null null

APPLICATION OF VIRTUAL INSTRUMENTATION IN POSITION CONTROL
SYSTEM USING DIRECT DIGITAL CONTROL VIA PID AND FUZZY LOGIC
CONTROLLER
MARIAM BINTI MD GHAZALY
UNVERSITI TEKNOLOGI MALAYSIA
PSZ 19:16 (Pind.1/97)
UNIVERSITI TEKNOLOGI MALAYSIA
BORANG PENGESAHAN STATUS TESIS♦
JUDUL : APPLICATION OF VIRTUAL INSTRUMENTATION IN POSITION
CONTROL SYSTEM USING DIRECT DIGITAL CONTROL VIA PID
AND FUZZY LOGIC CONTROLLER
SESI PENGAJIAN:
Saya
2004 / 2005
MARIAM BINTI MD GHAZALY
(HURUF BESAR)
mengaku membenarkan tesis (PSM/Sarjana/Doktor Falsafah)* ini disimpan di Perpustakaan
Universiti Teknologi Malaysia dengan syarat-syarat kegunaan seperti berikut:
1.
2.
Tesis adalah hak milik Universiti Teknologi Malaysia.
Perpustakaan Universiti Teknologi Malaysia dibenarkan membuat salinan untuk
tujuan pengajian sahaja.
Perpustakaan dibenarkan membuat salinan tesis ini sebagai bahan pertukaran antara
institusi pengajian tinggi.
**Sila tandakan ( 9 )
3.
4.
SULIT
(Mengandungi maklumat yang berdarjah keselamatan atau
kepentingan Malaysia seperti yang termaktub di dalam
AKTA RAHSIA RASMI 1972)
TERHAD (Mengandungi maklumat TERHAD yang telah ditentukan
oleh organisasi/badan di mana penyelidikan dijalankan)
9
TIDAK TERHAD
Disahkan oleh
(TANDATANGAN PENULIS)
Alamat Tetap:
24, JALAN SUASANA 3/1A,
BANDAR TUN HUSSEIN ONN,
43200 CHERAS,
SELANGOR.
Tarikh : 4th APRIL 2005
CATATAN:
(TANDATANGAN PENYELIA)
ASSOC. PROF. DR. MOHD FUA’AD
BIN HJ RAHMAT
Nama Penyelia
Tarikh : 4th APRIL 2005
* Potong yang tidak berkenaan.
** Jika tesis ini SULIT atau TERHAD, sila lampirkan surat daripada pihak
berkuasa/organisasi berkenaan dengan menyatakan sekali sebab dan
tempoh tesis ini perlu dikelaskan sebagai SULIT atau TERHAD.
♦ Tesis dimaksudkan sebagai tesis bagi Ijazah Doktor Falsafah dan Sarjana secara
penyelidikan, atau disertai bagi pengajian secara kerja kursus dan penyelidikan, atau
Laporan Projek Sarjana Muda (PSM).
“I hereby declare that I had read this thesis and in my opinion this thesis is sufficient
in terms of scope and quality for the award of the degree of Master of Engineering
(Electrical - Mechatronic & Automatic Control)”
Signature
: …………………………………………….
Name of supervisor
: Assoc. Prof. Dr. Mohd Fua’ad Haji Rahmat.
Date
: 4th April 2005.
APPLICATION OF VIRTUAL INSTRUMENTATION IN POSITION CONTROL
SYSTEM USING DIRECT DIGITAL CONTROL VIA PID AND FUZZY LOGIC
CONTROLLER
MARIAM BINTI MD GHAZALY
A project report submitted in partial fulfillment of the requirements for
the award of the degree of Master of Engineering
(Electrical - Mechatronic & Automatic Control)
Faculty of Electrical Engineering
Universiti Teknologi Malaysia
APRIL 2005
ii
I declare that this thesis entitled “Application of Virtual Instrumentation in Position
Control System Using Direct Digital Control via PID and Fuzzy Logic Controller “is the
result of my own research except as cited in the references. The thesis has not been
accepted for any degree and is not concurrently submitted in candidate of any other
degree.
Signature
: …………………………...
Name
: Mariam binti Md Ghazaly.
Date
: 4th April 2005.
iii
To my beloved mother and father
iv
ACKNOWLEDGEMENT
In preparing this thesis, I was in contact with many people, researchers,
academicians, and practitioners. They have contributed towards my understanding and
thoughts. In particular, I wish to express my sincere appreciation to my main thesis
supervisor, Assoc. Prof. Dr. Mohd Fua’ad Haji Rahmat for encouragement, guidance,
critics and friendship. Without their continued support and interest, this thesis would not
have been the same as presented here.
My fellow postgraduate students should also be recognized for their support. My
sincere appreciation also extends to all my colleagues and others who have provided
assistance at various occasions. Their views and tips are useful indeed. Unfortunately, it
is not possible to list all of them in this limited space. I am grateful to my father and
mother for their guidance, advices and motivation.
v
ABSTRACT
The objective is to design and developed a GUI software using Microsoft
Visual Basic 6.0 to ease students in performing position control system experiment,
expose students on position control system theoretically and practically and to
developed controller using software for position control system. The scope of the
project is to apply direct digital control (DDC) in position control system. PID
controller and a fuzzy logic controller will be use to control the output response. An
interactive software will be developed to help the student to visualize and analyze the
system. This project can be divided into two parts namely hardware and software.
Hardware parts involve more in interfacing MS150 Modular servo System and Data
Acquisition System with a personal computer. While the software part include
programming real-time software using Microsoft Visual Basic 6.0. In the earlier
stage, literature review and experiment are performed manually to understand the
concept of the controller. An interactive software will be designed using Microsoft
Visual Basic 6.0. The software will be equipped with a set of graphic instructions to
ease any mistake when performing experiment. Finally, the software will be
integrated with hardware to produce a GUI position control system.
vi
ABSTRAK
Satu perisian komputer akan di reka dengan menggunakan Microsoft Visual
Basic 6.0. Tujuan perisian ini di reka adalah untuk pembelajaran dan pemahaman
pelajar semasa menjalankan ujikaji mengawal kedudukan sistem motor servo. Selain
itu, tujuan perisian ini adalah untuk mereka sistem kawalan menggunakan Visual
Basic 6.0 yang dapat mengawal kedudukan motor servo. Skop projek ini adalah
untuk mengadaptasikan modul kawalan terus digital dalam mengawal kedudukan
motor servo. Pengawal PID dan pengawal “fuzzy logic” akan digunakan bagi
mengawal keluaran motor servo. Satu perisian yang interaktif akan di reka agar dapat
membantu pelajar menganalisis sistem motor servo tersebut. Projek ini terbahagi
kepada dua bahagian iaitu perkakasan dan perisian. Perkakasan terdiri daripada
membuat hubungan antaramuka antara modal MS150 Modular servo sistem dengan
DAQ cad dan computer peribadi. Manakala, perisian adalah lebih kepada mereka
bentuk perisian yakni mereka bentuk pengawal PID dan pengawal “fuzzy logic”
menggunakan Microsoft Visual Basic 6.0 bagi mengawal sistem motor servo.
vii
TABLE OF CONTENTS
CHAPTER
1
TITLE
PAGE
DECLARATION
ii
DEDICATION
iii
ACKNOWLEDGMENT
iv
ABSTRACT
v
ABSTRAK
vi
TABLE OF CONTENTS
vii
LIST OF TABLES
xi
LIST OF FIGURES
xii
LIST OF SYMBOLS
xiv
LIST OF APPENDICES
xviii
INTRODUCTION
1
1.1
Virtual instrumentation
1
1.2
Virtual Instruments
3
1.3
Project Objectives
5
1.4
Project Scope
7
viii
2
LITERATURE REVIEW
8
2.1
Introduction.
8
2.2
Previous Research.
9
2.2.1
10
“Motor Drive Control System For
Education and Research”.
2.1.2
“Development of a Distant Laboratory
10
using Labview”.
2.1.3
“Real-Time Position Control of
11
Free-Electron Laser Beams: Theory
and Experiments”.
2.1.4
“Closed-loop Position Control
11
System using Labview”.
2.3
Virtual Instrumentation in Position
12
Control System.
3
RESEARCH METHODOLOGY
13
3.1
Project Overview
13
3.2
Position Control System
15
3.2.1
General Servomotor System Model
20
3.2.2
Direct Digital Control (DDC)
24
3.2.3 PID controller
25
3.2.3.1 Analog PID controller
27
3.2.3.2 Digital PID controller
30
3.2.4 Fuzzy Logic Controller
33
3.2.4.1 Fuzzification
36
3.2.4.2 Knowledge Base
38
3.2.4.3 Inferencing
40
3.2.4.4 Defuzzification
41
ix
3.3
Software Review
42
3.3.1
42
Virtual Instrumentation - Graphical User
Interface (GUI)
3.3.2
3.4
3.3.1.1 General GUI Design Considerations.
43
3.3.1.2 Visual Presentation
45
3.3.1.3 VI Diagram Considerations
46
3.3.1.4 Additional GUI Object Considerations
46
Microsoft Visual Basic
47
Hardware Review
48
3.4.1
Data Acquisition System (DAS)
48
3.4.1.1 DAQ Hardware
50
3.4.1.2 Analog Input system
53
3.4.1.3 Analog Output System
54
3.4.1.4 D/A Converter
55
MS150 Modular Servo System
55
3.4.2.1 Pre-amplifier Unit – PA150C
56
3.4.2.2 Operational Amplifier Unit -OA150A
56
3.4.2.3 Input Potentiometer – IP150H &
57
3.4.2
Output Potentiometer –OP150K
3.4.2.4 Attenuator Unit – AU150B
58
3.4.2.5 DC Motor –DCM150F
58
3.4.2.6 Reduction Gear Tacho Unit – GT150X
59
3.4.2.6 Servo Amplifier Unit – SA150D
59
3.4.2.7 Power Supply Unit – PS150E
60
3.4.3
Hardware Connection
61
3.4.4
Classical Control System Experiment
62
x
4
RESULTS & ANALYSIS
63
4.1
Direct Digital Control
64
4.1.1
PID controller GUI
65
4.1.2
Fuzzy Logic Controller GUI
67
4.2
4.3
5
Preliminary Result Analysis
69
4.2.1
Closed Loop Position Control System
70
4.2.2
Effect of Gain Changes on Deadband
73
4.2.3
Unit Step Response to Gain Changes
74
4.2.4
Velocity Feedback
77
Result Statement
82
4.3.1
PID Controller Result Analysis
83
4.3.2
Fuzzy Logic Controller Result Analysis
85
4.4
Comparison between PID and Fuzzy Logic
86
4.5
Discussion
87
CONCLUSION
89
5.1
91
Suggestion for future work
REFERENCES
92
Appendices A-D
94
xi
LIST OF TABLES
NO.
TITLE
PAGE
3.1
Effect of each PID Controller
30
4.1
Output result of the experiment unit step response to
75
gain changes.
4.2
Output result of the experiment velocity feedback
78
4.3
Output result of the PID controller
84
4.4
The comparison output between response using the
87
PID and Fuzzy logic.
xii
LIST OF FIGURES
NO.
TITLE
PAGE
1.1
Structure of a measuring device.
2
1.2
Hardware architecture of virtual instruments.
4
3.1
General form of the block diagram for closed loop position
15
control system.
3.2
Standard measurements on Step Response of a Control System
17
3.3
Single position loop diagram can be modified to include
17
velocity feedback.
3.4
Simple Error Channel
18
3.5
DC-Motor (a) Wiring Diagram (b) Sketch (c) Schematic
20
3.6
Direct Digital Control Schematic.
24
3.7
General Feedback Architecture
26
3.8
Block diagram for PID Controller
28
3.9
Flow chart for the PID controller system.
32
3.10
Flowchart of Simulator
34
3.11
FLC Design Methodology
35
3.12
The input and output variable.
36
3.13
The error input variable
37
3.14
The rate input variable
37
3.15
The motor output variable
38
3.16
The fuzzy logic rules
39
xiii
3.17
The firing strength for the rule base using max-min inference.
41
3.18
Computerized data acquisition and control system.
49
3.19
A picture of AX5412H Data Acquisition Card
50
3.20
AX750 screw terminal panel
52
3.21
Many input signals come into the A/D system
53
3.22
General View of MS150 Modular Servo System.
55
3.23
PA150C- Pre-amplifier Unit
56
3.24
OA150A – Operational Amplifier Unit
57
3.25
Input Potentiometer and Output Potentiometer (IP150H, OP150K)
57
3.26
AU150B – Attenuator Unit
58
3.27
DCM150F – DC Motor with Tachogenarator
59
3.28
SA150D – Servo Amplifier Unit
60
3.29
PS150E – Power Supply Unit
60
3.30
Connection between PC and MS150 through AX750 Screw
61
Terminal Panel
4.1
Splash screen
64
4.2
Circuit diagram for direct digital control.
65
4.3
PID controller GUI
66
4.4
Fuzzy Logic Controller GUI
68
4.5
Fuzzy Logic Controller data capture form
68
4.6
The output result of closed loop position control system experiment.
72
4.7
The output result of gain changes on deadband
74
4.8
PID results which was capture within the GUI
83
4.9
The fuzzy logic result which was capture within the GUI.
