numerical simulation to improve engine control … · numerical simulation to improve engine...
TRANSCRIPT
NUMERICAL SIMULATION TO IMPROVE
ENGINE CONTROL
DURING TIP-IN MANOEUVRES
F. MILLO, C.V. FERRARO, F. MALLAMODIPARTIMENTO DI ENERGETICA
POLITECNICO DI TORINO
GT-Suite Users International ConferenceFrankfurt a.M., October 21st 2002
L. PILOFA-GM POWERTRAIN
Presentation overview
• Introduction
• Experimental set-up
• The engine model
• The vehicle and driveline model
• Model validation
• Evaluation of control strategies
• Conclusions
• Future work
Introduction
Vehicle driveability has nowadays undoubtedly become a key success factor for passenger cars, thus setting a further targetfor the designers besides pollutant emissions, fuel consumption and vehicle performance.
However, while the determination of fuel consumption and exhaust emissions follows well established standards, objective and reproducible criteria for the evaluation of a vehicle’s driveability are more difficult to be defined: a rating system based on the subjective assessments of experienced test drivers recorded during a sequence of relevant manoeuvres is usually employed.
One of the most common among these manouvres is the so called “tip-in”, that is a sudden opening of the throttle operated from conditions of low speed and low load.
Tip-in manoeuvres
Introduction
0
20
40
60
80
100
120
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5time [s]
loa
d [
%]
Sudden opening of the throttleVehicle jerkings and acceleration fluctuations
The abrupt change of the torque delivered by the engine excites the torsional natural frequencies of the driveline, causing vehicle jerking and acceleration fluctuations, which are the main responsible for the driver’s and passengers’perception and assessment of perfomance and comfort.
Tip-in manoeuvres:The role of numerical simulation
Introduction
The vehicle’s behavior during tip-in is obviously determined by the powertrain design, but the engine management system may play a decisive role, especially with modern “torque based” systems, which are able to decouple the engine response from the driver’s demand by means, for example, of a “drive by wire” (DBW) throttle, that electronically filters the throttle valve opening command during fast acceleration transients, smoothing the vehicle response to sharpdriver’s requests.
However, since designing and tuning the engine control system still remains a time consuming activity, several efforts have been made to explore ways that could lead to significant reductions of thedevelopment process: in particular, the use of numerical simulation to build engine models by which control strategies can be tested and tuned “on a desk” seems to be very promising.
Tip-in manoeuvres:The role of numerical simulation
Introduction
The aim of this work is therefore to evaluate the potential of numerical simulation in the analysis of the dynamic transient response of a vehicle during tip-in manoeuvres, so as to reduce the experimental tests required to optimize the control strategies of the engine management system.
Presentation overview
ü Introduction
• Experimental set-up
• The engine model
• The vehicle and driveline model
• Model validation
• Evaluation of control strategies
• Conclusions
• Future work
Experimental set-up
VEHICLE: FIAT PUNTO
MAIN ENGINE FEATURES
DOHC, 4 valves/cylinderDistribution
Multi-point electronic injectionFuel Metering System
107 Nm @ 4000 rpmMaximum Torque
52 kW @ 5000 rpmMaximum Power
Pent-roofCombustion Chamber
10.6 : 1Compression Ratio
1242 cm3Displacement
70.8 / 78.86 mmBore/Stroke
S.I. 4 cylinders in lineType
Experimental set-up
1 2 3 4
Alimentatore
Amplificatore Convertitore f-V
Analizzatore a/f
Amplificatore Convertitore f-V1. Giri ruota
2. Giri motore
3. Dosatura a/f
4. Apertura farfalla
5. Accelerazione longitudinale vettura
5
Segnali acquisiti
Elaborazione dati
Acquisizione dati digitale
Data analysisDigital acquisition
Amplifier
Amplifier
f-V Converter
f-V Converter
A/F AnalyzerPower supply
Acquired signals
1. Wheels angular speed
2. Engine angular speed
3. Air-Fuel ratio
4. Throttle opening
5. Vehicle longitudinalacceleration
DATA ACQUISITION SYSTEM
Presentation overview
ü Introduction
üExperimental set-up
• The engine model
• The vehicle and driveline model
• Model validation
• Evaluation of control strategies
• Conclusions
• Future work
Main features of the engine model
The engine model
air inlet
air filter + Helmotz resonator
Close coupled catalyst
exhaust manifold
intake plenum + primary intake runners
connectingpipe + throttle
Validation of the engine model: full load conditionsGlobal quantities
The engine model
0
40
80
120
160
200
240
0 1000 2000 3000 4000 5000 6000n [rpm]
Bra
ke t
orq
ue
[Nm
]
0
10
20
30
40
50
60
Bra
ke p
ow
er [
kW]
SIM
EXP
The engine model
Validation of the engine model: full load conditionsInstantaneous quantities
-360 -270 -180 -90 0 90 180 270 3600
10
20
30
40
50
60
Crank Ang le [de g ]
Pre
ssu
re
[b
ar]
experimentals imulation
W.O.T n = 4000 r.p.m.
