens the reed switch, shutting off the SMR. There is also a main fuse for the
HV circuit within the service plug assembly.
Upper control limit
Example of change in SOC
Overcharge region
70%
60%
SOC
50%
Undercharge region
Lower SOC control limit
Time
FIGURE A.3 Battery SOC control region.
Target SOC control
Control
region
506
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
SMR 2
+
SMR 1
Resistor
Service
plug
SMR 3
–
FIGURE A.4 SMRs and service plug.
Toyota Prius uses an absorbed glass mat 12-V maintenance-free auxiliary
battery. This 12-V battery powers the vehicle’s electrical system similar to a
conventional vehicle.
A.3.4
Inverter Assembly
The inverter assembly includes an inverter, a booster converter, a DC–DC
converter, and an AC converter as shown in Figures A.1 and A.5.
A.3.4.1
Booster Converter (2004 and Later)
The booster converter boosts the nominal voltages of 206.1 V DC that is
output by the HV battery to the maximum voltage of 500 V DC. To boost
the voltage, the converter uses a boost integrated power module with a builtin insulated-gate bipolar transistor (IGBT) for switching control and a reactor
to store the energy, as shown in Figure A.5.
When MG1 or MG2 acts as a generator, the inverter AC, generated by either
motor, is converted to DC. Then the booster converter drops the voltage to
201.6 V DC to charge the HV battery.
A.3.4.2
Inverter
The inverter changes the HV DC from the HV battery into three-phase AC for
MG1 and MG2 as shown in Figure A.5. The HV ECU controls the activation
507
Technical Overview of Toyota Prius
Boost converter
Inverter
IGBT
IPM
Inductor
MG1
HV
battery
Current sensor
DC–DC
converter
IPM
A/C
inverter
MG2
Inverter assembly
Power
transistor
HV ECU
FIGURE A.5 Inverter assembly.
of the power transistors. In addition, the inverter transmits information that
is needed to control the current, such as the output amperage or voltage, to
the HV ECU.
The inverter, MG1, and MG2 are cooled by a dedicated radiator and coolant
system that is separated from the engine coolant system. The HV ECU controls
the electric water pump for this system.
A.3.4.3
DC–DC Converter
A DC–DC converter is used to transform the HV into 12 V to recharge the
12-V auxiliary battery. The structure of the DC–DC converter is shown in
Figure A.6. In the 2001–2003 models, it transforms 273.6 V DC to 12 V DC. In
the 2004 and later models, it transforms 201.6 V DC to 12 V DC.
A.3.4.4 AC Inverter
The inverter assembly in the 2004 and later models includes a separate inverter
for the air conditioner system that changes the HV battery’s nominal voltage
of 201.6 V DC into 206.6 V AC to power the air conditioner system’s electric
motor as shown in Figure A.7.
A.3.5
Brake System
The hybrid vehicle brake system includes both standard hydraulic brakes
and a regenerative braking system that uses the vehicle’s kinetic energy to
508
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
DC201.6 V
DC/DC converter
Input filter
AMD
GND
Auxiliary
battery
IG
Converter control unit
S
NODD
VLO
HV ECU
IDH
A/C ECU
FIGURE A.6 DC–DC converter.
recharge the battery. As soon as the accelerator pedal is released, the HV
ECU initiates the regenerative braking. MG2 is turned by the wheels and
used as a generator to recharge the battery. During this phase of braking, the
hydraulic brakes are not used. When more rapid deceleration is required,
the hydraulic brakes are activated to provide additional stopping power. To
increase energy efficiency, the system uses the regenerative brakes whenever
possible. Selecting “B” on the shift lever will maximize regenerative efficiency
and is useful for controlling downhill speeds. The overall structure of the
hybrid brake system is shown in Figure A.8.
A/C inverter
Current
sensor
HV
battery
Electric
compressor
Power
supply
circuit
Gate drive
Voltage Temp.
sensor sensor
IG
HV ECU
Input/
output
interface
FIGURE A.7 AC inverter.
CPU
System protection
control circuit
509
Technical Overview of Toyota Prius
Hydraulic
adjustment
area
ABS function area
MG2
Power splitting
device
Engine
MG1
Inverter
HV battery
FIGURE A.8 The hybrid brake system.
A.3.5.1
Regenerative Brake Cooperative Control
Regenerative brake cooperative control balances the brake force of the
regenerative and the hydraulic brakes to minimize the amount of kinetic
energy lost to heat and friction. It recovers the energy by converting it into
electrical energy.
A.3.5.2
Electronic Brake Distribution Control (2004 and Later Models)
In the 2004 and later models, brake force distribution is performed under
electrical control of the skid control ECU. The skid control ECU precisely controls the braking force in accordance with the vehicle’s driving
conditions.
1. Braking force distribution, front and rear (2004 and later models) Generally,
when the brakes are applied the vehicle’s weight shifts forward, reducing the
load on the rear wheels. When the skid control ECU senses this condition
(based on the speed sensor output), it signals the brake actuator to regulate
rear braking force so that the vehicle will remain under control during the
stop. The amount of brake force applied to the rear wheels varies based on
the amount of deceleration. The amount of brake force that is applied to the
real wheels also varies based on road conditions. Figure A.9a and b show
the braking force on the front and rear wheels without and with load on the
rear wheels.
510
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
(b)
Ideal braking force
distribution
EBD control
Brake force on front wheels
Brake force on rear wheels
Brake force on rear wheels
(a)
Ideal braking force
distribution
EBD control
Brake force on front wheels
FIGURE A.9 Braking force on the front and rear wheels: (a) without load on the rear wheels and
(b) with load in the rear wheels.
2. Brake force distribution, left and right (2004 and later models) When the
brakes are applied while the vehicle is cornering, the load applied to the
inner wheels decreases whereas the load applied to the outer wheels increases.
When the skid control ECU senses this situation (based on speed sensor output), it signals the brake actuator to regulate the brake force between the left
and right wheels to prevent a skid.