85
xiv
LIST OF SYMBOLS
VI
-
Virtual Instrument
PC
-
Personal Computer
DAQ
-
Data Acquisition
DSP
-
Digital Signal Processing
GUI
-
Graphical User Interface
DDC
-
Direct Digital Control
ITL
-
Interactive Teaching and Learning
PID
-
Proportional Integral Derivative Controller
I&C
-
Instrumentation and Control
DC
-
Direct Current
ADC
-
Analogue to Digital Converter
DAC
-
Digital to Analogue Converter
FCE
-
Final Control Element
θ i (t )
-
Input shaft angle
θ o (t )
-
Output shaft angle
Ra
-
Armature resistance
La
-
Armature inductance
I(t)
-
Armature current
F
-
Magnetic field force
1
-
Length of conductor
β
-
Magnetic field strength
v(t )
-
Velocity of conductor normal to the magnetic force
xv
vb (t )
-
Back electromotive force
Kb
-
Motor dependant constant
τ (t )
-
Torque
J
-
Total moment inertia
B
-
Total viscous friction
KT
-
Motor torque constant
e(t)
-
Error signal
Kg
-
Gain
ζ
-
Damping ratio
wn
-
Natural frequency
Tp
-
Peak time
Tr
-
Rise time
%OS
-
Percent overshoot
Mp
-
Peak value of time response
fv
-
Final value
Ts
-
Settling time
CV
-
Control variable
SP
-
Set point
y (t )
-
Output position signal
r (t )
-
Reference signal
V (t )
-
Controller output
Kp
-
Proportional gain
Ki
-
Integral gain
Kd
-
Derivative gain
DMA
-
direct Memory Access
I/O
-
Input / output
RTDs
-
Resistance temperature detectors
IC
-
Integrated circuit
AI
-
Analogue input
xvi
AO
-
Analogue output
DI
-
Digital input
DO
-
Digital output
Hz
-
Hertz
LSB
-
Least significant bit
TTL
-
Transistor-transistor logic
PA150C
-
Preamplifier unit
OA150A
-
Operational amplifier unit
IP150H
-
Input potentiometer
OP150K
-
Output potentiometer
AU150B
-
Attenuator unit
DCMl50F
-
DC motor
GT150X
-
Reduction gear tacho unit
PS150E
-
Power supply
Ω
-
Ohm
RAD
-
Rapid Application Development
HTML
-
Hypertext Markup Language
VB6
-
Visual Basic Version 6.0
COM
-
Component Object Model
IIS
-
Internet Information Server
ASP
-
Active Server Pages
DLL
-
Dynamic. Link Libraries
API
-
Application Programming Interface
IDE
-
Integrated Development Environment
OLE DB
-
Object Link Embedded Database
MTS
-
Microsoft Transaction Server
Ne
-
negative error
Pe
-
positive error
Ze
-
zero error
Nde
-
negative de
Pde
-
positive de
xvii
Zde
-
zero de
No
-
negative output
Po
-
positive output
Zo
-
zero output
xviii
LIST OF APPENDICES
APPENDIX
TITLE
PAGE
A
PID results
94
B
Software snapshot
102
C
Slide presentation
106
D
Software source code
134
CHAPTER 1
INTRODUCTION
1.1
Virtual instrumentation
Virtual instrument have become the catchword in measurement technology.
Virtual instrumentation easily can be connected with the term “test engineer”. When
attempting to interpret the meaning of “test engineer”, almost everyone ties a different
idea or concept to this catchword. For many, it represents a control instrument based on
standard personal computers to store, evaluate, and represent test data. According to this
notion, data is acquired through special measuring devices attached to a personal
computer over a serial or parallel cable. Some thinks that it means a computer equipped
with application and driver software and a built in transmitter as sort of low-cost
alternative to relatively expensive standard alone measuring devices. Both ideas are
correct, but only up to a certain point. They cover only part of this concept. Before
discussing the exact definition, we will describe the principle types of computer-assisted
test data acquisition [13].
2
Test data can be acquired in a computer in a different ways. It is important to
understand the underlying architecture of a measuring device. A traditional measuring
device always consists of three components, shown in figure 1.1, which perform the
following tasks [13]:
ƒ
Acquire the measurement parameters (data acquisition)
ƒ
Adapt and process the measured signal (analysis)
ƒ
Output the measured value (presentation)
Figure 1.1: Structure of a measuring device.
Such measuring devices generally have fixed functions defined by the
manufacturer, and they are characterized by a manufacturer-specific architecture and an
inflexible user interface. Consequently, they cannot be adapted easily to changing needs.
3
Both the operation and the documentation are entirely manual. To add long measuring
sequences that require constant changes of settings, a large amount of time is used to set
the measuring devices and to document the measured values.
1.2
Virtual Instruments
Based on this background information, we are now able to define the term
"virtual instrument" in a more accurate way. We speak about a virtual instrument when
we create measuring systems composed of a standard personal computer, suitable
software, and appropriate measuring hardware tailored to the measuring task, which is
normally available only in specifically designed stand-alone measuring devices. Virtual
instruments represent a visualization and centralization of complex measurement
systems on a standard personal computer in the form of a virtual user interface [13]. The
user sees a uniform, comprehensive single system, i.e., a complete application,
consisting of many individual measuring components. This fundamental concept is the
quantum leap from the conventional measuring device over computer-assisted
measurement technologies to adaptable virtual measuring systems. This represents a
shift from manufacturer-defined measuring devices to user-defined measuring systems.
The main benefits of this concept are:
ƒ
A virtual instrument can contain any combination of industry standard hardware
to acquire or output data: IEEE-488.2, RS-232 devices, VXI/MXI systems, field
buses (CAN, Interbus-S, Profibus, Foundation Fieldbus, Axiom etc.),
multifunction plug-in cards, DAQ instruments, image processing components,
external black-box systems, or motion control. figure 1.2 illustrates the hardware
architecture.
4
ƒ
The capabilities to analyze and represent measured data reach far beyond the
boundaries of conventional measurement technology.
ƒ
A powerful software development environment and a set of hardware
components allow creation of a number of virtual instruments to cover a wide
range of test functions and applications.
Figure 1.2: Hardware architecture of virtual instruments.
The spectrum of virtual instruments used in practice includes classical laboratory
automation, process visualization and control, automotive and aviation industries,
medical applications, manufacturing industry, and research and science.
5
1.3
Project Objectives
The main objective of this project is to develop software with Graphical User
Interface (GUI) capabilities using Microsoft Visual Basic version 6.0 in performing their
position control system experiment. The GUI will be developed with the following
goals:
1)
To developed controller using software for position control system.
2)
To ease students in performing position control system experiment.
3)
Expose students on position control system theoretically and practically.
4)
Enable students to understand more on application that is based on
position control.
5)
Teaching electrical engineering students real-world data acquisition and
analysis.
6)
Allowing students manipulate the information with their graphics and
word processing packages to generate professional quality lab reports.
7)
Empowering students to acquire experimental data in this laboratory
experiment directly from the software.
8)
Permits students to meaningfully experiment with physical relationship in
a readily interpretable, graphics format.
Besides that, the objectives of each experiment that have to be performed by
student are basically divided into 4 classical control experiments listed as below:
(i)
Experiment 1 – Closed-loop Position Control System.
Objective:
To demonstrate a simple motor-driven closed-loop
position control system.
6
(ii)
Experiment 2 – Effect of Gain Changes On Deadband.
Objective:
To study on the deadband of a position control system and
the effect of gain changes upon the Deadband.
(iii)
Experiment 3 – Unit Step Response To Gain Changes.
Objective:
Visual study of step response to gain changes upon
position control system.
(iv)
Experiment 4 – Velocity Feedback.
Objective:
Study on the effect of velocity feedback in a position
control system.
Besides these experiments, an additional experiment using Direct Digital Control
(DDC) techniques will be implementing in this software to allow student to observe
system response behaviour upon changes on the proportional gain, integral gain and
derivative gain. The objective of these experiments is to enable student to differentiate
between Direct Digital Control and Classical Control. Hence, this will allow student to
control the position of the control system in real-time with a PC-based system and aware
them of the important role, played by personal computers in engineering nowadays.
Based on the above experiment, it also allow student to differentiate between
Supervisory Control and Direct Digital Control where in Supervisory Control,
computers are used for monitoring purpose only but in Direct Digital Control, control
loop is replaced by computers to interface with process measurement and control some
of the physical process parameters.
7
1.4
Project Scope
The scope of the project is listed below:
i.
To apply direct digital control (DDC) in position control system.
ii.
To develop a PID and a fuzzy logic controller which will be use to
control the output response.
iii.
To make comparison of the performance between PID controller and
fuzzy logic controller.
iv.
To developed Graphical User Interface to help the student to visualize
and analyze the system. Besides, it also covers the method on enhancing
the original experiment for student so that they can finish their
experiment on time and in the mean time, they can grab more
understanding through visual presentation.
CHAPTER 2
LITERATURE REVIEW
2.1
Introduction
The limits of my language mean the limits of my world.
-Ludwig Wittgenstein
Tractatus Logico-Philosophicus [13]
The evolution of programming paradigms was marked by the transition from
machine language to assembly languages, and from these to functional, procedural,
and, eventually, object- oriented, high-level languages. During these past years, a
totally new paradigm evolved along these lines, offering problem solving approaches
in a new dimension graphical data flow programming in the form of so called
"virtual instruments". Where, virtual instrumentation is defined as a visualization and
centralization of complex measurement system on a standard personal computer in
the form of a virtual user interface. A user sees a uniform comprehensive single
system, consisting of many individual measuring components in a VI.
9
There has been an increasing awareness in the importance of virtual
instrumentation in control engineering especially in position control system. Virtual
instrumentation in control and measurement system can reduce hardware drastically
thus lowering the cost. The Virtual Instrument approach addresses several of these
user equipments. Significant strides in development of application software
technology has made it possible to offer users intuitive, graphics or code based
programming features and state of the art connectivity. These factors have
dramatically eased the user’s pain in putting together a VI. The need for integrated
test products and suited in automated measurement environments is another factor
influencing the use of VIs in electronics manufacturing, data acquisition, automotive,
and laboratory research application.
It is evident form the structure of VIs that software plays a vital role in
defining the utility, control, access and applications of the VI. It seems that the
virtual instrumentation industry is on the launch pad to a new world of user friendly,
open architecture instrumentation. Vendors and users realize that VIs is not expected
to replace all bench top instrument applications in the future. However, it is certainly
gaining grounds in several applications in the electronics, telecommunications,
semiconductor, and automotive industries previously considered inaccessible for
computer based measurements.
2.2
Previous Research
Previous works have been done on real-time system and virtual
instrumentation. These researches that are relevant to this project will be discussed
next to demonstrate continuity from previous researches.
10
2.2.1
“Motor Drive Control System for Education and Research”
Jee Ming Lee and Jung-Tang Huang (1995) developed research on “Motor
Drive Control System For Education and Research”. An industrial level real time
motor control education tool has been designed for the enhancement of the related
basic conception between control theory and physical system. The educational tool
set presents a life of control methodology in a servo motor system. The set of DC
motor system is assembled as a physical experiment control system. The learning
process provides a friendly and humanized environment. The multitasking operating
module is adopted to enable the educational control package to swap the
environment flexibly. The real time control module accurately analyzes the motor
position control, which a set of experiment curriculum is designed to apply basic
automatic control course [5].
2.2.2
“Development of a Distant Laboratory using Labview”
K.K.Tan, T.H. Lee and F.M.Leu (1999) have conducted a research on how to
“Development of a Distant Laboratory using Labview”. This paper presents the
development of a distant laboratory experiment for engineering students. The
experiment is concerned with real-time modeling and control of a pilot-scale DC
servomotor. The World-Wide-Web (WWW), with its ability to deliver multimedia
and graphics across the world, is used as the medium to link the student to a remote
experimental setup located in the distant laboratory. This paper describes the
hardware and software constructs, which facilitate students to perform the
experiment remotely and efficiently, using only a commonly available, user-friendly
web browser [7].
11
2.2.3
“Real-Time Position Control of Free-Electron Laser Beams: Theory and
Experiments”
M.Moallen (2000) has done a research on “Real-Time Position Control of
Free-Electron Laser Beams: Theory and Experiments”. This paper addresses
feedback control of electron beam positions in Duke University Free-Electron Laser
Facility. The input-output relationship for this system is algebraic and the system is
subject to noise and disturbance inputs. The information about beam position is
obtained by sensory device called Beam Position Monitors distributed throughout the
storage ring. In contrast, the beam position profile is controlled by passing currents
through the coils of correction magnets distributed around the ring. A desired control
objective is to move the stored electron beam to specific locations in the vacuum
chamber and to stabilize the orbit in the presence of disturbances. The technical
details of implementing the data acquisition and control scheme and their integration
with the existing control system modules are presented along with experimental
results [10].
2.2.4
“Closed-loop Position Control System using Labview”
Kelvin R. Aaron, Noreen L. Foster, Danielle P.Hazel, A.M.Hasanul Basher
(2002) has designed a “Closed-loop Position Control System using Labview”. A
closed-loop DC motor control system is developed using National Instrument's (NI)
Data Acquisition (DAQ) Board (Model MIO 6040E), LabVIEW software package,
and DAQ Signal Accessory Board for smooth and accurate positioning. The motor
can be rotated either clockwise (CW) or counter clockwise (CCW) as needed. A
linear feedback that employs a potentiometer provides the current position of the
motor. The LabVIEW software is used to develop virtual instrument (VI) that
includes a front panel and a functional diagram. The VI reads the desired position of
the motor entered by user on the front panel and the current motor position via
12
feedback, and determines its amount and direction of rotation. A smooth cubic
polynomial that is continuous and has a continuous first and second derivative is
utilized to represent the angular position of the motor. The VI uses the polynomial to
generate a velocity profile, which in turn is converted into a voltage profile for the
motor armature through a conversion formula. This conversion formula is
determined through the use of motor speed voltage characteristics obtained from
laboratory experiments. Reading the motor position and sending voltage to motor
circuit are accomplished via the DAQ's input and output ports respectively. The
system is tested in the laboratory for different angular displacements and initial
positions of the motor [8].
2.3
Virtual Instrumentation in Position Control System
Mainly the goal of this research is to very effectively fill the gap between
textbooks, which work on theoretical manipulation, and physical systems, which
work on the performance running. Essential features of an educational automatic
control system program in control engineering will be developed. These experiment
contents of an undergraduate engineering course in control system analysis and
designed are greatly enhanced.
It is the aim of this research to develop a user-friendly graphical user
interface, which is easy to navigate. Thus providing a high-quality GUI that
improves usability, end-user perception and marketability.
CHAPTER 3
RESEARCH METHODOLOGY
3.1
Project Overview
The main purpose of this project is to design a software based
program in controlling the position of a servomechanism or also known as
servomotor. It involves understanding on how to interface personal computer (PC)
with position control system through Data Acquisition System (DAS). This project
can be divided into two parts namely hardware and software. Hardware parts involve
more in interfacing MS150 Modular servo System and Data Acquisition System with
a personal computer. While the software part include programming real-time
software using Microsoft Visual Basic 6.0. In this chapter, three main topics will be
discussed. We will discuss on:
i.
