Integral Data act. SURGE cont. + opt. IV – PARETO set

Integral Data act. SURGE cont. + opt. IV – PARETO set
Transient Performance of
Large-bore SI Engine
Oldřich VÍTEK, Jan MACEK
Josef Božek Research Center
Czech Technical University, Prague
Martin Vacek, Jiří Klíma
PBS Turbo, Velká Bíteš
Company
Logo
Outline
Introduction
Engine Model
Computed Cases
Results
Conclusion
2
Goals
To optimize 2-stage turbocharger group for
required BMEP target (24 bar) while satisfying
level of NOx (TA Luft).
Both steady operation and transient response
(from BMEP 9 to 24 bar) considered.
Different engine flexibility assumed – from
standard design to fully flexible ICE (VGT,
intake/exhaust VVA, var. compression ratio).
3
Basic Engine Parameters
Bore-to-Stroke Ratio
[-]
Compression Ratio
[-]
Engine speed
Mean piston speed
[m/s]
Charging
Fuel
Air Excess
No. of Intake Valves
No. of Exhaust Valves
4
0.86
14
Constant
11
2-stage Turbocharged
Methane
Variable (TA Luft)
2
2
Engine Layout
Air excess control:
• based on free oxygen
(including EGR)
BMEP control:
• intake throttle
• compressor blow-by (‘green’
path)
• waste-gate (2 variants)
• variable geom. turbine (VGT)
Compressor blow-by control:
• customer requirement (~5%)
• compressor surge
External cooled EGR control:
• EGR valve eff. area
• dedicated EGR compressor
driven by electric motor
• exhaust throttling (if needed)
5
LP-WG control:
• dedicated LP turbine by-pass to
limit LP stage boost pressure (if
needed)
Application of Available Software
GT-POWER code used (1-D approach to
pipes, 0-D approach to other engine parts).
Compressor/turbine maps provided by PBS
Turbo/MAN.
mFRONTIER applied – optimization using
genetic algorithm (MOGA-II).
6
Engine Model
Virtual engine model – based on existing
design while using experience with very
similar engine.
Limited experimental data available =>
missing parameters estimated by the authors.
Predictive sub-models applied – heat transfer,
comb. chamber wall temp., FMEP, NOx, etc.
Air excess adjusted to keep constant
NOx level (TA Luft).
Complex control.
7
Turbocharger Model
Classical approach (lumped model) –
application of compressor/turbine maps.
The maps provided by turbocharger
manufacturer.
Dedicated HP turbocharger –
compressor/turbine design optimized for 2stage turbocharging applications.
8
Turbocharger Model
Experimentally observed increase of total
boost group efficiency (complete 2-stage
system tested) => applied in calculations.
9
Computed Cases
Sensitivity studies:
Different control means
Different EGR levels => optimized
Different NOx levels (TA Luft requirements)
Other studies:
Transient advanced control to speed up response
Surge control
Optimization of reference 1-stage variant
10
Optimization Procedure
Complex multi-variable multi-constraint multitarget optimization:
Variables: 4 multipliers to select proper HP/LP
compressor/turbine size + 1 value of EGR amount
+ 2 values of intake valve (IV timing and IV
profile width) + 2 values of exhaust valve (EV
timing and EV profile width) + 1 value of
compression ratio.
11
Optimization Procedure
Complex multi-variable multi-constraint multitarget optimization:
Constraints: control targets (under both steady
and unsteady operation) – BMEP, compressor
blow-by, EGR, NOx level (TA Luft), compressor
surge; no max. in-cylinder limit was imposed.
Optimization targets: min. BSFC and min.
transient duration (simultaneously)
12
Results: Integral Data
pseudo-transient simulation:
•
•
•
All variables are solved in transient mode
however engine speed is kept constant
(‘continental’ electricity network).
Significantly easier to control the engine
(when compared with fully transient case).
Qualitative trends are preserved – if any
variant is faster in pseudo-transient mode, it is
also faster in fully transient mode.
13
Results: Integral Data
pseudo-transient PARETO set:
Pseudo-trans. value
Low BSFC,
slow response
Compromise?
high BSFC,
fast response
14
Results: Integral Data
PARETO set – pseudo-trans. vs trans.:
+5.5s
+3.8s
+1.6s
15
Results: Integral Data
2-stage variant optimizations – PARETO set:
HP + EGR comp.
blow-by 5%
16
Results: Integral Data
2-stage variant optimizations – PARETO set:
HP + EGR comp.
blow-by 5%
17
Results: Integral Data
2-stage variant optimizations – PARETO set:
HP + EGR comp.
blow-by 5%
18
Results: Integral Data
2-stage variant optimizations – PARETO set:
HP + EGR comp.
blow-by 5%
19
Results: Integral Data
2-stage variant optimizations – PARETO set:
HP + EGR comp.
blow-by 5%
20
Results: Integral Data
2-stage variant optimizations – PARETO set:
HP + EGR comp.
blow-by 5%
21
Results: Integral Data
2-stage variant optimizations – PARETO set:
HP + EGR comp.
blow-by 5%
22
Results: Integral Data
2-stage variant optimizations – PARETO set:
•
HP + EGR comp.
blow-by 5%
23
The most important factor
(in terms of both BSFC and
response time) is HP turbine
size.
Results: Integral Data
2-stage variant optimizations – PARETO set:
HP + EGR comp.
blow-by 5%
24
Results: Integral Data
2-stage variant optimizations – PARETO set:
HP + EGR comp.
blow-by 5%
25
Results: Integral Data
2-stage variant optimizations – PARETO set:
•
HP + EGR comp.
blow-by 5%
EGR 7%
TA Luft 100%
There are differences among variants (with
respect to BMEP control) – throttle control is
the fastest one while VGT has the best
efficiency => combine both approaches
26
Results: Integral Data
2-stage variant optimizations – PARETO set:
LP EGR
blow-by 5%
EGR 7%
TA Luft 100%
27
Results: Integral Data
act. SURGE cont. + opt. IV/EV – PARETO set:
HP + EGR comp.
blow-by 5%
TA Luft 25%
28
cont.: throttle
opt. amount of EGR
Results: Integral Data
act. SURGE cont. + opt. IV/EV – PARETO set:
HP + EGR comp.
blow-by 5%
TA Luft 25%
29
cont.: throttle
opt. amount of EGR
Results: Integral Data
act. SURGE cont. + opt. IV/EV – PARETO set:
HP + EGR comp.
blow-by 5%
TA Luft 25%
30
cont.: throttle
opt. amount of EGR
Results: Integral Data
act. SURGE cont. + opt. IV/EV – PARETO set:
•
HP + EGR comp.
blow-by 5%
TA Luft 25%
31
Intake VVA can significantly
improve engine
performance (both BSFC
and transient response).
cont.: throttle
opt. amount of EGR
Results: Integral Data
act. SURGE cont. + opt. IV – PARETO set:
low BSFC
BMEP=24bar
HP + EGR comp.
blow-by 5%
TA Luft 25%
fast response
BMEP=24bar
34
cont.: Throttle
opt. amount of EGR
Results: Integral Data
act. SURGE cont. + opt. IV/EV – PARETO set:
low BSFC
BMEP=24bar
fast response
BMEP=24bar
narrow cam (optimized for
Miller cycle) + phasing?
•
IVC is a critical parameter.
HP + EGR comp.
blow-by 5%
TA Luft 25%
35
cont.: Throttle
opt. amount of EGR
Conclusions:
•
•
Optimization of engine setting was performed
under both steady state and transient state.
Contradictory tendency:
•
•
•
Low BSFC setting leads to slow response time.
Fast response time leads to high BSFC.
The most important factor (in terms of both
BSFC and response time) is HP turbine size.
36
Conclusions:
•
•
•
There are significant differences among
variants (with respect to BMEP control) –
throttle control is the fastest one while VGT
has the best efficiency.
Optimal EGR value is usually very low – it is
typically smaller for LP EGR variant when
compared with HP EGR + EGR comp.
The lean concept (without any EGR) is quite
efficient in terms of low BSFC and low NOx.
37
Conclusions:
•
•
•
Intake VVA can significantly improve engine
performance (both BSFC and transient
response).
IVC is a critical parameter.
If exhaust valve parameters are properly
optimized, exhaust VVA is not needed.
38
Thank you for your attention
39
Was this manual useful for you? yes no
Thank you for your participation!

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

Download PDF

advertising