A.3.5.3
Brake Assist System (2004 and Later Models)
In emergencies, drivers often panic and do not apply sufficiently fast pressure
to the brake pedal. So in the 2004 and later models, a brake assist system (as
shown in Figure A.10) is used to interpret a quick push of the brake pedal as
emergency braking and supplements braking power accordingly.
To determine the need for an emergency stop, the skid control ECU looks
at the speed and the amount of brake pedal application based on signals from
the master cylinder pressure sensors and the brake pedal stroke sensor. If the
skid control ECU determines that the driver is attempting an emergency stop,
it signals the brake actuator to increase the hydraulic pressure.
A key feature of the brake assist system is that the timing and the degree
of braking assistance are designed to ensure that the driver does not discern
anything unusual about the braking operation. As soon as the driver eases up
on the brake pedal, the system reduces the amount of assistance it provides.
A.3.6
Electric Power Steering
A 12-V motor powers the EPS system so that steering feel is not affected
when the engine shuts off. The EPS ECU uses torque sensor output along
with information from skid control ECU about vehicle speed and torque
assist demand to determine the direction and force of the power assist. It then
511
Technical Overview of Toyota Prius
Master cylinder
pressure
sensor signal
Skid control
ECU
Brake pedal stroke
sensor signal
Brake
actuator
FIGURE A.10 Brake assist.
actuates the DC motor accordingly. The structure of the EPS system is shown
in Figure A.11.
The EPS ECU uses signals from the torque sensor to interpret the driver’s
steering intention. It combines this information with data from other sensors
regarding the current vehicle condition to determine the amount of steering
assist that will be required. It can then control the current to the DC motor
that provides steering assist.
Gateway EUC
Skit control ECU
Vehicle speed signal
assist torque demand
signal (for enhanced
VSC)
Torque
sensor
Master ECU
master warning light
EPS warning
Display request
Signal to Multi display
EPS ECU
DC
motor
HV ECU
Ready signal
CAN
DLC3
FIGURE A.11 The EPS system.
HV ECU
Ready signal
512
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
When the steering wheel is turned, torque is transmitted to the pinion causing the input shaft to rotate. The torsion bar that links the input shaft and the
pinion twists until the torque and reaction force equalize. The torque sensor
detects the twist of the torsion bar and generates an electrical signal that is
proportional to the amount of torque applied to the torsion bar. The EPS ECU
uses that signal to calculate the amount of power assist that the DC motor
should provide.
A.3.7
Enhanced Vehicle Stability Control (VSC) System
(2004 and Later Prius)
The enhanced vehicle stability control (VSC) system in the 2004 and later models helps maintain stability when the vehicle’s tires exceed their lateral grip.
The system helps control the vehicle by adjusting the motive force and the
brakes at each wheel when
•
the front wheels lose traction but the rear wheels do not (front wheel
skid tendency known as “understeer,”
•
the rear wheels lose traction but the front wheels do not (rear wheel
skid tendency, or “oversteer”).
When the skid control ECU determines that the vehicle is in understeer or
oversteer condition, it decreases engine output and applies the brakes to the
appropriate wheels individually to control the vehicle.
•
When the skid control ECU senses understeer, it brakes the front and
rear inside wheel. This slows the vehicle, shifts the load to the outside
front wheel, and limits the front wheel skid.
•
When the skid control ECU senses oversteer condition, it brakes the
front and rear outside wheel. This restrains the skid and moves the
vehicle back toward its intended path.
Enhanced VSC also provides the appropriate amount of steering assist based
on driving conditions by coordinating EPC and VSC control as shown in
Figure A.12.
A.4
Hybrid System Control Modes
Toyota Prius hybrid system uses series–parallel hybrid configuration that has
many operation modes, as was discussed in previous chapters. Prius uses the
following control strategies:
1. When starting off and traveling at low speed, MG2 provides the primary motive force. The engine may start immediately if the HV battery
513
Technical Overview of Toyota Prius
DC motor
Torque sensor
EPS ECU
Skid control ECU
Controls the
steering motor
after calculating
required steering
assist torque
- Detects a state of
the vehicle
- Calculates
required torque
for the steering
assists
Speed sensor
Wheel cylinder
pressure sensor
Steering angles
sensor
Yaw & deceleration
rate sensor
CAN
FIGURE A.12 Cooperative control with EPS.
SOC is low. As the speed increases above 24 km/h to 32 km/h (15–20
mph), the engine will start.
2. When driving under normal conditions, the engine’s power is divided
into two paths: a portion drives the wheels and a portion drives MG1
to produce electricity. The HV ECU controls the energy distribution
ratio for maximum efficiency.
3. During full acceleration, the power generated by the engine and MG1
is supplemented by power from the HV battery. Engine torque combined with MG2 torque delivers the power required to accelerate the
vehicle.
4. During deceleration or braking, the wheels drive MG2. MG2 acts as
a generator for regenerative energy recovery. The recovered energy
from braking is stored in the HV battery pack.
The operation modes of the engine, MG1, and MG2 are depicted in response
to different driving conditions as follows.
Stopped: If HV battery is fully charged and the vehicle is not moving, the engine may stop. The engine will start up automatically if the
HV battery needs charging. Also, if MAX AC is selected in the 2001–
2003 models, the engine will run continuously due to the engine-driven
compressor. The 2004 and later models use an electrically driven compressor. Figure A.13 shows the depicted operation modes of the engine, MG1,
and MG2.
Starting out: When starting out under light load and light throttle, only
MG2 turns to provide power. The engine does not run and the vehicle runs
on electric power only. MG1 rotates backwards and just idles. It does not
generate electricity, as illustrated in Figure A.14.
Engine starting: As the speed increases above 24 km/h to 32 km/h (15–
20 mph), the engine starts. The engine is started by MG1. The operations
of the engine, MG1, and MG2 are shown in Figure A.15.
514
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Ready on
Vehicle is stopped
Vehicle
speed
Planetary gear
MG2 & final drive MG1 Engine
0
Time
Inverter
HV
ECU
Drive
Driven
HV battery
MG1
MG2
Engine
FIGURE A.13 Operation with stopped vehicle.