Basic concept of control system:
¾
Position Control System
¾
Direct Digital Control (DDC) system
¾
PID controller
¾
Fuzzy Logic Controller
14
ii.
iii.
Software Review:
¾
Virtual instrumentation (VI)
¾
Graphical User Interface (GUI)
¾
Microsoft Visual Basic 6.0 software
Hardware Review:
¾
Data Acquisition System (DAS)
¾
MS150 Modular Servo System
In the earlier stage, literature review and experiment are performed manually
to understand the concept of the controller. Interactive software will be designed
using Microsoft Visual Basic 6.0. The software will be equipped with a set of
graphic instructions to ease any mistake when performing experiment. Finally, the
software will be integrated with hardware to produce a GUI position control system.
In designing digital controller in controlling the position control system, each
theory and concept play an important part in understanding the operation of position
control system. In contrast, understanding on the concept of Direct Digital Control
(DDC), Data Acquisition System (DAS), Proportional, Integral and Derivative (PID)
controller and fuzzy logic controller also play an important role in succeeding this
project. The following topic will be described briefly on each concept and theory.
15
3.2
Position Control System
In a position control system, a common requirement is for a motor to rotate
an output shaft to the same angle as an input shaft. The general form of the block
diagram for closed loop position control system is shown in figure 3.1.
Correction
Op-amp
Input position
Potentiometer
Output position
of load
Propotional
Control
+
Sum
-
Servoamplifier
Motor
Gear &
Screw
Load
Voltage
Feedback
related to actual
position
Measurement
Potentiometer
Figure 3.1: general form of the block diagram for closed loop position control
system.
In a simple position control system, an error signal proportional to the
difference between input of set position and actual position is generated. The error
signals then fed into controller. In the figure the controller will be pure gain
controller. The output of the controller drives the motor closer to the desired position.
The overall of position control system is likely second order system. This mean when
there is step voltage input, the output will take time to rise to the required output and
likely oscillate about it for a while before settling down to steady state value. We can
think of the situation like this. The voltage step input causes the motor shaft to
accelerate and change position of the load, and because of time delays inherent in the
system there is a lag before the position of the load is fed back to the summing
amplifier, so the load overshoots its position and oscillates about the required
position.
16
Figure 3.2 shows the typical form of the response of an under damped system
to the step input [12]. Certain terms are used to specify such a performance:
•
Rise time, Tr
Define as the time taken for response from 10% final value to the 90% of
final value. It shows how fast a system response to input.
•
Peak time, Tp
Define as the time taken for the response to rise from 0 to the first peak value
or the maximum value that the system can achieve.
•
Overshoot, Mp
Define as the maximum amount by which response overshoots the steady
state value or final value.
•
Settling time, Ts
Define as the time taken for the response to fall within and remain within
some specified percentage e.g. 2% of the steady state va1ue. It used as a
measure of the time taken for the oscillation to die away.
17
Figure 3.2 Standard measurements on Step Response of a Control System
An alternative and more common way of achieving the same effect is to use a
second feedback, loop which gives measurement related to the rate at which the
position of load is changing. This called velocity feedback. Figure 3.3 shows how
single position loop diagram can be modified to include velocity feedback.
Summing
amplifier
Input position
+
+
-
-
Rotational
Output
Servoamplifier
Motor
Output
position
Gear &
Screw
Load
Voltage
Feedback of
speed,
stabilizing
feedback
Speed
measurement
Feedback of position,
monitoring feedback
Measurement
Potentiometer
Figure 3.3: single position loop diagram can be modified to include velocity
feedback.
18
The overall system performance will improve by changing type of feedback.
Overshoot is caused when the motor shaft is going too fast to be stopped suddenly
due to inertia. Thus by negatively feedback some signal proportional to a shaft
velocity, we can counteract the effect of inertia. This type of feedback is
appropriately called velocity feedback. The most common way to achieve this is to
use the tachometer signal, which is proportional to the velocity, in some minor
feedback loop hence the name minor loop feedback. Note that to obtain minor loop
feedback control system, include velocity feedback by means of tachometer and
another controller to the simple position control system. Also note that the feedback
signal is subtracted from the error signal, as required for stability.
For the position control the input and output, transducers must measure input
and output shaft angles and produces a control signal (or known as error)
proportional to the angle between the shafts. In this project, a simple form of error
channel is to mount a potentiometer onto each shaft and then added together the
voltages from the potentiometer sliders with an operational amplifier shown in figure
3.4 [11].
Figure 3.4: Simple Error Channel
19
If the two shaft angles θi, θo are equal, the two slider-voltages are equal and
opposite and cancel at the amplifier output. Any misalignment will give an error
signal proportional to the misalignment (θi - θo), which can operate the forward path.
An important factor in an error channel is the error factor Ke, which gives the V/° (or
possibly radian) of misalignment.
In order that the closed-loop system output shaft follows the input shaft
movement correctly. It is necessary that the polarity of the various connections be
such that an error signal causes the motor to rotate in a direction to reduce the error
and not increase it [11].
20
3.2.1
General Servomotor System Model
In this section, the model of a servo system (also known as positioning
system) is derived. A simple DC motor with its schematic is shown in figure 3.5
[12].
(c)
Figure 3.5: DC-Motor (a) Wiring Diagram (b) Sketch (c) Schematic
The motor is armature controlled DC motor with a constant field. The motor
has an armature resistance, Ra and inductance, La, respectively. A magnetic filed is
developed by stationary permanent magnets or a stationary electromagnet called the
21
fixed field. A rotating circuit called the armature through which current, i(t) flows,
passes through this magnetic field at the right angle and produces force,
Force, F = Bl i (t )
(3.1)
Where, i(t) is the current passing through the armature circuit, l is the length
of conductor and B is the magnetic field strength. The force produced resulting
torque that turns the rotor (the rotating member of the motor). There is another
phenomenon that occurs in the motor, a conductor moving at the right angles to a
magnetic field generates a voltage at the terminal of the conductor,
Voltage, e = Bl v
(3.2)
Where v is the velocity of the conductor normal to the magnetic field. Since
the current carrying armature is rotating in a magnetic field, its back electromotive
force (or back emf) is proportional speed. (Bishop R.H. Dorf R.C, 1998) Thus,
Back emf, vb (t ) = K b
dθ (t )
dt
From the above equation, Kb is the motor dependent constant,
(3.3)
dθ (t )
is the
dt
angular velocity of motor and θ (t ) is the shaft position.
The typical equivalent mechanical loading on a motor that connected to the
motor shaft including total moment of inertia, J and total viscous friction, B. Assume
that τ(t) is the torque developed by the motor, then writing the following equation
22
Torque, τ (t ) = J
dθ (t )
d 2θ (t )
+B
2
dt
dt
(3.4)
The developed torque for this motor can be given by
Torque, τ (t ) = K T i (t )
(3.5)
With i(t) is the armature current and KT is the motor torque constant.
The equivalent equation of the armature inductance is relatively small
compared to the armature resistance and can be ignored [12], thus we assume the
inductance La can be ignored, which in the case for many servomotors [2], then
v(t ) = i (t ) Ra + vb (t )
(3.6)
i (t ) =
v(t ) − vb (t )
Ra
(3.7)
i (t ) =
v(t ) K b dθ (t )
−
Ra
Ra dt
(3.8)
τ (t ) = K T i(t )
τ (t ) =
J
K K dθ (t )
KT
dθ (t )
d 2θ
=J 2 +B
v(t ) − b T
Ra
dt
Ra
dt
dt
d 2θ (t ) ⎛ BRa + K b K T
+ ⎜⎜
Ra
dt 2
⎝
⎞ dθ (t ) K T
⎟⎟
=
v(t )
Ra
⎠ dt
(3.9)
(3.10)
23
If the armature inductance cannot be neglected, then the model is third order
system. By using Laplace transform and solving the transfer function becomes [11]
[12],
θ ( s)
V (s)
KT
= Gp( s ) =
Ra
⎛ BR + KT K b ⎞
⎟⎟ s
Js 2 + ⎜⎜ a
Ra
⎝
⎠
⎛ Kt ⎞
⎜⎜
⎟
Ra J m ⎟⎠
θ m (s)
⎝
=
∴
E a ( s)
⎡
KK
1 ⎛
⎜⎜ Dm + t b
s ⎢s +
Ra
Jm ⎝
⎣
⎞⎤
⎟⎟⎥
⎠⎦
(3.11)
(3.12)
where;
J m = 30 × 10 −6 kgm 2
K b = 60 × 10 −3 Vsrad −1
K t = 17 × 10 −3 NmA −1
Ra = 3.2Ω
Dm = small (can' t be quote )
The value of Jm, Kb, Kt, Ra and Dm is given in the MS150 modular servo
system. Thus when substitute these values into equation (3.12) we get the plant
transfer function in equation (3.13).
G(s) =
(177.0833)
s[s + 10.625]
(3.13)
24
3.2.2
Direct Digital Control (DDC)
Direct Digital Control (DDC) technology gives today’s industrial user the
flexibility to tailor near any system to specific needs using convenient, softwarebased control. A simple concept that has blossomed into a complex array of
components and systems, DDC is basically a microprocessor-based technology in
which a controller performs closed-loop functions via sophisticated algorithms and
strategies. A typical schematic of direct digital control is shown in figure 3.6.
Computer
R(s),
Input
position
C(s),
Output
position
Control Algorithm
DAC
Final
Control
Element
ADC
Servomotor
Plant
Analogue
feedback
transducer
Digital
feedback
transducer
Figure 3.6: Direct Digital Control Schematic.
The controller function handles inputs and outputs electronically while the
software provides the logic. Within the controller, the analogue signal enters the
system where it is converted to a digital form so that the data can be acted upon
according to a set of parameters or formulas. The result is the execution of
commands directly to end device. Because the programs used by the controllers are
in digital form. The operations are known as direct digital control [3] [4].
DDC control brings speed, precision, flexibility to control functions at low
cost. It can range from the simple control of a single loop to the application of larger
systems controlling a multitude of loops. Controllers may function as stand-alone
units or be networked into systems using a PC as a host to provide additional
25
functions. Most intelligent DDC devices residing on a network today are capable of
sophisticated equipment control and remote communication.
3.2.3
PID controller
PID stands for Proportional-Integral-Derivative. This is a type of feedback
controller whose output, a control variable (CV), is generally based on the error
between some user-defined set point (SP) and some measured process variable (PV).
Each element of the PID controller refers to a particular action taken on the error:
There are many situations that require some type of servo-control system.
This section reviews the fundamental of PID controllers and presents detailed
simulations or design for development of digital servomotor controller. PID
controllers are commonly used to regulate the time-domain behaviour of many
different types of dynamic plants. These controllers are extremely popular because
they can usually provide good closed loop response characteristics, can be tuned
using relatively simple rules and are easy to construct using either analogue or digital
components. Consider the feedback system show in figure 3.7 where we assume that
the plan is a DC motor whose shaft position must be accurately regulated.
26
R(s),
Input
position
C(s),
Output
position
PID
controller
+
-
-
DC
ServoMotor
plant
System
dynamic
Figure 3.7: General Feedback Architecture
The PID controller K(s) is placed in the forward path, so that its output
becomes the voltage applied to the motor’s armature. The feedback signal is either an
angular shaft position or velocity, measured by a potentiometer or a tachometer,
respectively. In the block diagram, these transducer dynamics are in the feedback
path. The output position signal y(t) (or velocity signal) is summed with a reference
signal r(t), or command signal, to form the error signal e(t). Finally, the error signal
is the input to the PID controller.
The concept for this closed-loop system is simple. The sensor, which is
attached to the motor shaft, provides a voltage that is compared to the reference
voltage r(t) at the summer. When this error signal is non-zero, there will be an input
to the controller, and hence some action taken by the DC motor. Once the sensor
signal is equal to the reference signal, there is no input to the controller and no
voltage applied to the motor, causing the motor to stop. However, this simple
explanation does not provide us a with a method whereby the motor position can be
brought to an exact position (with absolute accuracy) and does not tell us how to
make the motor perform this positioning task as quickly as possible.
27
3.2.3.1 Analog PID controller
The PID controller is defined by the following relationship between the
controller input e(t) and the controller output v(t) that is applied to the motor
armature:
r
v(t ) = K p E (t ) + K I ∫ e(τ )dτ + K D
0
de(t )
dt
(3.14)
Taking the Laplace transform of this equation gives the transfer function of
K(s):
K ( s) =
K
V (s)
= ( K P + I + K D s)
E (s)
s
(3.15)
This transfer function clearly illustrates the proportional, integral, and
derivative gains that make up the PID compensator. Select new definitions for the
gain terms according to;
KP = K
(3.16)
K
TI
(3.17)
KI =
K D = KTD
(3.18)
28
Then, the transfer function can be expressed to easily show that the PID
controller leads to a pole at the origin of the Laplace plane and design freedom over
two zeros:
K ( s) =
KTD 2 1
1
(s +
)
s+
s
TD
TI TD
(3.18)
The block diagram for the PID controller can be sketched in several different
ways. One possible block diagram is shown in figure 3.8.
PID controller
Ki/s
Kds
R(s),
Input
position
+
+
-
Kp
+
-
-
DC
ServoMotor
plant
C(s),
Output
position
+
Figure 3.8: Block diagram for PID Controller
Proportional mode is the heart of a closed-loop control system. It provides an
output, which is directly proportional to the error signal. It is the only control mode,
which can be used alone. Integral and derivative modes can only be used when
combined with at least the proportional mode. Proportional control amplifies the
error to motivate the plant towards the desired response. It can reduce but not
eliminate stead-state error. Proportional control can also produce excessive overshoot
and oscillation. Remember that by increasing K will increase the overshoot of the
29
system, but it will decrease the system rise time. If the process response to a change
in set point is not sufficient and a large offset error persists then, perhaps the
proportional gain is too low. Increase the proportional gain till a satisfactory response
is obtained. On the other hand if the proportional gain is too high, then the
oscillations may be developed in the system.
Integral can eliminate steady state error. Usually small integral gains are used
to avoid destabilizing the closed-loop system. In many systems Ki is responsible for
driving error to zero, but to set Ki too high is to invite oscillation or instability or
integrator windup or actuator saturation.