Validation of the engine model: part load conditions
The engine model
0
1
2
3
4
5
6
7
8
9
10
0 200 400 600 800 1000inlet manifold pressure [mbar]
bm
ep [
bar
]
EXPSIM
1500 rpm
ü Introduction
üExperimental set-up
üThe engine model
• The vehicle and driveline model
• Model validation
• Evaluation of control strategies
• Conclusions
• Future work
Presentation overview
• Front-wheel drive
The vehicle and driveline model
Schematic diagram of the driveline
Axle shaft
• 5 speed gear box
• Asymmetric axle shafts
9 DOF Torsional Model(Matlab-Simulink)
The vehicle and driveline model
1 2
3 4
7
5 6
8
9
C m o t C r e s
Equivalent inertia:1. Engine2. Clutch + Gear-box + Differential3. 5. Hub + rim4. 6. Tyres9. 9’. Vehicle
Engine torque
Load torque
9’
Load torque
9
865
743
21
Simulation Scheme
GT-Power Engine model
MATLAB -SIMULINK Driveline and Vehicle Model
Engine speed
Engine torque
Throttle
opening
A / F
ü Introduction
üExperimental set-up
üThe engine model
üThe vehicle and driveline model
• Model validation
• Evaluation of control strategies
• Conclusions
• Future work
Presentation overview
Example: 2nd gear - Initial engine speed: 1500 rpm
Model validation
Validation of the complete engine-vehicle model
0
20
40
60
80
100
120
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5time [s]
load
[%
]
Time history of the throttle opening
Engine angular speed: simulation vs experimental results
Model validation
Example: 2nd gear transmission ratio
Vehicle longitudinal acceleration: simulation vs experimental results
Model validation
Example: 2nd gear transmission ratio
ü Introduction
üExperimental set-up
üThe engine model
üThe vehicle and driveline model
üModel validation
• Evaluation of control strategies
• Conclusions
• Future work
Presentation overview
Use of the model for the evaluation of control strategies
After the assessment of the accuracy and of the reliability of the complete engine-vehicle model, thenumerical simulation has been used as a prediction tool to analyze the impact of different control strategies on vehicle driveability during tip-in. The two following strategies were evaluated:
· drive by wire control
· spark advance control
Evaluation of control strategies
“Drive by wire” (DBW) throttle control
Evaluation of control strategies
0
20
40
60
80
100
120
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5time [s]
load
[%
]
direct controlDBW
Time history of the throttle opening
Evaluation of control strategies
“Drive by wire” (DBW) throttle control
Engine angular speed: comparison between DBW and direct control in 2nd gear
Evaluation of control strategies
“Drive by wire” (DBW) throttle control
Vehicle acceleration: comparison between DBW and direct control in 2nd gear
Evaluation of control strategies
“Drive by wire” (DBW) throttle control
Jerk and vehicle acceleration during tip-in:comparison between DBW and direct control in 2nd gear
Spark advance controller
Evaluation of control strategies
60
70
80
90
100
110
120
500 1000 1500 2000 2500 3000 3500 4000 4500 5000 5500 6000n [rpm]
En
gin
e to
rqu
e [N
m]
basis advancebasis advance -6°basis advance -9°basis advance +3°basis advance +6°
Evaluation of control strategies
Spark advance controller
Engine angular speed: comparison between simulation with and without spark advance controller
Evaluation of control strategies
Spark advance controllerVehicle acceleration:
comparison between simulation with and without spark advance controller
ü Introduction
üExperimental set-up
üThe engine model
üThe vehicle and driveline model
üModel validation
üEvaluation of control strategies
• Conclusions
• Future work
Presentation overview
A one-dimensional fluid-dynamic engine model was employed in conjunction with a Matlab-Simulink vehicle and driveline model to analyze the dynamic transient response of a gasoline passenger car during tip-in manoeuvres.
The numerical simulation was shown to be reliable and helpful for the study of proper control strategies of the engine management system aiming to enhance the vehicle driveability.
A detailed validation process of the complete engine-vehicle model, based on several sets of experimental data, wasperformed.
CONCLUSIONS
ü Introduction
üExperimental set-up
üThe engine model
üThe vehicle and driveline model
üModel validation
üEvaluation of control strategies
üConclusions
• Future work
Presentation overview
Further investigations will be devoted to the improvement of the simulation of the vehicle acceleration, by means of more detailed models for the driveline and the vehicle.
FUTURE WORK
On the engine side the analysis will be extended to theevaluation of more detailed ECU models, in order to better represent the behavior of torque based systems.
Moreover, the use of more detailed combustion models to obtain essential information such as knock likelihood caused by sparktiming increases will also be evaluated.
FUTURE WORK
Example of a more detailed ECU model of a torque based system.
ACKNOWLEDGMENTS
The authors wish to thank:
• Gamma Technlogies for the support in the simulation activities;
• FA-GM Powertrain S.p.A. for the support and the permission to publish this work;
• Dr. Ferreri, Dr. Franchino and Dr. Tuttobene (Politecnico di Torino) for their helpful contribution during the experimental and simulation data processing.
A/F during tip-in