Light acceleration with engine: In this mode, the engine delivers its power to
the driven wheels and MG1, which is generating. MG2 may assist the engine
for propulsion if required, depending on the engine power and the requested
driving power. In this mode, the energy generated by MG1 may be equal to the
energy delivered to MG2. The drive train operates as an EVT. The operations
of the engine, MG1, and MG2 are shown in Figure A.16.
Starting out
Vechicle is driving by MG2
Vehicle
speed
Planetary gear
MG2 & final drive MG1 Engine
0
Time
Inverter
HV
ECU
Drive
Driven
HV battery
MG1
Engine
MG2
FIGURE A.14 Operation mode of starting out.
515
Technical Overview of Toyota Prius
Engine starting
MG1 turns the engine
Vehicle
speed
Planetary gear
MG2 & final drive MG1 Engine
0
Time
HV
ECU
Inverter
Drive
Driven
HV battery
MG1
Engine
MG2
FIGURE A.15 Operation mode of engine starting.
Low-speed cruising: This mode is similar to the mode of light acceleration
with the engine as shown in Figure A.17.
Full acceleration: In this mode, the engine delivers its power to the wheels
and to MG1, which is in the generating mode. MG2 adds its power to the
engine power and is delivered to the wheels as shown in Figure A.18. The
power drawn by MG2 from HV battery power is greater than the power
Light acceleration with engine
Driven by MG2 & engine
Engine turns MG1 as generator
Vehicle
speed
Planetary gear
MG2 & final drive MG1 Engine
0
Time
Inverter
Drive
Driven
HV battery
MG1
Engine
MG2
FIGURE A.16 Operation mode of light acceleration with engine.
HV
ECU
516
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
Driven by MG2 & engine
Engine turns MG1 as genetator
Low speed cruising
Planetary gear
MG2 & final drive MG1 Engine
0
Time
Inverter
HV
ECU
Drive
Driven
HV battery
MG1
Engine
MG2
FIGURE A.17 Operation mode of low-speed cruising.
generated by MG1. Thus, the HV battery pack contributes energy to the drive
train and its SOC drops.
High-speed cruising: In this mode, the shaft of MG1 is fixed to the vehicle
stationary frame and the drive train is operated in pure torque-coupling mode.
Both the engine and MG2 propel the vehicle as shown in Figure A.19.
Driving with maximum speed: In this mode, both MG1 and MG2 receive
power from the HV battery pack and deliver their mechanical power to the
Driven by MG2 & engine
HV battery adds power
Full acceleration
Planetary gear
MG2 & final drive MG1 Engine
0
Time
Inverter
Drive
Driven
HV battery
MG1
Engine
MG2
FIGURE A.18 Operation mode in full acceleration.
HV
ECU
517
Technical Overview of Toyota Prius
Driven by MG2 & engine
MG1 locked in place
High speed cruising
Vehicle
speed
Planetary gear
MG2 & final drive MG1 Engine
0
Time
Inverter
HV
ECU
Drive
Driven
HV battery
MG1
Engine
MG2
FIGURE A.19 Operation mode at high-speed cruising.
drive train. In this case, MG1 turns in the opposite direction as shown in
Figure A.20.
Deceleration or braking: When the vehicle is decelerating or braking, the
engine is shut down. MG2 becomes a generator and is turned by the
drive wheels and generates electricity to recharge the HV battery pack. The
operation is shown in Figure A.21.
Driven by MG2 & engine
MG1 drives sun gear backward
Max speed
Planetary gear
MG2 & final drive MG1 Engine
0
Time
Inverter
+
Drive
Driven
HV battery
–
MG1
Engine
MG2
FIGURE A.20 Operation mode in driving with maximum speed.
HV
ECU
518
Modern Electric, Hybrid Electric, and Fuel Cell Vehicles
MG2 driven by wheels
and acts as a generator
Deceleration or braking
Planetary gear
MG2 & final drive MG1 Engine
0
Time
Inverter
HV
ECU
Drive
Driven by wheels
HV battery
MG1
Engine
MG2
FIGURE A.21 Operation mode in deceleration or braking.
Reverse: When the vehicle moves in reverse direction, MG2 turns in the
reverse direction as an electric motor. The engine is shut down. MG1 turns in
the forward direction and just idles as shown in Figure A.22.
MG2 drives backwards
Engine is off
Reverse
Planetary gear
MG2 & final drive MG1 Engine
0
Time
Inverter
Drive
HV battery
MG1
Engine
FIGURE A.22 Reverse operation.