Derivative Control acts on the slope of the error, thereby minimizing
overshoot. It provides "rate" feedback, resulting in more damping. High derivative
gains can increase trise and tsettling. It is difficult to realize in practice because
differentiation leads to "noisy" signals. In many systems Kd is responsible for system
response: too low and the process variable (PV) will oscillate; too high and the PV
will respond sluggishly. The designer should also note that derivative action
amplifies any noise in the error signal. Derivative mode is effective only at high
frequencies. It does not respond to any steady state error value but to acts swiftly
against the rate of change of error. This gives the derivative mode the ability to
respond to the upcoming changes by reacting to the rest in process output. For this
reason, derivative mode of control is also known as anticipatory control. Since the
controller output is dependent upon the slope of the error signal, it is not very
satisfactory when noise is present in measurement. For processes with very small
time constant, the derivative mode of control can be troublesome.
The general characteristic of a PID controller is that a proportional controller
(Kp) will have the effect of reducing the rise time and will reduce, but never
eliminate, the steady-state error. An integral control (KI) will have the effect of
eliminating the steady-state error, but it may make the transient response worse. A
derivative control (KD) will have the effect of increasing the stability of the system,
30
reducing the overshoot, and improving the transient response. Effects of each of
controllers Kp, KD, and KI on a closed-loop system are summarized in the table 3.1.
Table 3.1 Effect of Each PID Controller
Closed Loop
Rise Time
Overshoot
Settling Time
Response
Steady-state
Error
Kp
Decrease
Increase
Small change
Decrease
KI
Decrease
Increase
Increase
Eliminate
KD
Small Change
Decrease
Decrease
Small change
Note that these correlations may not be exactly accurate, because KP, KI, and
KD are dependent of each other. In fact, changing one of these variables can change
the effect of the other two.
3.2.3.2 Digital PID controller
To discretise the controller, we need to approximate the integral and the
derivative terms to forms suitable for computation by a computer. By using ztransform, we may discretise the analogue PID controller to obtain the pulse transfer
function for the digital PID controller as the following equation [6].
Positional Form of the PID control scheme is given as ;
G(z ) =
where
C (z )
KI
= Kp +
+ K D (1 − z −1 )
1 − z −1
R(z )
(3.19)
31
K P = proportional gain = K −
K I = Integral gain =
K
KT
=K− I
2Ti
2
KT
Ti
KTd
T
K D = Derivative gain =
Equation (3.19) can be written in a sequence operation as in equation (3.20).
Equation u (k) will be used to develop the coding program for PID controller which
corresponding to sequence operation [16].
k
u (k ) = K p e(k ) + K i Ts ∑ e(n ) + K d Ts
−1
[e(k ) − e(k − 1)]
(3.20)
n =o
where Ts = time sampling = 1 ms
e(k ) = error between desired value and actual value
t
∑ e(k ) = total error
i =0
e(k ) − e(k − 1) = deviation error
The value of Kp, Ki & Kd was calculated using root locus technique. The tuning
method used is the Ziegler Nichols. The PID controller has been simulated using
matlab to ensure the manual calculation is correct. The value of Kp, Ki & Kd is as
follow.
K p = 1.266
K i = 0.1261
thus Ti =
Kp
Ki
= 10.0444 ≈ 10
K d = 0.0563
thus Td =
Kd
= 0.044 ≈ 1.0
Kp
32
After calculating these values, now we can write the coding for the PID
controller. The flow chart for the system can be virtualized as in figure 3.9.
Start
Read Input from
DAQ
Compute Error
Compute Proportional
action
Delay until
next sample
Compute Integral
action
Compute Derivative
action
Send output to DAQ
End
Figure 3.9.: Flow chart for the PID controller system.
33
3.2.4
Fuzzy Logic Controller
In this chapter, we show the design flow of the simulator and FLC for the
position control system in flow charts. The position control system simulator has
been developed using Microsoft visual basic 6.0 [15].
The steps that we are needed to do are:
Step 1: define quantization for state (input) variables, v and h and output
variable, force in universe of discourse.
Step 2: define a membership function for the input and control output.
Step 3: define the rules base and summarize them in an FAM table. The
values in the FAM table are the output control.
Step 4: define the initial condition
Step 5: conduct the simulation. For this part we use max-min configuration.
We defuzzify using centroid method.
Step 6: Compare the calculation with the simulation using matlab.
Figure 2.1 is the flow chart in designing the FLC. In developing the design,
we need to initialize and declare what are the inputs and outputs variables. Then,
define the math model so that we can carry out a simulation using the previous value
and the changes in the output. All the parameter is then used by FLC to be simulated.
The FLC design flow is shown in Figure 3.10. FLC is a subroutine to the main
simulator program. The FLC design methodology is shown in Figure 3.11.
34
Start
Initialization
&
Declaration
Plant/ Math Model
Fuzzy Logic Controller
FLC Subroutine:
1)
2)
3)
4)
Plotting
N
k=k+1
Yes
K =end
end
Figure 3.10 Flowchart of Simulator
Fuzzification
Knowledge Base
Inferencing
Defuzzification
35
Start
Study the System
-
-
determine the objective
identify process and
controller’s input & output
Fuzzification
-
quantize the input and output
variables
define the membership
function
Inferencing
-
-
Derive the fuzzy control
rules-based
Define the inference engine
define the membership
function
Defuzzification
-
Parameter Tuning
-
-
Chooses defuzzification
method function
Fuzzy Controller Operation
Mapping of
membership
function
Fuzzy inference rules
Scaling factor
-
Fuzzification
Fuzzy Inference
Defuzzification
Simulation & Testing
No
Performance OK?
Yes
End
Figure 3.11 FLC Design methodology
36
3.2.4.1 Fuzzification
The position control of the servo system configuration was designed based on
2 inputs and 1 output. Inputs for this controller are error and rate (de), and the motor
output. Fuzzification is where we define the quantization and membership function
for input variable, error and rate (de) and output variable, motor output in universe of
discourse. It involves the conversion of the input and output signals into a number of
fuzzy represented values (fuzzy sets).
Usually fuzzy sets are overlapping by about 25%. It can be observed that in
fuzzy logic control the transition from one fuzzy set to another provides a smooth
transition from one control action to another. If there is no overlapping in the fuzzy
sets then the control action would assemble bivalent control (transition from one
error region to another is rather abrupt). On the other hand if there is no overlapping,
there would be a lot of fuzziness and this blurs the distinction in the control action.
Figure 3.12 show the input and output variable that is used in the system.
Input variables
Output variable
Fuzzy logic
rule
Figure 3.12 show the input and output variable.
37
Quantization of input and output variables:
Input variables:
1)
•
Error(e)
Quantized into 3 fuzzy sets: negative error (Ne), zero error (Ze) and
positiveerror (Pe). Figure 3.13 show the error input variable.
Figure 3.13: The error input variable.
•
Rate(de)
Quantized into 3 fuzzy sets: negative rate (Nde), zero rate (Zde) and positive
rate (Pde). Figure 3.14 show the rate input variable.
Figure 3.14: The rate input variable.
38
Output variable:
2)
•
Motor output
Quantized into 3 fuzzy sets: negative output (No), zero output (Zo) and
positive output (Po). Figure 3.15 show the motor output variable.
Figure 3.15: The motor output variable.
3.2.4.2 Knowledge Base
The knowledge base of a fuzzy logic controller consists of a data base and a
rule base. The basic function of the data base is to provide the necessary information
for the proper functioning of the fuzzification module, the rule base and the
defuzzification module. This information includes the meaning of the linguistic
values of the membership functions of the process state and the control output
variables, physical domains and their normalized counterparts together with the
normalization, renormalization and scaling factors and the type of the membership
function of a fuzzy set.
39
The basic function of the rule base is to represent the expert knowledge in a
form of if-then rule structure. For this project, we derive the fuzzy logic rules based
on expert experience and control engineering knowledge. This method is the least
structured of the four methods and yet it is one of the most widely used today. It is
based on the derivation of rules from the experience based knowledge of the process
operator and/or control engineer. The fuzzy logic can than be derive into a 3 x 3 rules
which consist of 9 rule. Figure 3.16 show the fuzzy logic rules.
error
Ne
Ze
Pe
Nde
No
No
Zo
Zde
No
Zo
Po
de
Antecedent
Consequents
Pde
Zo
Po
Po
Figure 3.16: The fuzzy logic rules.
The fuzzy logic rules-based are:
1) If error is Ne AND de is Nde THEN output is No
2) If error is Ne AND de is Zde THEN output is No
3) If error is Ne AND de is Pde THEN output is Zo
4) If error is Ze AND de is Nde THEN output is No
5) If error is Ze AND de is Zde THEN output is Zo
6) If error is Ze AND de is Pde THEN output is Po
7) If error is Pe AND de is Nde THEN output is Zo
8) If error is Pe AND de is Zde THEN output is Po
9) If error is Pe AND de is Pde THEN output is Po
40
3.2.4.3 Inferencing
The inference mechanism provides the mechanism for invoking or referring
to the rule base such that the appropriate rules are fired. There are several
methodologies to derive the Inferencing method. Two most common methods used
in fuzzy logic control are the max-min composition and the max-(algebraic) product
composition. The inference or firing with this fuzzy relation is performed via the
operations between the fuzzified crisp input and the fuzzy relation representing the
meaning of the overall set of rules. As a result of the composition, one obtains the
fuzzy set describing the fuzzy value of the overall control output.
For this system, max-min composition is used for the Inferencing. Figure
3.17 shows the firing strength for the rule base using max-min inference.
41
Output of the
controller
Figure 3.17: The firing strength for the rule base using max-min inference.
3.2.4.4 Defuzzification
Defuzzification is a mapping from a space of fuzzy control actions defined
over an output universe of discourse into a space of non-fuzzy (crisp) control action.
This process is necessary because in many practical applications crisp control action
is required to actuate the plant. The defuzzifier also performs an output
denormalization if normalization is performed in the fuzzification module. There are
many methods of defuzzification that have been proposed in recent years.
42
For this system, centroid method is used for defuzzification. It is given by the
algebraic expression:
z* =
∑ u (z ) ⋅ z
∑ u (z )
z
(3.21)
z
3.3
Software Review
Software review is basically a discussion on the virtual instrumentation and
graphical user interface that will be developed in the project. Here, the software part
of the project will be discussed which is programming with Microsoft Visual Basic
[17] [18].
3.3.4
Virtual Instrumentation - Graphical User Interface (GUI)
Graphical User Interface (GUI) can be defined as an intuitive, easy-to-use
method of interfacing with computer program by means of graphical screen display.
It can resemble the front panel of instrumentations or other objects associated with a
computer program [9] [14].
43
3.3.4.2 General GUI Design Considerations.
The GUI design will reflects the user’s mental model rather than the
implementation model. Program features and functions support only required user
goals. No superfluous features have been added simply because they are easy to
implement, or as a result of the personal biases of the programmers. The GUI design
reflects the expected characteristics and capabilities of the user population - visual,
physical and cognitive capabilities; cultural and technical background; domain
experience; education level and so forth.
The GUI design reflects any unusual characteristics of the users’ environment
- dangerous or hazardous work areas, excessive noise, bright or dim lighting, and so
forth. The design is optimized for human perception and information processing
capabilities. For example, the “seven plus or minus 2” item limitations of the shortterm memory, and “recognition over recall” figure prominently in the design choices.
All user classes are adequately represented, and the potentially diverse needs of all
user classes are balanced appropriately.
All GUI items are prioritized. Critical and frequently accessed items are
prominent - and more easily accessed - than less important items. All unnecessary
GUI controls and indicators have been eliminated. The GUI hierarchy geometry is
optimized to reduce panel clutter, and to minimize the total number of clicks required
to access each function.
Panel layouts and GUI object placement logically reflect user task sequences.
In particular, the user is never required to jump between panels or applications to
complete a single task. Where possible, all necessary controls for each task are
accessible from a single panel. The user is not required to manually copy information
displayed on one panel into a control on another panel. The program automatically
44
transfers shared information between panels and eliminates unnecessary “busy work”
for users.
Tedious, mundane, and predictable tasks are automated to improve user
efficiency, but not at the expense of adequate user control. Qualified users are
permitted to modify or bypass automation as necessary. Tool tips, Control
Descriptions, and keyboard shortcuts have been included for power users.
User actions and task sequences lead naturally from one to the next.
The function of all GUI items is visually apparent, and all objects (including
custom controls) behave as expected. For example, GUI buttons look and respond
like real buttons from the physical world, and don’t produce unexpected outcomes.
GUI metaphors, visual or otherwise, are natural and consistent with their real-world
counterparts. The GUI adheres as necessary to applicable standards documents and
company-wide style guides.
Where measurable usability targets have been set, the final design meets or
exceeds these targets. Direct user feedback has been collected, and all design
shortcomings have been (or will be) addressed in subsequent iterative cycles.
Documentation and “Help” files have been prepared with the same level of
enthusiasm and attention to detail as the software.
45
3.3.1.2 Visual Presentation
Panels supposed to immediately appear uncluttered and organized. The style
and mood of the GUI design is appropriate for the application and its expected users.
The design style is consistent from one panel to the next, and all panels look as
though they belong to the same application. A limited number of unique design
elements - control and indicator types, colors, fonts, proportions, and so forth - are
used consistently and thematically, thereby promoting a sense of application-wide
unity. Limited color palettes have been selected and applied consistently to enhance
both aesthetics and mental model development. Bright colors are used sparingly for
highlights, or to attract the user’s attention.
The layout of each panel creates a visual hierarchy, drawing users’ eyes to the
most important items first. GUI items are arranged and ordered to reflect natural
visual scanning patterns (left to right, top to bottom in Western cultures). Task
sequences are mapped directly to the natural scanning patterns to improve user
productivity. Positive and negative space has been used effectively to make panel
designs appear balanced and uncluttered. GUI text, control labels, and menu text is
clear, descriptive, and concise. Controls have been grouped to permit the elimination
of redundant label text. Error messages are brief, informative, and designed to help
users locate and overcome difficulties. Unnecessary technical jargon has been
eliminated from panels and dialog boxes. The selected font style, size, and color
combinations provide adequate readability for users with common visual
deficiencies.
Where panel resizing is permitted, panels have been designed to resize
gracefully. Bitmapped graphics have been avoided on resizing panels, and the
number of decoration elements has been kept to a minimum. GUI panels targeted for
cross-platform deployment have been verified visually on all target platforms. Panels
have been designed to provide a pleasing visual presentation, but form always
follows function.
46
3.3.1.3 VI Diagram Considerations
The VI architecture supports the most logical user task sequences - user
efficiency was not sacrificed in order to simplify the VI architecture or wiring. GUI
objects are divided between panels based on priority and task requirements. GUI
organization was not permitted to evolve simply as a by-product of dataflow
considerations, the application architecture, or the organization of the sub VI
hierarchy. Potential dataflow limitations have been overcome to provide continuous,
real-time GUI feedback in response to user actions. Modal GUI architectures have
been avoided unless the underlying user task sequences are also modal. Event-driven
architectures with numerous simultaneous options are favored to promote exploration
and enhance mental model development.
Cancel and programmatic Undo functions are included to promote
exploration and enhance GUI forgiveness. Presets, default values, and recently
accessed item menus are included to alleviate repetitive tasks. Similarly, state
information is preserved from one run (or sub VI call) to the next; thereby
minimizing the time users spend resetting configuration parameters.
3.3.1.4 Additional GUI Object Considerations
Custom menus are included to reduce panel clutter while providing nonmodal access to numerous functions and parameters. Menu items accurately and
concisely describe the functions they reference. Menu items and other GUI objects
are disabled (grayed-out) when the functions they invoke are not compatible with the
current operational mode. Tab control implementations are consistent with the
underlying file folder metaphor. In non-standard implementations, the tabs are
hidden. Nested tabs are avoided. ActiveX objects, DLLs, CINs, and other platform-
47
specific extensions have been used only when a 100% G solution is impractical or
impossible.
3.3.2
Microsoft Visual Basic
The software that will be used in the project is Microsoft Visual Basic
Version 6.0. The Microsoft Visual Basic is the most productive tool for creating
high-performance components and applications. Visual Basic 6.0 offers developers
the ability to create robust applications that reside on the client or server, or operate
in a distributed n-tier environment. Visual Basic 6.0 is the Rapid Application
Development (RAD) tool available either as a stand-alone product or as a part of the
Visual Studio 6.0 suite of tools [17] [18].
Visual Basic 6.0 provides a complete set of tools for integrating databases
with any application. It visually creates reusable record set command objects with
drag-and-drop functionality. Bind to multiple data sources for data aggregation and
manipulation. By using the new Data Environment designer, developers can quickly
drag-and-drop custom data-bound controls to create forms; they can create reports
with the new Data Report Designer. Creation of custom data hierarchies is as easy as
filling out a dialog box and dragging the command to the form. It can also develop
multimedia-rich applications using the document object model and Dynamic HTML
surface as the user interface. The Uses of new String Functions and features such as
"Retain in Memory" can be able to create faster, more scalable applications and
components.
Microsoft VB6 (Visual Basic 6) represents a significant leap forward in some
aspects, particularly database programming and connectivity and Internet
programming, both for IIS on the server-side and for Internet Explorer 4.x on the
48
client-side. VB6 is definitely a robust package and, with over 3.2 million developers
using it, it remains one of the most popular development languages for Windows.
There are tens of thousands of ActiveX controls from third party developers that can
be implemented in the projects, and which is remarkably easy to create ActiveX
controllers.
The visual basic can also share code with Labview programming. While
some programming projects begin with a blank slate, many others would be greatly
expedited if existing code could be reused. This is especially true when the new
solution is to be developed in a different environment. This is not to say that it is
necessary to develop components of a LabVIEW application in another language.
LabVIEW is a rich language that is designed for rapid development and can be used
to create an application from the ground up.
3.4 Hardware Review
Generally, two sets of hardware components used in this project, one the
MS150 Modular Servo System and Data Acquisition System (DAS). The Data
Acquisition System (DAS) consists of AX5412H PC Bus Data Acquisition Card and
control board.
3.4.1
Data Acquisition System (DAS)
Data acquisition is defined as collecting and measuring electrical signals from
sensors, transducers, and test probes or fixtures and inputting them to a computer for
49
processing. Sometimes, it is defined as collecting and measuring the same kind of
electrical signals with A/D or digital input/output (DIO) boards plugged into a PC,
and possibly generating control signals with D/A or DIO boards in the same PC.
Many applications use plug-in boards to acquire data and transfer it directly
to computer memory. Others use DAS hardware remote from the PC that is coupled
via parallel or serial port. A PC-based DAQ system depends on each of the following
system elements:
i)
Sensors and Transducers
ii)
Signal Conditioning
iii)
DAQ Hardware
iv)
Personal Computer
v)
Software
Typically a computerized data acquisition and control system is constructed
as figure 3.18.
Figure 3.18: Computerized data acquisition and control system.
The following sessions will gives an overview of each element and explains
the most important criteria of each element of a PC-based DAS.
50
3.4.1.1 DAQ Hardware
The AX5412 data acquisition board is used in this project as an interface
between computer and the servomotor plant. Accompanied with the AX5412X is the
AX750 board [1]. It is a screw terminal panel with standoffs to allow tabletop
operation. The Axiom Data Acquisition Card Driver is designed to support all
Axiom’s DAS card for ease of use under Windows. Axiom’s driver includes a
standard DAS API (Application program Interface) and some device-specific drivers.
While programming, we only need to work with standard DAS API. A general view
of AX5412H data acquisition board is shown in figure 3.19.
Figure 3.19: A picture of AX5412H Data Acquisition Card
Some of the AXIOM Data Acquisition Card features are:
ƒ
32-bit DLL.
ƒ
Supported compilers: Visual C/C++, Borland C/C++, Visual Basic, Delphi
ƒ
Support AXIOM data acquisition hardware [1].
ƒ
Easy to use
51
ƒ
Implemented Functions:
¾ Analog Input via software trigger
¾ Analog Input via interrupt
¾ Analog Input via DMA
¾ Digital Input
¾ Digital Output
¾ Temperature Measurement
¾ Counter/Timer
The AX750 is a screw terminal panel with standoffs to allow tabletop
operation. It provides easy connection, via 80 barrier strip connections, to all analog
and digital I/O signals to the AXIOM DA&C series boards. The AX750 includes a
one-meter long flat ribbon cable connector for direct connection to any AXIOM
DA&C board. The AX750T, consists of AX750 and an additional thermocouple
cold-junction compensation circuit, is designed to be used with AX5611-HG or
AX5412-HG, for accepting various type of thermocouple input. The AX750 also
provides features to accommodate the requirement for current loop input operation
with a 0 to 20mA full-scale input range. By installing 250½ current shunt resistors,
the user can convert the 0 to 20mA input range to a 0 to 5V range. Figure 3.20 shows
the AX750 screw terminal panel.
52
Figure 3.20: AX750 screw terminal panel
The output from the servomotor plant i.e. the input voltage that is in analog
signal from the servo amplifier will be converted to digital signal before the
computer can process the data. The AX5412 board Analog to Digital Converter
(A/D) with 12 bit AD resolution for a input range from 0V to 10V, 0.0025 Volt
precision can be achieved, which it is good enough for the control purpose. There are
16 channels for single-ended & 8 for differential analogue input.
After the computer has calculated and processes the data, the controller signal
has to be sent back to the servomotor plant. The data acquisition board with 2
channels of 12-bit Digital to Analog (D/A) output will convert the digital signal data
to analog signal of the relevant voltage level. All the connection can be done easily
with the AX750 screw panel connected to the output parallel port of the AX5412
boards.
53
3.4.1.2 Analog Input system
The basic function of analog input system (A/D system) is to convert the
analog input signal to corresponding digital format that the computer can be read. In
order to benefit A/D system’s stability and obtain good performance, several other
additional parts such as multiplexed, amplifiers, sample and hold are required. Figure
3.21 exhibits many input signals come into the A/D system.
Figure 3.21: Many input signals come into the A/D system
A/D converter has been developed for different application needs. The most
widely used A/D converters are integrating and successive approximation.
Integrating type has the advantage of high noise rejection but its speed is lower. Most
data acquisition system needs at least 12-bit resolution to recognize the measured
signal. Most high resolution A/D will elapse more time and normally the price is
higher.
54
Multiplexed is a device containing a group of switched to select exact signal
to go through A/D converter. Because the amplifier and A/D ate shared, the cost will
be lower, but the channel’s acquisition speed will be reduced too.
Sample and hole (S/H) is designed for A/D converter to keep track of input
signals. When a high speed A/D converter is used, such as successive approximation,
it requires a no changing input signal during the conversion stage. The S/H circuit
can keep the input signal with a constant level until the A/D is completed. Often the
limiting factor in the application of the S/H is the uncertainty in the time actual
sample is taken – i.e. the “The aperture jitter” or Taj. The aperture jitter causes an
amplitude uncertainty for any input where the voltage is changing. The
approximation voltage error due to aperture jitter depends on the slew rate of the
signal at the sample point.
Amplifier is a device to enlarge the input small signal to a congruous signal
that A/D converter can resolvable. Notice if the input signal change is below the A/D
system’s minimum resolution, the system cannot recognize the deviation. The exact
gain is usually better the A/D resolution.
3.4.1.3 Analog Output System
Analog output (digital control analog output D/A) are typically used to
generate programmable level signal for yielding a loop control system Familiarly
analog output range are 0 – 5V, 4 – 20 mA and so on. They can be employed to
control such a dc power supply, frequency converter or to drive chart recorder, or as
in the project the servomotor system itself. Normally analog out merely provides
little power (voltage output is typically no more than 5mA), power amplifier or
current boosters are required attachment when large load is used.
55
3.4.1.4 D/A Converter
A digital to analog (D/A) converter affords an analog output proportional to a
digital number on the input. Most converters are 8 bits, 12 bits or even 16 bits, with
correspondingly higher resolution. Cost rises rapidly with resolution.
3.4.2
MS150 Modular Servo System
Feedback Instrument Ltd. manufactured this set of experiment learning tools
for the aim to help student to understand not only on position control system but also
speed control of a servo motor system. The MS150 Modular Servo System contains
few equipment such as the pre-amplifier unit (PA150C), Operational amplifier unit
(OA150A), input potentiometer (IP150H), output potentiometer (OP150K),
Attenuator unit (AU150B), DC motor (DCM150F), reduction gear tacho unit
(GT150X), servo amplifier unit (SA150D) and power supply unit (PS150E). Each
specification of the unit will be elaborate on the following session. A general view of
MS150 Modular Servo System is shown in figure 3.22.
Figure 3.22: General View of MS150 Modular Servo System.
56
3.4.2.1 Pre-amplifier Unit – PA150C
The pre-amplifier provides two output signals to the servo amplifier. One
output or the others goes positive to drive the motor in one direction or the other in
response to positive or negative input. The schematic of PA150C is shown in figure
3.23.
Figure 3.23 PA150C- Pre-amplifier Unit
3.4.2.2 Operational Amplifier Unit – OA150A
This OA150A functions as to amplify the input signal with either unity
feedback or external feedback determined by user. The schematic of OA150A is
shown in figure 3.24.
57
Figure 3.24 : OA150A – Operational Amplifier Unit
3.4.2.3 Input Potentiometer – IP150H & Output Potentiometer –OP150K
The input and output potentiometer is used to measure displacement. These
are precision servo mount type potentiometers fitted with calibrated positioningindicating dials. A non-inverting buffer amplifier in each unit is designed to be dcaccurate and allow the potentiometers to be used practically as potential dividers.
The schematic of IP150H and OP150K are shown in figure 3.25.
Figure 3.25: Input Potentiometer and Output Potentiometer (IP150H, OP150K)
58
3.4.2.4 Attenuator Unit – AU150B
The AU150B has two 100kΩ potentiometer connected to the terminal
sockets, and otherwise unconnected. The schematic of AU150B is shown in figure
3.26.
Figure 3.26 AU150B – Attenuator Unit
3.4.2.5 DC Motor –DCM150F
The DC motor is a permanent magnet type, which connects to the servoamplifier via a 12-way plug. It has a nominal maximum speed of
3000rotation/minute. Its torque at 2A is in the order of 0.1Nm and the rotor inertia is
3.9 * 10-5 kgM2. The schematic of DCM150F is shown in figure 3.27.
59
Figure 3.27 DCM150F – DC Motor with Tachogenarator
3.4.2.6 Reduction Gear Tacho Unit – GT150X
This unit has a 30/1-reduction gear and a tachogenarator; it also provides a
digital voltmeter and tachometer for display of motor speed directly in r/min. The
schematic of GT150X is shown in figure 3.27.
3.4.2.6 Servo Amplifier Unit – SA150D
The servo amplifier is used to operate the DC motor (DCM150F). The
schematic is shown in figure 3.28.
60
Figure 3.28:SA150D – Servo Amplifier Unit
3.4.2.7 Power Supply Unit – PS150E
This unit provides supply to all units required for MS150. The schematic is
shown in figure 3.29.
Figure 3.29: PS150E – Power Supply Unit
61
3.4.3
Hardware Connection
Hardware connection is done by interfacing between Personal Computer and
MS150 Modular Servo System through AX5412H Data acquisition card. The
connection is shown in figure 3.30. There are 8 differential input (or 16 single input)
and 2 output from the AX5412 data acquisition card. Only 2 differential input is used
for measurement and both output are used to control the direction and the speed of
the motor. Connection between MS150 Modular Servo System and AX5412H data
acquisition card will be shown in the experiment procedure respectively.
Driver used in this project is the Axiom Device Manager version 1.1 where
AX5412H is set at based address 300 with differential input and output range set at
10V bipolar.
MS150 Modular Servo System
Personal Computer
AX750 Screw Terminal Panel
Figure 3.30: Connection between PC and MS150 through AX750 Screw Terminal
Panel
62
3.4.4
Classical Control System Experiment
The classical control consists of 4 experiments which are:
1)
Closed Loop Position Control System.
2)
Effect of Gain Changes on Deadband.
3)
Unit Step Response to Gain Changes.
4)
Velocity Feedback.