MG2
HV
ECU
Index
1.5n switch converter, 225
1NZ-FEX engine, 501, 502
2S engines, 89–93
42-V Ultracapacitor, 409
4S, compression-ignition IC
engines, 88–89
4S, spark-ignited IC engines, 67
basic techniques for improving
engine performance,
efficiency, and emissions,
85–88
exhaust gas recirculation, 87
forced induction, 85–86
gasoline direct injection and
lean-burn engines, 86
intelligent ignition, 87
multi- and variable-valve
timing, 86–87
new engine materials, 87–88
throttle-less torque control, 87
variable compression ratio, 87
design and operating variables
affecting SI engine
performance, efficiency, and
emission characteristics,
78–84
compression ratio, 79–80
fuel/air equivalent ratio, 82–84
spark timing, 80–82
emission control, 84–85
engine operation characteristics,
76–78
engine performance parameters,
76–77
fuel consumption characteristics,
78, 79
indicated and brake power and
torque, 77–78
operating principles, 67–69
operation parameters, 69–75
fuel/air and air/fuel ratios, 73–74
indicated work per cycles and
mean effective pressure,
69–71
mechanical efficiency, 71–72
and performance parameters,
75–76
rating values of engines, 69
specific emissions, 73
specific fuel consumption and
efficiency, 72–73
volumetric efficiency, 74–75
2600F Maxwell Technologies
ultracapacitor, 395
Acceleration performance
of vehicle, 45–48
verification of, 273
Acceleration time and distance versus
vehicle speed, 115
Acid rain, 2
Activation losses, 437
Actively controlled hybrid
battery/ultracapacitor
energy storage, 405, 407
AER. See All electric range
Aerodynamic drag, 23–24
Air gap, in SRM design, 245
Air pollution, 1
carbon monoxide, 2
nitrogen oxides, 2
pollutants, 3
unburned HCs, 3
Alkaline fuel cells (AFCs), 444–446
Alkaline metal hydrides, 453
All electric range (AER), 335
AER-focused control strategy,
335–341
519
520
Alnico, 206–207
AM method. See Amplitude
modulation method
Ammonia
AFCs, 457–458
as hydrogen carrier, 457
Amplitude modulation (AM)
method, 234–235
ANN. See Artificial neural networks
Armature voltage, and DC motor
performance, 155, 158
Artificial intelligence controller, 211
Artificial neural networks (ANNs), 216
self-tuning using, 238–240
AshmanTechnology, 404
ATR. See Autothermal reforming
Automatic transmission, operating
efficiency of, 354–355
Automobile power train, 30
Autothermal reforming (ATR), 456
Auxiliary subsystem, 105
AVCON, 404
Back EMF integration, 216
Back-propagation training algorithm
(BPN), 216
Bacon, Francis, 17
Base speed, 226
Batteries. See also Electrochemical
batteries
and ultracapacitors, combination of,
494–496
Battery charging
from engine, 143
mode, 129, 355, 357, 371
Bell Laboratories, 13
Bidirectional DC/DC converter, 261, 262
BLDC motor derives, and PM, 153,
200–203
advantages, 201–202
construction and classification, 203
control, 211–212
disadvantages, 202–203
extend speed technology, 213
operation principle, 203
performance analysis, 208, 209–211
sensorless techniques, 213–217
unique technique, 216–217
using measurables and math, 214
Index
using observers, 215
using back EMF sensing, 215–216
sinusoidal-shaped back EMF BLDC
motor, 205
trapezoidal back EMF BLDC
motor, 204–205
Blended control strategy, 341–346
Boil-off, 452
Boost/buck DC/DC converter,
262, 263
BOOSTCAPi® ultracapacitor, 396
BPN. See Back-propagation training
algorithm
Brake performance, of vehicle, 53
braking distribution, on front and
rear axles, 55–60
braking force, 53–55
braking regulation and braking
performance analysis, 61
braking performance analysis,
62–65
braking regulation, 61–62
Brake power, 77
Brake specific fuel consumption
(bsfc), 72
Brake torque, 77
Braking and transmission, energy
consumed in, 353–355
Braking energy
versus braking power, 416
consumption in urban driving,
411–413
on front and rear axles, 419
versus vehicle deceleration rate,
417–419
versus vehicle speed, 413–416
Braking force, 53–55
Braking force distribution
on front and rear axles, 55–60
on front wheel, 423–424
Braking mode, 462
Braking performance analysis, 62–65
Braking power
versus braking energy, 416
versus vehicle speed, 416–417
Braking regulation, 61–62
Braking torque and braking force,
relationship between, 54
Briggs & Stratton Corporation, 16
Index
Brushless DC motor. See BDLC
Bsfc. See brake specific fuel consumption
Buick Skylark, 16
Bulldozing resistance, 475
C-dump inverter, 225, 226
Carbon dioxide, 2, 4, 5
Carbon monoxide, 2, 455
Carbon nanotubes, 454
CDR. See Charge-depleting range
Charge-depleting mode, 335, 343, 344
Charge-depleting range (CDR), 335
Charge-sustaining mode, 335
Chico, 17
Choppers, and DC motor control,
158–159, 160
wave forms, 161–162
CHPS (Combat Hybrid Power
System), 493
Battery Alternative at Standard
Testing, 408
Chrysler TE Van, 387
Citroën AX, 387
Civic Hybrid, 17
Class A choppers, 162
Class B choppers, 162–163
Class C choppers, 165–167
Class E choppers, 167–168
Classic converter, 224, 225
CLC. See Current limit control
Combustion, 1–2
Commutator motors, 151–152
Commutatorless motors, 152, 168. See
also Induction motor drives
Complex hybrid, 128
Compressed hydrogen, 450–452
Compression ratio, 79–80
Compression stroke, 68, 70
Concentration voltage drop, 438
Constant frequency TRC, 161
Constant volt/hertz control, 174–175,
177, 179
Constrained engine on–off control
strategy, 288–290
Continuously variable transmission
(CVT), 42–43
Control strategies, 256, 283–284
constrained engine on–off control
strategy, 288–290
521
dynamic programming technique,
292–295
engine on–off (thermostat) control
strategy, 257–258, 287–288
fuzzy logic control technique,
290–292
Max. SOC-of-PPS control strategy,
256–257, 284–287
for optimal braking performance,
427–429
for optimal energy recovery, 429–430
in vehicle controller, 461–463
“Cracking”, 457
Cryogenic liquid hydrogen, 452–453
Cumulative compound DC motor, 155
Cumulative oil consumption, 12
Current limit control (CLC), 161
CVT. See Continuously variable
transmission
Daily driving distance, statistics of,
333–334
Darracq, M. A., 13
DC/AC inverter with sinusoidal
pulse-width modulation,
178
DC/DC converter, 155, 260–261
DC motor drives, 152
armature, steady-state equivalent
circuit, 155, 156
armature voltage and field control,
combined, 158
chopper control, 158–163
multi-quadrant control of
chopper-fed drive, 163–168
four-quadrant operation, 167–168
two-quadrant operation, 163–167
operation principle, 154–155
performance, 155–158
wound-field DC motor, 154–155, 156
Depth of discharge (DOD), 349
Diagnostic pulse-based method, 235
Direct methanol fuel cells (DMFCs),
449–450
Direct methanol PEMFCs, 457
DMFCs. See Direct methanol fuel cells
DOD. See depth of discharge
Dodge, 16
522
Double-layer capacitor
technology, 391–392
Drag coefficients, for different body
shapes, 25
Drawbar pull, 477–478
Drive train, 123
auxiliary subsystem, 105
configurations
with speed coupling, 142–144
with torque coupling, 133–138
control strategy, 370
electric motor propulsion
subsystem, 105
energy source subsystem, 105
with floating-stator motor, 371–372
Drive train, parametric design of, 479
traction motor power design, 480
motor power and acceleration
performance, 481–482
motor power and gradeability,
482–484
tracked vehicle, steering
maneuver of, 485–489
vehicle thrust versus speed,
480–481
Driver’s expectation, 151
Dupont® , 443
Duty interval choppers, 160
Duty ratio, 159, 198
Dynamic equation, 26–27
Dynamic hydraulic torque converter, 354
Dynamic power, 126
Dynamic programming technique,
292–295
Dynamic Tequivalent circuit, of
induction motor, 189
EGR. See Exhaust gas recirculation
Electric Auto Corporation, 16
Electric braking, 15
Electric drive train, 105, 106
Electric motor
drive power design, 299–302
efficiency characteristics, 121
propulsion subsystem, 105
speed–torque (power) characteristics
of, 264
for traction, performance
characteristics of, 34
Index
Electric propulsion systems, 151–154
DC motor drives
armature voltage and field
control, combined, 158
chopper control, 158–163
multi-quadrant control of
chopper-fed drive, 163–168
operation principle, 154–155
performance, 155–158
functional block diagram, 152
induction motor drives, 168–169
constant volt/hertz control,
174–176
field orientation control, 179–193
operation principle, 169–172
power electronic control, 176–179
steady-state performance, 172–174
voltage source inverter for FOC,
193–200
permanent magnetic BLDC motor
devices, 200–203
construction and classification,
203
extend speed technology, 213
operation principle, 203
performance analysis and control,
208–212
PM materials, properties of,
205–208
sensorless techniques, 213–217
SRM drives, 217–218
basic magnetic structure, 218–221
design, 243–247
drive converter, 224–226
operation modes, 226–227
regenerative braking, 227–230
self-tuning techniques of, 236–240
sensorless control, 230–236
torque production, 222–224
vibration and acoustic noise,
240–242
Electric vehicle Kilometers (EVKM), 335
Electric vehicle miles (EVM), 335
Electric vehicles (EVs), 105
configurations, 105–108
energy consumption, 120–122
fully controllable hybrid brake
system, 426–430
history of, 12–14
523
Index
parallel hybrid braking system,
420–426
performance, 108
traction motor characteristics,
108–109, 110
tractive effort and transmission
requirement, 109–112
vehicle performance, 112–115
tractive effort, in normal driving,
115–120
Electric weapon systems, 491
Electrical angular velocity, 170
Electrical coupling device, 259–264
Electrical variable transmission
(EVS), 149
Electroboat, 13
Electrochemical batteries, 375–377
battery technologies, 385
lead–acid battery, 385–386
lithium-based batteries, 388–390
nickel-based batteries, 386–388
electrochemical reactions, 378–379
energy efficiency, 384–385
specific energy, 380–381, 383
specific power, 382, 384
thermodynamic voltage, 379
Electrode potential and current–voltage
curve, 437–440
Electromotive force (EMF), and DC
motor performance, 155
Electrovan, 13
EMF. See Electromotive force
Emission control, 84–85
Energy capacity, of PPS, 271–272,
305, 377
Energy consumption, 120–122
Energy management strategy, 335
AER-focused control strategy,
336–341
blended control strategy, 341–346
Energy source, 151
Energy source subsystem, 105
Energy storage
design, 346–351
hybridization of
battery and ultracapacitor size
design, 406–409
concept, 404
passive and active hybrid energy
storage with battery and
ultracapacitor, 404–406
power and energy design of,
490–491
batteries and ultracapacitors,
combination of, 494–496
batteries/ultracapacitors, energy
design of, 494
nontraction, peaking power for,
491–494
traction, peaking power for, 491
Energy/power ratio, 351, 407
Engine displacement, 71
Engine on–off control strategy, 257–258,
287–288
Engine operation characteristics, 76–78
engine performance parameters,
76–77
fuel consumption characteristics,
78, 79
indicated and brake power and
torque, 77–78
Engine performance parameters, 76–77
Engine traction, with battery charging
mode, 129
Engine/generator
power design, 489–490
power rating design of, 267–270
PPS charging from, 255–256
size, design of, 275–277
Engine/generator-alone traction
mode, 255
Engine-alone propelling mode, 285–286
Engine-alone traction, 142–143, 314–315,
355, 357, 369–370
ESX-1, 16
EVKM. See Electric vehicle Kilometers
EVM. See Electric vehicle miles
EVs. See Electric vehicles
EVS. See Electrical variable transmission
Exhaust gas recirculation (EGR), 87
Exhaust stroke, 69, 70
Expansion stroke, 69, 70
Ferrites, 208
Field flux, and DC motor
performance, 155
524
Field orientation control (FOC), 153,
179–193