The closed loop position control system experiment is to demonstrate a
simple motor-driven closed-loop position control system. From this experiment,
student can understand the operating characteristic of the pre-amplifier and can
observe the action of a simple motor driven closed loop position control system.
The effect of gain changes on deadband experiment is to demonstrate the
term ‘deadband’. From this experiment, student can observed the response of a
position control system to a step input and also investigate the effect of gain on
deadband and step response.
The effect of unit step response to gain changes experiment is to visual study
of step response to gain changes upon position control system. From this experiment,
student can understand the effect of gain changes to output response.
The effect of velocity feedback experiment is to demonstrate a simple closed
loop position control system using velocity feedback. From this experiment, student
can understand the operating characteristic of the pre-amplifier and can observe the
action of a simple motor driven closed loop position control system.
CHAPTER 4
RESULTS & ANALYSIS
This software is mainly divided into 2 parts that is Classical Control and the
artificial intelligence control using Direct Digital Control (DDC) technique. The
Classical Control consists of PID controller while the artificial intelligence control
consists of Fuzzy Logic control. Both the experiment will be discussed later in this
chapter.
For Direct Digital Control (DDC), computer was used as controller to control
the position of the servo system. A Proportional, Integral and Derivative (PID)
Controller and Fuzzy Logic control was implemented into the software to control the
position control system. Basically, this software was written for the purpose to
enhance students understanding on some of the principle on position control system
so that they are more familiar some of the practical applications based on Position
Control System and how it works.
This software starts with a splash screen shown in figure 4.1. After 10
seconds, the PID controller GUI will appear. From this screen, user can also select
which controller to use, either PID or the fuzzy logic controller. Each controller will
be discussed one by one on the following session.
64
Figure 4.1: Splash screen
4.1
Direct Digital Control
The objective of this experiment is to perform visual study on the difference
between direct digital control and classical control in controlling the position control
system. Besides, it is also introduce computer control in replacing hardware
components and function as digital controller in controlling the position control
system. For this experiment, user is required to connect the circuit shown in figure
4.2.
65
Figure 4.2: Circuit diagram for direct digital control.
4.1.1
PID controller GUI
A PID controller was implemented using software programming to control
the position control system response. This PID controller was based on the positional
algorithm with the flow diagram of positional PID control algorithm shown in figure
3.9. This PID controller is adjustable which allow user to change the gain setting of
PID respectively to observe the changes on the output response with different
combination of PID controller gain such as undamped, underdamped, overdamped
and critically damped response.
For this experiment, a total of four components were reduced compared to
classical control. Hardware components that are reduced are the input potentiometer
66
(IP150H), Preamplifier unit (PA150C), Operational Amplifier Unit (OA150A) and
the attenuator unit (AU150B). All these 4 units are replaced by personal computer.
The PID controller GUI is shown in figure 4.3.
After connecting the circuit, user can press START button to start the
controlling process and STOP button is pressed to stop the controller action. A
digital oscilloscope was implemented where user can adjust its Volt per division
setting, time per division setting and output waveform can also be pause to stop its
scrolling waveform. All the data can be recorded into the list box shown in figure
4.3.
Data
collector
Waveform
is display
here
Oscilloscope
control
display
Square
wave
generator
control
display
To see hardware
connection for
DDC
Start button
to activate
system
Figure 4.3: PID controller GUI
Slider to tune
the PID
controller
67
4.1.2
Fuzzy Logic Controller GUI
A fuzzy logic controller was implemented using software programming to
control the position control system response. This fuzzy logic controller was based
on the rule-base that has been discussed in chapter 3. The fuzzy logic control that
was design is fix (can’t manipulate the output from the GUI). Thus to improve the
system response, we have to redesign the input and out put variable by adding more
fuzzy sets, or by changing the fuzzy rule base, inference technique or even the
defuzzification technique.
For this experiment, a total of four components were also reduced compared
to classical control. Hardware components that are reduced are the input
potentiometer (IP150H), Preamplifier unit (PA150C), Operational Amplifier Unit
(OA150A) and the attenuator unit (AU150B). All these 4 units are replaced by
personal computer. The fuzzy logic controller GUI is shown in figure 4.4.
After connecting the circuit, user can press START button to start the
controlling process and STOP button is pressed to stop the controller action. A
digital oscilloscope was implemented where user can adjust its Volt per division
setting, time per division setting and output waveform can also be pause to stop its
scrolling waveform. All the data can be recorded into the list box shown in figure
4.5.
68
Error
Input
Variable
Waveform
is display
here
Rate
Output
Variable
Output
Motor
Square
wave
generator
control
display
Figure 4.4: Fuzzy Logic Controller GUI
Figure 4.5: Fuzzy Logic Controller data capture form
Start button
to activate
system
Go to
Data
collector
69
4.2
Preliminary Result Analysis
For the first part of the project, results are obtained manually from the MS150
Modular Servo System by performing the classical control experiments. Output
responses are being recorded using Yokogawa Digital Oscilloscope model DL1546.
The waveform is then analyzed manually to obtain its peak time (Tp), peak-to-peak
voltage (Vp-p), maximum voltage (Vmax), rise time (Tr) and percent overshoot (%OS).
The following terms are defined to ease analysis data:
i)
Rise time, Tr
Define as the time taken for response from 10% final value to the 90% of
final value. It shows how fast a system response to input.
ii)
Peak time, Tp
Define as the time taken for the response to rise from 0 to the first peak value
or the maximum value that the system can achieve.
iii)
Overshoot, Mp
Define as the maximum amount by which response overshoots the steady
state value or final value.
iv)
Settling time, Ts
Define as the time taken for the response to fall within and remain within
some specified percentage e.g. 2% of the steady state va1ue. It used as a
measure of the time taken for the oscillation to die away.
70
4.2.1
Closed Loop Position Control System
The closed loop position control system experiment is to demonstrate a
simple motor-driven closed-loop position control system. From this experiment,
student can understand the operating characteristic of the pre-amplifier and can
observe the action of a simple motor driven closed loop position control system.
Figure 4.3 shows the output result of the experiment.
The gain is measured on the straight part of the curve because this is the part
that will be used in practice. From figure 4.6, we can observe that the output of the
position sensor varies proportionally to the input (indicated angle).
71
i := 0 .. 30
xi :=
y i :=
−150
−140
−130
−120
−110
−100
−90
−80
−70
−60
−50
−40
−30
−20
−10
0
10
20
30
40
50
60
70
80
90
100
110
120
130
140
150
−8.42
−8.03
−7.54
−6.69
−5.94
−5.31
−4.98
−4.17
−3.75
−3.46
−2.99
−2.41
−1.8
−1.21
−0.73
−0.03
0.64
1.27
1.64
2.58
3.06
3.33
3.82
4.71
4.95
5.37
5.89
6.29
6.65
7.80
8.30
where;
xi = angle
yi = voltage
mean( x) = 0
mean( y ) = −0.037
stdev ( x) = 89.443
stdev ( y ) = 4.947
3
var ( x) = 8 × 10
var ( y ) = 24.477
m := slope ( x, y )
m = 0.055
c := intercept ( x, y )
c = −0.037
cc := corr( x, y )
cc = 0.999
72
Voltage10
Position Sensor Output vs Angle
10
Position Sensor Output
5
m ⋅xi + c
yi
0
5
− 10
10
150
− 150
100
50
0
50
xi
Indicated Angle
100
150
150
Angle
Figure 4.6: The output result of closed loop position control system experiment.
The graph is almost a straight line with a correlation coefficient = 0.999.
From figure 4.6, we can observe that the graph intercept the y-axis at –0.037 Volts.
We also can conclude that the angle sensor constant;
K θ = m(slope )
= 0.055V / 0
.
73
4.2.2
Effect of Gain Changes on Deadband
The effect of gain changes on deadband experiment is to demonstrate the
term ‘deadband’. From this experiment, student can observed the response of a
position control system to a step input and also investigate the effect of gain on
deadband and step response. Figure 4.7 shows the output result of the experiment.
i := 0 .. 10
xi :=
y i :=
1
2
3
4
5
6
7
8
9
10
37
27
20
18
16
13
10
9
8
7
where;
xi = Proportional gain
yi = Voltage
mean( x) = 5.5
mean( y ) = 16.5
stdev ( x) = 2.872
stdev ( y ) = 9.047
var ( x) = 8.25
var ( y ) = 81.85
m := slope ( x, y )
m = −2.939
c := intercept ( x, y )
c = 32.667
cc := corr( x, y )
cc = −0.933
74
Dead Zone angle
Voltage
40
Dead Zone angle VS Porpotional Gain
40
m ⋅xi + c
yi
20
0
0
0
5
10
0
xi
Porpotional Gain
15
Proportional
15
Gain
Figure 4.7: The output result of gain changes on deadband
The gain is measured on the straight part of the curve because this is the part
that will be used in practice. From figure 4.7, we can observe that the relationship
between gain and deadband is that the deadband error reduces as gain is increased.
The graph is almost a straight line with a correlation coefficient = -0.933.
From figure 4.7, we can observe that the graph intercept the y-axis at 32.6670 with a
slope of m = -2.939.
4.2.3
Unit Step Response to Gain Changes
The effect of unit step response to gain changes experiment is to visual study
of step response to gain changes upon position control system. From this experiment,
student can understand the effect of gain changes to output response. Table 4.1
shows the output result of the experiment.
75
From table 4.1 we can observe that in a position control system, that while
increasing the gain has the advantage of reducing deadband, it does also increases the
problem of overshoot. We can see that, eventually the system become unstable, thus
it continuously oscillate about the required position.
Table 4.1: Output result of the experiment unit step response to gain changes.
Kp
System
Characteristic
1
Vp-p = 5.840 V
Vmax = 5.040 V
%OS = 11.9%
Tr = 180m s
Output waveform
76
Table 4.1 (continue)
3
Vp-p = 6.560 V
Vmax = 4.8 V %
OS = 20.3%
Tr = 160m s
6
Vp-p = 6.88 V
Vmax = 4.8 V %
OS = 4.1%
Tr = 200m s
77
Table 4.1 (continue)
10
Vp-p = 6.88 V
Vmax = 4.8 V %
OS = 4.1%
Tr = 200m s
Width = 1.040 s
4.2.4
Velocity Feedback
The effect of velocity feedback experiment is to demonstrate a simple closed
loop position control system using velocity feedback. From this experiment, student
can understand the operating characteristic of the pre-amplifier and can observe the
action of a simple motor driven closed loop position control system. Table 4.2 shows
the output result of the experiment.
We can observe that in a position control system, that while increasing the
gain has the advantage of reducing deadband, it does also increases the problem of
overshoot. We can see that, eventually the system become unstable, thus it
continuously oscillate about the required position.
78
The problem of overshoot arises due to the motor’s speed carrying it past the
point of alignment, the higher the gain the greater the speed of motor response so that
above a certain gain the system can never align. Thus to overcome the overshoot, we
use velocity feedback from the tachogenarator.
Table 4.2: Output result of the experiment velocity feedback
Gain
System
Characteristic
Kv = 1
Kp = 6
Vp-p = 6.08 V
Vmax = 4.72 V
%OS = 18.6%
Tr = 160m s
Output waveform
79
Table 4.2 (continue)
Kv = 3
Kp = 6
Vp-p = 4.88 V
Vmax = 4.88 V
%OS = 0%
Tr = 1.1 s
Kv = 6
Kp = 6
Vp-p = 4.8 V
Vmax = 4.8 V
%OS = 0%
Tr = 300m s
Width = 1.090 s
80
Table 4.2 (continue)
Kv = 6
Kp = 10
Vp-p = 4.88 V
Vmax = 4.88 V
%OS = 0%
Tr = 280m s
Width = 1.100 s
Kv = 10
Kp = 6
Vp-p = 4.72 V
Vmax = 4.72 V
%OS = 0%
Tr = 440m s
Width = 1.120 s
By observing the output response of the system via the gain changes in the
Proportional controller gain via velocity feedback respectively, we can conclude that
for the P controller gain, the output of the controller is proportional to error, the
proportionality given by the controller gain KP determines how much the output from
the controller changes for a given change in error, thus KP establishes the sensitivity
81
of the controller to an error, how much the controller output changes per unit error.
While, velocity feedback is introduced into the system to overcome the overshoot,
which was used from the tachogenarator, but excessive velocity feedback gives a
slow response to the system.
Proportional controller offers the advantage of having only one adjustable or
tuning parameter, KP. However they suffer a major disadvantage that is the
controlled variable is operated with an offset. This offset can be described as a steady
state deviation of the controlled variable from set point or simply a steady state error.
While, velocity feedback is introduce into the system to overcome the overshoot,
which was used from the tachogenarator.
By increasing the proportional gain, it will create smaller offset. But, it makes
process becomes more oscillatory. Thus to remove the offset, integral action is
implemented where it function to integrate the error.
Thus to make the system to have less overshoot and a fast Tr, a PID controller
will be implemented. This is the objective of the project, which is to develop a PID
controller that will be implemented in a DDC control system.
A PID controller has 3 terms, KP, KI, and KD, which must be adjusted or
tuned to obtain satisfactory control. The derivative action gives the controller the
capabilities to anticipate where the process is heading. The derivative action reduces
overshoot and decrease oscillation around set point but PID controller is
recommended for slow process, which is usually free of noise. Fast processes
(processes with slow time constant) are easily susceptible to process noise. The
application of derivative mode will only result in amplification of the noise, because
the derivative of the fast changing noise is a large value.
82
4.3
Result Statement
After the GUI is developed, experiment will be performed via software, data
and output waveform will then be collected directly from the digital oscilloscope
from the software. This software is mainly divided into two Direct Digital Control
(DDC) experiments, that is:
1)
PID Controller
2)
Fuzzy Logic Controller
83
4.3.1
PID Controller Result Analysis
The waveform obtained through the direct digital control experiment is being
analyzed. The terms defined above are being used for each combination of PID
controller. Figure 4.8 shows the result which was capture within the GUI. All
analysis on collected data is shown in appendices A. Table 4.3 shows the analysis of
the best output result which was capture within the GUI. Here, should note that all
the system characteristic term is analyze using Yokogawa DL1546 Digital
Oscilloscope to analyze. All values are written in two decimal place. The unit used
for rise time, Tr, settling time, Ts, peak time, Tp is milliseconds (ms), %OS is in
percentage (%), ζ is dimensionless, ωn is rad/s and unit for steady state error, ess is in
Volt.
Figure 4.8: PID result which was capture within the GUI
84
Table 4.3: The analysis of PID controller output result.
Gain
System
(%)
Characteristic
P=10
Tr = 415.38
I=100
Tp = 476.92
D=50
Ts = 538.46
%OS = 0
ζ = 0.66
ωn = 11.27
ess = 0.10
Output Waveform
85
4.3.2
Fuzzy Logic Controller Result Analysis
The waveform obtained through the direct digital control experiment is being
analyzed. The terms defined above are being used for the fuzzy logic controller.
Figure 4.9 shows the result which was capture within the GUI. From the capture
result, it can be concluded that the output response of the fuzzy logic controller has a
percentage overshoot of 0.8V that is 16% overshoot.
Overshoot
(0.8V)
Figure 4.9: The fuzzy logic result which was capture within the GUI.
86
4.4
Comparison between PID and Fuzzy Logic
After designing the software for both the PID and Fuzzy logic controller, we
can conclude that:
PID controller:
-
Is easy to tune the parameter or P, I and D.
-
The PID parameter can easily be tune with the slide and is more user friendly
Fuzzy Logic:
-
The fuzzy logic controller can’t be easily be tune as the PID. If the output is
not satisfactory, thus need to add more fuzzy sets into the input and output
variable and define a new rule-base.
-
trial and error method approach will be used until satisfactory result obtained
-
The simulation output response was satisfactory. But when connected to the
MS150 servo motor, the results are not satisfactory.
-
It is time consuming
The output response of both the PID and Fuzzy logic is being compared.
Table 4.4 shows the comparison between the output response using the PID and
Fuzzy logic. From the analysis and comparison of both controllers, the PID
controller performs better compared to Fuzzy logic in terms of percentage overshoot,
with 0% overshoot compared to fuzzy logic with an overshoot of 16%. But, when
comparing both of the time rise and time settling of both controllers, the fuzzy logic
controller performs better, with a Tr of 400.00 ms and Ts of 100 ms.
87
Table 4.4: The comparison output between response using the PID and Fuzzy
logic.
PID
FLC
(best response)
4.5
Tr
415.38ms
400.00ms
Ts
538.46ms
100ms
%Os
0%
16.0%
Discussion
Proportional controller offers the advantage of having only one adjustable or
tuning parameter, KP. However they suffer a major disadvantage that is the
controlled variable is operated with an offset. This offset can be described as a steady
state deviation of the controlled variable from set point or simply a steady state error.
While, velocity feedback is introduce into the system to overcome the overshoot,
which was used from the tachogenarator.
By increasing the proportional gain, it will create smaller offset. But, it makes
process becomes more oscillatory. Thus to remove the offset, integral action is
implemented where it function to integrate the error.
88
Thus to make the system to have less overshoot and a fast Tr, a PID controller
was implemented. This is the objective of the project, which is to develop a PID
controller that will be implemented in a DDC control system.
A PID controller has 3 terms, KP, KI, and KD, which must be adjusted or
tuned to obtain satisfactory control. The derivative action gives the controller the
capabilities to anticipate where the process is heading. The derivative action reduces
overshoot and decrease oscillation around set point but PID controller is
recommended for slow process, which is usually free of noise. Fast processes
(processes with slow time constant) are easily susceptible to process noise. The
application of derivative mode will only result in amplification of the noise, because
the derivative of the fast changing noise is a large value.
For the fuzzy logic controller, the controller can’t be easily being tune as the
PID. If the output is not satisfactory, thus need to add more fuzzy sets into the input
and output variable and define a new rule-base. Trial and error method approach will
be used until satisfactory result obtained. The simulation output response was
satisfactory. But when connected to the MS150 servo motor, the results are not
satisfactory. The designing of the fuzzy logic controller is also very time consuming.
CHAPTER 5
CONCLUSION
Virtual instrumentation or Graphical User Interface (GUI) can be
defined as an intuitive, easy-to-use method of interfacing with computer program by
means of graphical screen display. It can resemble the front panel of instrumentations
or other objects associated with a computer program. The GUI design will reflects
the user’s mental model rather than the implementation model. Program features and
functions support only required user goals. No superfluous features have been added
simply because they are easy to implement, or as a result of the personal biases of the
programmers. The GUI design reflects the expected characteristics and capabilities
of the user population - visual, physical and cognitive capabilities; cultural and
technical background; domain experience; education level and so forth.
In designing digital controller in controlling the position control system, each
theory and concept play an important part in understanding the operation of position
control system. In contrast, understanding on the concept of Direct Digital Control
(DDC), Data Acquisition System (DAS), Proportional, Integral and Derivative (PID)
controller and fuzzy logic controller also play an important role in succeeding this
project. The software designing is tough to be done especially in designing the
coding for controllers. It takes a long time and great patient to make sure the
software design is in good progress.
90
There are many situations that require some type of servo-control system.
PID controllers are commonly used to regulate the time-domain behavior of many
different types of dynamic plants. These controllers are extremely popular because
they can usually provide good closed loop response characteristics, can be tuned
using relatively simple rules and are easy to construct using either analogue or digital
components. To discretise the PID controller, we need to approximate the integral
and the derivative terms to forms suitable for computation by a computer. By using
z-transform, we may discretise the analogue PID controller to obtain the pulse
transfer function for the digital PID controller. Then, equation u (k) was used to
develop the coding program for PID controller which corresponding to sequence
operation.
To design the fuzzy logic controller, the position control of the servo system
configuration was designed based on 2 inputs and 1 output. Inputs for this controller
are error and rate (de), and the motor output. Fuzzification is where we define the
quantization and membership function for input variable, error and rate (de) and
output variable, motor output in universe of discourse. It involves the conversion of
the input and output signals into a number of fuzzy represented values (fuzzy sets).
The knowledge base of a fuzzy logic controller consists of a data base and a rule
base. The basic function of the data base is to provide the necessary information for
the proper functioning of the fuzzification module, the rule base and the
defuzzification module. The basic function of the rule base is to represent the expert
knowledge in a form of if-then rule structure. The fuzzy logic was derived into a 3 x
3 rules which consist of 9 rules. For this system, max-min composition is used for the
Inferencing to obtain the fuzzy set describing the fuzzy value of the overall control
output. Defuzzification is a mapping from a space of fuzzy control actions defined
over an output universe of discourse into a space of non-fuzzy (crisp) control action.
For this system, centroid method is used for defuzzification.
91
5.1
Suggestion for future work
Since the software being developed is still remains as a trial version,
suggestions on some of the features to be implemented for future work which can be
implemented are:
1) Design a new controller which can improve the output response of the
position of servo motor.
2) Design the software using Labview, C++ or any suitable programming
language.
3) Design a virtual instrumentation using National Instrument DAQ card to
interface with the MS150 Modular Servomotor system. By using NI DAQ
card, student can design the GUI of the controllers using Labview which
is much simpler than designing controller with Visual Basic.
4) Design the fuzzy logic controller such that user can change the rule-base
immediately in the GUI, rather than have to rewrite a new coding.
92
REFERENCES
[1]
Axiom AX5412-H Data Acquisition Board User Manual
[2]
Constantine H. Houpis & Gary B. Lamout, (1985), Digital Control System –
Theory, Hardware, Software, Mc. Graw Hill Book
[3]
David M. Auslander, (1990), Real-time Software for Control, Prentice Hall
[4]
George A. Perdikaris, (1991), Computer controlled Systems, Theory and
Application, Kluwer Academic Publisher
[5]
Jee Ming Lee and Jung-Tang Huang; (1995), Motor Drive Control System
For Education and Research
[6]
Katsuhiko Ogata (1995), Discrete Time Control Systems 2nd edition,
Prentice Hall.
[7]
K.K.Tan, T.H. Lee & F.M.Leu; (1999), Development of a Distant Laboratory
using Labview
[8]
Kelvin R. Aaron, Noreen L. Foster, Danielle P.Hazel, A.M.Hasanul Basher;
(2002), Closed-loop Position Control System using Labview
[9]
Michael F. Hordeski, (1992), Control System Interfaces – Design and
Implementation Using PC, Prentice Hall
93
[10]
M.Moallen; (2000), Real-Time Position Control of Free-Electron Laser
Beams:Theory and Experiments
[11]
MS150 Modular Servo System Experiment Book 1-3
[12]
Norman S. Nise (1995) Control System Engineering, Addison Westley
Publishing
[13]
Rahman Jamal (2001), Labview Application and solutions, National
Instruments Visual Instrumentation Series
[14]
Stuart Bennert,(1995) Real-time Computer Control – An Introduction,
Prentice Hall
[15]
Timothy J. Ross, (1997), Fuzzy Logic with engineering applications,
McGraw-Hill International Edition.
[16]
Hugh F. VanLandingHam, (1985), Introduction to digital control system,
Macmillan Publishing Company.
[17]
Deitel & Deitel T. R. Nieto, (1999), Visual Basic 6: How to Program,
Prentice Hall.
[18]
Eric A., Valor Whisler and Hank Marquis, (1998), Visual Basic 6 Bible,
Hungry Minds, Inc.
94
APPENDIX A
PID RESULTS
95
Gain
System
(%)
Characteristic
P=10
Tr = 366.25
I=10
Tp = 450.33
D=10
Ts = 840.24
%OS = 9.09
ζ = 0.61
ωn = 7.85
ess = 0.47
P=10
Tr = 390.44
I=10
Tp = 0.00
D=50
Ts = 390.44
%OS = 0
ζ>1
ωn = 10.24
ess = 0.35
P=10
Tr = 540.05
I=10
Tp = 0.00
D=100 Ts = 540.05
%OS = 0
ζ>1
ωn = 7.41
ess = 0.05
P=10
Tr = 375.25
I=50
Tp = 471.33
D=10
Ts = 841.22
%OS = 12.85
ζ = 0.55
ωn = 8.70
ess = 0.29
Output Waveform
96
P=10
Tr = 405.22
I=50
Tp = 0.00
D=50
Ts = 405.22
%OS = 0
ζ>1
ωn = 9.87
ess = 0.12
P=10
Tr = 585.74
I=50
Tp = 0.00
D=100 Ts = 585.74
%OS = 0
ζ>1
ωn = 6.23
ess = 0.31
P=10
Tr = 353.85
I=100
Tp = 470.77
D=10
Ts = 892.31
%OS = 13.04
ζ = 0.54
ωn = 8.24
ess = 0.30
P=10
Tr = 415.38
I=100
Tp = 476.92
D=50
Ts = 538.46
%OS = 0
ζ = 0.66
ωn = 11.27
ess = 0.10
97
P=10
Tr = 498.46
I=100
Tp = 0
D=100 Ts = 498.46
%OS = 0
ζ>1
ωn = 8.02
ess = 0.16
P=50
Tr = 384.62
I=10
Tp = 467.69
D=10
Ts = 1046.15
%OS = 18.04
ζ = 0.48
ωn = 7.99
ess = 0.15
P=50
Tr = 384.62
I=10
Tp = 461.54
D=50
Ts = 895.38
%OS = 3.04
ζ = 0.67
ωn = 6.64
ess = 0.13
P=50
Tr = 488.64
I=10
Tp = 0
D=100 Ts = 488.64
%OS = 0
ζ>1
ωn = 8.19
ess = 0.24
98
P=50
Tr = 371.12
I=50
Tp = 479.37
D=10
Ts = 1020.58
%OS = 21.54
ζ = 0.44
ωn = 8.93
ess = 0.08
P=50
Tr = 340.19
I=50
Tp = 417.51
D=50
Ts = 618.54
%OS = 0
ζ = 0.57
ωn = 11.25
ess = 0.23
P=50
Tr = 541.22
I=50
Tp = 0
D=100 Ts = 541.22
%OS = 0
ζ>1
ωn = 7.39
ess = 0.13
P=50
Tr = 371.12
I=100
Tp = 479.37
D=10
Ts = 1051.51
%OS = 16.67
ζ = 0.50
ωn = 7.68
ess = 0.14
99
P=50
Tr = 371.12
I=100
Tp = 448.44
D=50
Ts = 757.71
%OS = 9.66
ζ = 0.60
ωn = 8.84
ess = 0.14
P=50
Tr = 366.48
I=100
Tp = 562.87
D=100 Ts = 634.22
%OS = 3.44
ζ = 0.73
ωn = 8.62
ess = 0.10
P=100
Tr = 346.38
I=10
Tp = 456.17
D=10
Ts = 1623.66
%OS = 18.21
ζ = 0.48
ωn = 5.13
ess = 0.21
P=100
Tr = 268.59
I=10
Tp = 467.20
D=50
Ts = 1237.07
%OS = 9.09
ζ = 0.61
ωn = 5.33
ess = 0.14
100
P=100
Tr = 395.86
I=10
Tp = 432.98
D=100 Ts = 541.23
%OS = 1.52
ζ = 0.80
ωn = 9.24
ess = 0.17
P=100
Tr = 386.59
I=50
Tp = 501.01
D=10
Ts = 1577.27
%OS = 16.08
ζ = 0.50
ωn = 5.04
ess = 0.15
P=100
Tr = 195.86
I=50
Tp = 463.90
D=50
Ts = 748.43
%OS = 10.02
ζ = 0.59
ωn = 9.05
ess = 0.12
P=100
Tr = 346.38
I=50
Tp = 426.79
D=100 Ts = 556.68
%OS = 5.04
ζ = 0.55
ωn = 12.98
ess = 0.10
101
P=100
Tr = 389.68
I=100
Tp = 463.9
D=10
Ts = 1212.33
%OS = 17.88
ζ = 0.48
ωn = 6.87
ess = 0.12
P=100
Tr = 386.59
I=100
Tp = 432.98
D=50
Ts = 488.64
%OS = 3.04
ζ = 0.74
ωn = 11.01
ess = 0.11
P=100
Tr = 383.52
I=100
Tp = 430.21
D=100 Ts = 486.81
%OS = 1.45
ζ = 0.75
ωn = 11.02
ess = 0.11
102
APPENDIX B
SOFTWARE SNAPSHOT
103
Splash screen
Circuit diagram for direct digital control.