control, 187–189
direct rotor flux orientation scheme,
189–192
indirect rotor flux orientation
scheme, 192–193
principles, 179–187
voltage source inverter, 193–200
current control, 198–200
voltage control, 195–198
Floating-stator motor, drive train with,
371–372
FOC. See Field orientation control
Forced induction, 85–86
Ford Hybrid Electric Vehicle
Challenge, 16
Ford Motor Corporation, 16
Ford Prodigy, 16
Four-quadrant operation, 167–168
Freewheeling diode conduction, 216
Freewheeling interval, choppers, 160
French Renault Next, 16
Frequency modulation method, 234
FTP75 highway drive cycle, 330–332, 468
simulation in, 364, 365
FTP75 urban drive cycle, 305–306, 336,
337, 338–339, 345, 346, 412,
413, 468
simulation in, 363
Fuel and oxidant consumptions, in fuel
cells, 440–441
Fuel cell
characteristics, 441–442
electrode potential and
current–voltage curve,
437–440
fuel supply, 450
ammonia as hydrogen carrier, 457
hydrogen production, 454–456
hydrogen storage, 450–454
non-hydrogen fuel cells, 457–458
operating principles, 433–437
and oxidant consumptions, 440–441
technologies, 443
alkaline fuel cells, 444–446
direct methanol fuel cells, 449–450
molten carbonate fuel cells,
447–448
Index
phosphoric acid fuel cells, 446–447
proton exchange membrane fuel
cells, 443–444
solid oxide fuel cells, 448–449
Fuel cell hybrid electric drive train
design, 459
configuration, 459–461
control strategy, 461–463
design example, 466–469
parametric design, 463
fuel cell system, power design
of, 464
motor power design, 463–464
PPS, power and energy capacity
of, 465–466
Fuel cell vehicles (FCVs), 411
brake system of
fully controllable hybrid brake
system, 426–430
parallel hybrid braking system,
420–426
history of, 17
Fuel consumption
characteristics, 78
of Chrysler upgraded turbine, 102
of Kronograd KTT gas turbine, 102
of different development strategies of
next-generation vehicles, 11
in drive cycles, 279
Fuel/air and air/fuel ratios, 73–74,
82–84
Fully controllable hybrid brake system,
420, 426–430
Fuzzy logic, 216–217, 290–292
F–V converter, 234
Gas turbine engines, 100
advantage, 103
disadvantage, 103
Gasoline, 455
Gasoline direct injection, 86
Gasoline engine
fuel economy characteristics, 48
performance characteristics, 33
tractive effort characteristics, 36
Gear ratio, design of, 272
General Motors (GM), 13, 14
Gibbs free energy, 379
Index
Global Earth atmospheric
temperature, 4, 5
Global warming, 3–5
GM Ovonic, 388
GM Precept, 16
GP, 388
Grade, 24, 25
Gradeability, 44–45, 112
verification of, 274
Grading resistance, 24–26
Greenhouse effect, 3
Gross indicated work, 70
Grove, Sir William, 17
GS, 388, 390
Half-bridge converter, 224, 225
Hall sensors, 189, 190, 192
HC. See Hydrocarbon
H-EBSs. See Hydraulic electric brake
systems
HEVs. See Hybrid electric vehicles
Holtzapple, Mark, 103
Honda, 17, 388
Honda Insight vehicles, 17
Hybrid battery charging mode, 129
Hybrid braking system, 286, 411, 420
fully controllable hybrid brake
system, 426–430
parallel hybrid braking system,
420–426
Hybrid electric drive trains, 123–126
architectures, 126–149
parallel hybrid electric drive
trains, 130–149
series hybrid electric drive trains,
128–130
with speed and torque coupling of
transmotor
and double shaft, 148
and single shaft, 148
Hybrid electric vehicles (HEVs), 123
classifications, 127
design, 390
fully controllable hybrid brake
system, 426–430
history of, 14–17
hybrid electric drive trains, 123–126
architectures, 126–149
525
parallel hybrid braking system,
420–426
Hybrid energy storage operation,
404–405, 406
Hybrid propelling mode, 285
Hybrid traction mode, 129, 142,
254–255, 356
Hybrid vehicle, 123
Hydraulic electric brake systems
(H-EBSs), 426
Hydrocarbon (HC), 2, 454
Hydrodynamic transmission, 38–42
advantages, 38–39
disadvantages, 39
Hydrogen on-board, storage of, 450
compressed hydrogen, 450–452
cryogenic liquid hydrogen, 452–453
metal hydrides, 453–454
Hydrogen production, 454–456
autothermal reforming, 456
POX reforming, 455–456
steam reforming, 454–455
Hydrogen–air fuel cell system, 441–442
Hydrogen–oxygen fuel cell, 438–439, 440
Hysteretic current controller, 198–199
IC engines. See Internal combustion
engines
ICEV. See Internal combustion engine
vehicle
Ihrig, Harry Karl, 17
Improved magnetic equivalent circuit
approach, 246
Indicated power, 77
Indicated torque, 77
Induced costs, 8–9
Induction motor drives, 152, 168–169
constant volt/hertz control, 174–176
field orientation control, 179–193
control, 187–189
direct rotor flux orientation
scheme, 189–192
indirect rotor flux orientation
scheme, 192–193
principles, 179–187
field orientation control, voltage
source inverter, 193–200
current control, 198–200
voltage control, 195–198
526
Induction motor drives (continued)
operation principle, 169–172
per-phase equivalent circuit, 173
power electronic control, 176–179
steady-state performance, 172–174
torque–slip characteristics, 174
Induction stroke, 68, 69–70
Insight, 388
Intake manifold, 85
Intelligent ignition system, 87
Interface circuitry, 151
Internal combustion (IC)
engines, 67
2S engines, 89–93
4S, compression-ignition IC engines,
88–89
4S, spark-ignited IC engines, 67
basic techniques for improving
engine performance,
efficiency, and emissions,
85–88
design and operating variables
affecting SI engine
performance, efficiency, and
emission characteristics,
78–84
emission control, 84–85
engine operation characteristics,
76–78
operating principles, 67–69
operation and performance
parameters, relationships
between, 75–76
operation parameters, 69–75
fuel economy characteristics, 48–49
gas turbine engines, 100–103
quasi-isothermal Brayton cycle
engines(QIBCE), 103–104
Stirling engines, 95–100
Wankel rotary engines, 93–95
Internal combustion engine vehicle
(ICEV), 105
Jenatzy, Camille, 13, 15