104
Data
collector
Waveform
is display
here
Oscilloscope
control
display
Square
wave
generator
control
display
Start button
to activate
system
To see hardware
connection for
DDC
Slider to tune
the PID
controller
PID controller GUI
Error
Input
Variable
Waveform
is display
here
Rate
Output
Variable
Output
Motor
Square
wave
generator
control
display
Fuzzy Logic Controller GUI
Start button
to activate
system
Go to
Data
collector
105
Fuzzy Logic Controller data capture form
106
APPENDIX C
SLIDE PRESENTATION
107
Virtual Instrumentation
Of Position Control
System Using
Direct Digital Control
(DDC) Techniques
Presented by:
Mariam binti Md Ghazaly
Supervised by:
PM Dr. Mohd Fua’ad Rahmat
(Head Department
Off Campus Program)
1
Objectives
Design and developed a GUI software using
Microsoft Visual Basic 6.0 with the following goal
To visualize position control system experiment.
To apply Direct Digital Control technique in
position control system.
To make comparison of the performance
between PID controller and fuzzy logic
controller.
2
108
Scope of Work
To apply direct digital control (DDC) in position
control system.
PID controller and fuzzy logic controller will be
use to control the output response.
An interactive software will be developed to
help the student to visualize and analyze the
system using Visual Basic.
Type of motor that will be used is a DC
Servomotor.
3
Problem Statement
To visualize position control system equipment in
more interactive ways.
Developed Graphical User Interface (GUI) to represent
the Virtual Instrument system.
Comparison between conventional Controller and AI
controller.
4
109
Methodology
This project can be divided into two parts:
a)
Hardware:
Involve more in interfacing MS150 Modular Servo
System Feedback Instrument LTD and Data
Acquisition System with a personal computer.
b)
Software:
Include programming real-time software using
Microsoft Visual Basic 6.0.
5
Methodology (cont.)
Literature review and
experiment are performed
manually to understand the
concept of the controller
An interactive software will
be designed using
Microsoft Visual Basic 6.0
The software will be
equipped with a set of
graphic instructions to ease
any mistake when
performing experiment.
The software will be
integrated with hardware
to produce a GUI position
control system.
6
110
Planning Project I
Progress
1
Getting Project Title
2
Determining objectives
3
Gathering information and references
4
Literature Review
5
Learning Visual Basic
6
Preparing and Handling Proposal
7
Performing Experiment
8
Familiarize with Digital Oscilloscope
9
Identify the equipment characteristics
10
Installing AX5412-H Driver
11
Studies on AX5412-H User Manual
12
Programming Simple Software
13
Test-run and debugging
14
Preparing Presentation
15
Preparing Report
16
Presentation
17
Handling in report
July
August
September
October
7
Planning Project II
Progress
1
Learning Visual Basic
Programming
2
Literature Review
3
Software Programming
4
Updating
5
Testing and Debugging
6
Finalize and Retest
7
Preparation for Thesis
8
Correct mistakes
9
Preparation for presentation
10
Presentation
11
Handling in Thesis
November
December
January
February
March
April
8
111
Direct Digital Control (DDC)
9
Direct Digital Control (DDC)
Is where the decision of the computer are
applied directly to the process.
Is basically a microprocessor-based
technology in which a controller performs
closed-loop functions via sophisticated
algorithms and strategies.
Computer
Reference
variable
data
Control
Algorithm
Mechanization
ADC
DAC
Final
Control
Element
Plant
Process
Analogue
Feedback
Transducer
Digital Feedback
Transducer
10
112
Advantages of DDC
Bring speed, precision, flexibility to control specific
function.
Low cost.
Can distributed over a bus almost anywhere in
plant allowing condition to be displayed locally and
remotely.
Eliminate drift.
Have self diagnosing features.
Can handle advance control algorithms.
11
Position control system
12
113
Position control system
It is a control system that required motor to
rotate an output shaft to the same angle as
an input shaft.
Block diagram for position control system:
Correction
Op-amp
Input position
Potentiometer
Output position
of load
Proportional
Control
+
Servoamplifier
Sum
-
Motor
Gear &
Screw
Load
Voltage
Feedback
related to actual
position
Measurement
Potentiometer
13
Position control system (cont.)
General Motor Modeling.
Transfer function of motor:
⎛ Kt ⎞
⎜⎜
⎟
Ra J m ⎟⎠
θ m ( s)
⎝
=
Ea ( s )
⎡
1 ⎛
K K ⎞⎤
⎜⎜ Dm + t b ⎟⎟⎥
s ⎢s +
Ra ⎠⎦
⎣ Jm ⎝
14
114
Position control system (cont.)
Transfer function of motor in closed loop form:
⎛ Kt ⎞
⎜⎜
⎟
Ra J m ⎟⎠
θ m ( s)
⎝
=
Ea ( s )
⎡
1 ⎛
K K ⎞⎤
⎜⎜ Dm + t b ⎟⎟⎥
s ⎢s +
Ra ⎠⎦
⎣ Jm ⎝
where;
J m = 30 × 10 −6 kgm 2
K b = 60 ×10 −3Vsrad −1
K t = 17 × 10 −3 NmA−1
Ra = 3.2Ω
G ( s) =
(177.0833)
s[s + 10.625]
Dm = small (can' t be quote )
15
PID Controller
16
115
PID Controller
It provides a good closed loop response
characteristics.
Easy to construct either in analog or digital.
The PID control action in analog controller is
given by:
⎡
1
m(t ) = K ⎢e(t ) +
Ti
⎣
∫
t
0
e(t ) + Td
de(t )⎤
⎥
dt ⎦
where;
K = proportional gain
Ti = integral time(reset time)
TD = derivative time(rate time )
17
PID Controller (cont.)
The PID control action in digital form
maybe written as:
(
KI
⎡
+ K D 1 − z −1
M (z ) = ⎢ K P +
−1
1− z
⎣
)⎤⎥ E (z )
⎦
where;
K P = proportional gain
K I = integral gain
K D = derivative gain
18
116
PID Controller (cont.)
Equation u(t) will be used to developed the coding
program for PID controller which corresponding to
sequence operation.
u ( k ) = K p e( k ) +
Ts
TI
t
∑ e( k ) +
i =0
Td (e(k ) − e(k − 1))
Ts
where Ts = time sampling = 1 ms
t
∑ e(k ) = total error
i =0
e(k ) − e(k − 1) = deviation error
19
PID Controller (cont.)
Value of Kp, Ki & Kd was calculated using root
locus technique.
Tuning method : Ziegler Nichols
The PID controller have been simulated using
matlab to ensure the manual calculation is correct.
Kp = 1.266
Ki = 0.1261
Kp
= 10.0444 ≈ 10
Ki
Kd = 0.0563
thus Ti =
thus Td =
Kd
= 0.044 ≈ 1.0
Kp
20
117
PID Controller (cont.)
Start
Read Input from DAQ
Compute Error
Compute Proportional
action
Delay until
next sample
Compute Integral
action
Compute Derivative
action
Send output to DAQ
End
21
Fuzzy Logic Controller
22
118
Fuzzy Logic Controller
The steps that we are needed to do are:
Step 1: Define quantization for state (input)
variables, error and de and output variable, motor in
universe of discourse.
Step 2: define a membership function for the input
and control output.
Step 3: define the rules base and summarize them
in an FAM table. The values in the FAM table are
the output control.
Step 4: define the initial condition
Step 5: conduct the simulation. For this part, use
max-min configuration. Defuzzify using centroid
method.
Step 6: Compare the results with the conventional
control; PID controller.
23
Flowchart of Simulator
Start
Initialization
&
Declaration
Plant/ Math Model
Fuzzy Logic Controller
FLC Subroutine
Plotting
No
k=k+1
Yes
K =end
end
24
119
FLC Design Methodology
Start
Study the System
Fuzzification
-determine the objective
-identify process and controller’s
input & output
-quantize the input and
output variables
-define the membership function
Inferencing
Parameter Tuning
-Mapping of membership
function
-Fuzzy inference rules
-Scaling factor
-Derive the fuzzy control
rules-based
-Define the inference engine
-define the membership function
Defuzzification
-Chooses defuzzification
method function
No
Performance OK?
Simulation & Testing
Fuzzy Controller Operation
-Fuzzification
-Fuzzy Inference
-Defuzzification
Yes
25
End
Fuzzification
input variables
output variable
Fuzzy logic
rule
26
120
Fuzzification (cont.)
define a membership function for the input and control
output.
Fuzzy Input Variable :
(1) Error
27
Fuzzification (cont.)
(2) Rate
Output: Motor
28
121
Inferencing
Derive the fuzzy control rules-based : 3 x 3 rules
error
Ne
Ze
Pe
Nde
No
No
Zo
Zde
No
Zo
Po
de
antecedents
consequents
Pde
Zo
Po
Po
29
Inferencing (cont.)
1)
2)
3)
4)
5)
6)
7)
8)
9)
Fuzzy logic rules:
If error is Ne AND de is Nde THEN output is No
If error is Ne AND de is Zde THEN output is No
If error is Ne AND de is Pde THEN output is Zo
If error is Ze AND de is Nde THEN output is No
If error is Ze AND de is Zde THEN output is Zo
If error is Ze AND de is Pde THEN output is Po
If error is Pe AND de is Nde THEN output is Zo
If error is Pe AND de is Zde THEN output is Po
If error is Pe AND de is Pde THEN output is Po
30
122
Defuzzification
For this system, maxmin is used for
compositional and
centroid method is used
for defuzzification.
The picture shows the
firing strength for the
rule base using max-min
inference. We can see
the output of the
controller on the bottom
right of the rule base
output of the
window.
controller
31
Hardware & Software Review
32
123
Hardware & Software Review
The components used MS150 Modular Servo
system.
AX5412 Data Acquisition Card with AX750
Screw Terminal Panel
Software Used:
Microsoft Visual Basic™ 6.0
33
Hardware Connection
MS150 Modular Servo System
Personal Computer
AX750 Screw Terminal Panel
34
124
Software Development
Includes 2 direct digital control experiment i.e.: PID
controller and fuzzy logic controller.
For the DDC design:
z A digital PID controller was implemented based
on the positional algorithm.
z Totally 4 equipments from MS150 Modular
Servo System has been reduced.
z Digital Oscilloscope and Function Generator
also included.
35
Snapshot taken from the software
36
125
PID results & Analysis
37
Matlab Simulation
Input
Output
Overshoot is quite
high, thus need
tuning.
38
126
Software User’s Interface
Data
collector
Waveform
is display
here
Oscilloscope
control
display
Square
wave
generator
control
display
To see hardware
connection for
DDC
Start button
to activate
system
Slider to tune
the PID
controller
39
PID Results
Gain
(%)
System
Characteristic
(time in ms)
Comparison on output waveform
Overshoot is quite
Yokogawa DL1546
high, KD need
tuning.
P=10
I=10
D=10
Software
Tr = 366.25
Tp = 450.33
Ts = 840.24
%OS = 9.09
ζ = 0.61
ωn = 7.85
ess = 0.47
40
127
PID Results (cont.)
Gain
(%)
System
Characteristic
(time in ms)
Comparison on output waveform
Yokogawa DL1546
P=10
I=100
D=50
Software
Tr = 415.38
Tp = 476.92
Ts = 538.46
%OS = 0
ζ = 0.66
ωn = 11.27
ess = 0.10
41
Effect of PID
Closed Loop
Response
Rise
Time, Tr
Percent
Overshoot,
%OS
Settling
Time, Ts
Steady
State Error,
ess
Proportional
Gain, KP
Decrease
Increase
Small
Change
Decrease
Integral Gain,
KI
Decrease
Increase
Increase
Eliminate
Derivative
Gain, KD
Small
Change
Decrease
Decrease
Small
Change
42
128
Fuzzy Logic results
& Analysis
43
Software Development Results
Error
Input
Variable
Waveform
is display
here
Rate
Output
Variable
Output
Motor
Square
wave
generator
control
display
Start button
to activate
system
Go to
Data
collector
44
129
Simulation result
No
Overshoot
45
Simulation result (cont.)
46
130
FLC Real-time results
Overshoot
(0.8V)
47
FLC results (cont.)
48
131
Conclusion
49
Conclusion
1)
PID
easy to tune the parameter or P, I and D.
simulation result by matlab is quite similar to the real
time result.
2)
Fuzzy Logic
trial and error method approach will be used until
satisfactory result obtained
must find the best membership function and rule to
come out with the best output response.
the simulation output is very good. But when we
connect it to hardware, the results is not as good.
Time consuming
50
132
Conclusion (cont.)
PID
FLC
Tr
415.38ms
400.00ms
Ts
538.46ms
100ms
%Os
0%
16.0%
51
Conclusion (cont.)
3)
Software design
software designing is tough to be done especially in
designing the coding for controllers.
take a long time and great patient to make sure the
software design is in good progress.
52
133
Proposed Work
1)
2)
Design a new controller which can improved the
output response of the position of servo motor.
Design the software using Labview , C++ or any
suitable programming language.
53
References
1)
2)
3)
Axiom AX5412-H Data Acquisition Board User Manual
MS150 Modular Servo System Experiment Book 1-3
Norman S. Nise (1995) Control System Engineering, Addison
Westley Publishing
4)
Stuart Bennert,(1995) Real-time Computer Control – An
Introduction, Prentice Hall
5)
George A. Perdikaris, Computer controlled Systems, Theory
and Application, 1991, Kluwer Academic Publisher
6)
Katsuhiko Ogata(1995), Discrete Time Control Systems 2nd
edition, Prentice Hall.
7)
Development of a Distant Laboratory using Labview-K.K.Tan,
T.H. Lee and F.M.Leu; 1999.
8)
Motor Drive Control System For Education and Research-Jee
Ming Lee and Jung-Tang Huang; 1995.
9)
Real-Time Position Control of Free-Electron Laser
Beams:Theory and Experiments-M.Moallen; 2000.
10) Closed-loop Position Control System using Labview-Kelvin R.
Aaron, Noreen L. Foster, Danielle P.Hazel, A.M.Hasanul
Basher; 2002.
54

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project