Kalman filter, 215
Knocking, 80
Krieger, H., 15
Index
“La Jamais Contente”, 13
LA92 driving cycle, 341, 342, 343, 347,
348, 349, 350
Langer, Charles, 17
Lawrence Livermore National
Laboratory, 404
Lead, 3
Lead–acid batteries, 358–359, 385–386
electrochemical processes of, 378–379
Lean-burn engines, 86
L–F converter, 234
Linear Alpha Inc., 16
Lithium-based batteries, 388–389
Li–I battery, 389–390
Li–P battery, 389
Lohner-Porsche vehicle, 15
Lunar Roving Vehicle, 13
Manual gear transmission, 35–38
Max. SOC-of-PPS control strategy,
256–257, 284–287
Maximum rated power, 69
Maximum speed, of vehicle, 43–44
Maximum brake torque (MBT) timing,
76, 80
Maxwell Technologies, 396, 397
Mazda Roadster, 387
Mazda rotary engine, 16
MBT timing. See Maximum brake torque
MCFCs. See Molten carbonate fuel cells
Mean effective pressure (mep), 71
and indicated work per cycles, 69–71
Mechanical angular velocity, 170, 171
Mechanical coupling, 131
Mechanical efficiency, of engine, , 71–72
Mechanical electric brake systems, 426
mep. See Mean effective pressure
Metal hydrides, 453–454
Methanol, 449
Microprocessor-based rotor flux
calculator, 190
Mild hybrid electric drive train
braking and transmission, energy
consumed in, 353–355
parallel mild hybrid electric drive
train
configuration, 355
drive train design, 356–360
Index
operating modes and control
strategy, 355–356
performance, 360–365
series–parallel mild hybrid electric
drive train
configuration, 365–367
control strategy, 370–371
drive train with floating-stator
motor, 371–372
operating modes and control,
367–370
Miller converter. See (n+ 1) switch
inverter
Mitsubishi EV, 387
Modern transportation, environmental
impact and history of, 1
air pollution, 1
carbon monoxide, 2
nitrogen oxides, 2
pollutants, 3
unburned HCs, 3
EVs, history of, 12–14
fuel cell vehicles, history of, 17
global warming, 3–5
HEVs, history of, 14–17
induced costs, 8–9
petroleum resources, 5–8
transportation development
strategies, to future oil
supply, 9–12
Modulated signal injection methods,
233–235
amplitude modulation (AM) method,
234–235
diagnostic pulse-based method, 235
frequency modulation method, 234
phase modulation (PM) methods,
234–235
Modulation index, 177
Molten carbonate fuel cells (MCFCs),
447–448, 458
Mond, Ludwig, 17
Morris and Salom’s Electroboat, 13
Motion resistance, 471–472
caused by terrain bulldozing, 475–476
caused by terrain compaction,
472–474
drawbar pull, 477–478
527
running gear, internal resistance
of, 476
terrain, tractive effort of, 476–477
Motor-alone propelling mode, 285
Motor-alone traction, 143, 355
Motor Drive Laboratory, 153
Motor/generator-alone traction,
315–316
Multi- and variable-valve timing, 86–87
Mutual-induced voltage-based
method, 236
(n + 1) switch inverter, 225, 226
Nafion, 443
National Aeronautics and Space
Administration (NASA), 17
NdFeB magnets, 208
Neodymium, 208
Nernst relationship, 379
New engine materials, 87–88
Newton’s second law
vehicle acceleration, 19
Nickel/cadmium battery, 387
Nickel/iron battery, 386–387
Nickel-based batteries, 386
Ni–MH battery, 388
nickel/iron battery, 386–387
nickel/cadmium battery, 387
Nissan, 42
Nitrogen oxides (NOx ), 2
Non-hydrogen fuel cells, 457–458
Nontraction, peaking power for, 491–494
Normal rated power, 69
Northrop Grumman, 404
Observer-based methods, 236
Off-road vehicles, series hybrid drive
train design for, 471
drive train, parametric design of, 479
traction motor power design,
480–489
energy storage, power and energy
design of, 490–491
batteries and ultracapacitors,
combination of, 494–496
batteries/ultracapacitors, energy
design of, 494
peaking power for nontraction,
491–494
528
Off-road vehicles, series hybrid drive
train design for (continued)
traction, peaking power for, 491
engine/generator power design,
489–490
motion resistance, 471–472
caused by terrain bulldozing,
475–476
caused by terrain compaction,
472–474
drawbar pull, 477–478
running gear, internal resistance
of, 476
terrain, tractive effort of, 476–477
tracked series hybrid vehicle drive
train architecture, 478–479
Oil consumption trends, 7
Oil supply
transportation development
strategies to, 9–12
Operating fuel economy, 48
fuel economy characteristics, of IC
engines, 48–49
vehicle fuel economy
computation, 49–51
techniques to improve, 51–53
Operation and performance
parameters, relationships
between, 75–76
Ovonic, 388
PAFCs. See Phosphoric acid fuel cells
Panasonic, 388, 390
Parallel hybrid electric drive train
(mechanical coupling), 127,
130–149, 281
advantages, 131
control strategies, 283–284
constrained engine on–off control
strategy, 288–290
dynamic programming technique,
292–295
engine on–off (thermostat) control
strategy, 287–288
fuzzy logic control technique,
290–292
Max. SOC-of-PPS control strategy,
284–287
disadvantages, 131
Index
drive train configuration and design
objectives, 281–282
drive train, parametric design of, 295
electric motor drive power design,
299–302
engine power design, 295–298
PPS design, 302–305
transmission design, 298–299
simulations, 305–306
with speed coupling, 138–144
with torque and speed coupling
optional coupling mode, 144–146
with speed and torque coupling
modes, 146–149
with torque coupling, 132–138
Parallel hybrid braking system, 420–426
Parallel mild hybrid electric drive train
configuration, 355
drive train design, 356–360
operating modes and control
strategy, 355–356
performance, 360–365
Paris Salon, 14, 15
Partial oxidation (POX)
reforming, 455–456
Partnership for New Generation of
Vehicles (PNGV), 16
Peaking power
for nontraction, 491–494
for traction, 491
Peaking power source (PPS), 333,
460, 464
charge mode, 285
design, 270, 302–305
energy capacity, 271–272
power capacity, 271
energy capacity, 465–466
design of, 277–279
SOC, 293, 327–328, 330
power capacity, 465
design of, 277
PPS-alone traction mode, 255
Peaking power sources, and energy
storages, 375
electrochemical batteries, 375–377
battery technologies, 385–390
electrochemical reactions,
378–379
energy efficiency, 384–385
Index
specific energy, 380–381, 383
specific power, 382, 384
thermodynamic voltage, 379
energy storage, hybridization of
battery and ultracapacitor size
design, 406–409
concept, 404
passive and active hybrid energy
storage with battery and
ultracapacitor, 404–406
ultracapacitors, 390
basic principles, 391–392
features, 390–391
performance, 392–396
technologies, 396–397
ultra-high-speed flywheels, 397
operation principles, 397–400
power capacity, 400–401
technologies, 402–404
PEMFCs. See Proton exchange
membrane fuel cells
Perfluorosulfonic acid, 443
Performance factor, 45
Permanent magnetic
BLDC motor devices. See BLDC
motor devices
hybrid motor drives, 213
materials, properties of, 205–208
Alnico, 206–207
ferrites, 208
rare-earth materials, 208
Per-phase equivalent circuit, of
induction motor, 173
Petroleum resources, 5–8
Peugeot 106, 387
Peugeot Société Anonyme (PSA), 14
Phase flux linkage-based method,
231–232
Phase inductance-based method,
232–233
sensorless control
based on phase bulk inductance,
232–233
on phase incremental
inductance, 233
Phase modulation (PM) methods,
234–235
PHExx, 335
529
Phosphoric acid fuel cells (PAFCs),
446–447
Pieper vehicle, 14
Planetary gear unit, 139, 366
Plug-in hybrid electric vehicles (PHEVs),
333–334, 335
design and control principles of, 333
daily driving distance, statistics
of, 333–334
energy management strategy, 335
AER-focused control strategy,
336–341
blended control strategy,
341–346
energy storage design, 346–351
PM methods. See Phase modulation
methods
PM synchronous motors, 153
PNGV. See Partnership for New
Generation of Vehicles
Point-by-point control. See Current limit
control
Post-transmission configuration, 136
Potassium hydroxide, 444
Power capacity
of flywheel systems, 400–401
of PPS, 271
Power converter, 479
Power design, of fuel cell system, 464
Power Electronics, 153
Power plant characteristics, 32–35
Power rating versus speed
factor, 115
Power train, 123
Power train tractive effort
and vehicle speed, 30–32
Power Center at UT Austin, 404
Power R&D, 404
POX. See Partial oxidation
PPS. See Peaking power source
Pressure versus volume, 70
Pretransmission configuration, 136
Priestly, 14, 15, 16
Primary EV power train, 106
Proportional-integral controller, 211
Proton exchange membrane fuel cells
(PEMFCs), 443–444
Proved petroleum reserves, 6
PSA. See Peugeot Société Anonyme
530
Pure electric traction mode, 129
Pure engine traction mode, 129
Quasi-isothermal Brayton cycle engines
(QIBCE), 103
advantage, 104
disadvantage, 104
R/P ratio, 6
Rare-earth PM materials, 208
Rated speed, 69
Rating values of engines, 69
R-dump-type inverter, 225
Reforming, 454
Regenerative braking factor, 121–122
Regenerative braking, 143, 164–167, 228,
256, 319, 356, 370, 411
braking energy
versus braking power, 416
versus vehicle deceleration rate,
417–419
versus vehicle speed, 413–417
design and control principles,
422–426
EV, HEV, and FCV, brake system
of, 420
fully controllable hybrid brake
system, 426–430
parallel hybrid braking system,
420–426
front and rear axles, braking energy
on, 419
urban driving, braking energy
consumed in, 411–413
Regenerative-alone brake mode, 286
Renault Clio, 387
Resistive-plus-inductive equivalent
circuit, 181
Road resistance, 25
Rocketdyne/Rockwell Trinity Flywheel
US Flywheel Systems, 404
Rolling resistance coefficient, 22, 23
Rolling resistance, 20–23
Running gear, internal resistance of, 476
SAFT America, 387, 388, 390, 493
Samarium–cobalt (SmCo5 ), 208
“Saturnism”, 3
Index
“Scavenge process”, 90
Self-tuning techniques, for SRM
drives, 236
with arithmetic method, 237–238
optimization in the presence of
parameter variations, 238
optimization with balanced
inductance profiles, 237–238
using ANN, 238–240
Sensorless control, 230–236
for SRM drives
modulated signal injection
methods, 233–235
mutual-induced voltage-based
method, 236
observer-based methods, 236
phase flux linkage-based method,
231–232
phase inductance-based method,
232–233
Sensorless technology, for BLDC drives,
213–214
unique technique, 216–217
using back EMF sensing
back EMF integration, 216
freewheeling diode conduction,
216
terminal voltage sensing, 215
third harmonic back EMF
sensing, 215
using measurables and math, 214
using observers, 215
Separately excited DC motor, 155, 157
Series (electrical coupling) hybrid
electric drive train, 127,
128–130, 253, 259
advantages, 129–130
control strategies, 256
engine on–off/thermostat control
strategy, 257–258
Max. SOC-of-PPS control strategy,
256–257
design example, 272
acceleration performance,
verification of, 273
engine/generator size, 275–277
fuel consumption, 279
gear ratio, 272
gradeability, verification of, 274
Index
PPS, energy capacity of, 277–279
PPS, power capacity of, 277
traction motor size, 272
disadvantages, 130
electrical coupling device, 259–264
engine/generator, power rating
design of, 267–270
operation patterns, 254–259
PPS design, 270
energy capacity, 271–272
power capacity, 271
traction motor, power rating design
of, 264–267
Series DC motor, 155, 157–158
Series–parallel (torque and speed
coupling) hybrid drive train,
127–128, 309
drive train configuration
drive train configuration, 313–319
speed-coupling analysis, 309–313
drive train control methodology
control system, 320
drive train control strategies,
323–328
engine speed control approach,
320–321
traction torque control approach,
321–323
drive train parameters design,
328–329
example vehicle, simulation of,
329–332
Series–parallel mild hybrid electric
drive train
configuration, 365–367
control strategy, 370