© Ricardo plc 2005RD. 05/52402.1
Gasoline Engine Performanceand Emissions
Future Technologies andOptimization
Paul Whitaker - Technical Specialist -Ricardo
8th June 2005
© R
icard
o p
lc 2
005
RD. 05/52402.1 2
Contents
Fuel Economy Trends and Drivers – USA and Europe
Production Advanced Gasoline Technologies
Boosting for Downsizing Technologies
– Lean Boost Direct Injection
– EGR Boost Direct Injection
– 2-stroke / 4-stroke Switching
Optimization of Advanced Gasoline Combustion Systems
– Example: Optimization of a CAI enabled flexible valvetrain engine
Summary and Conclusions
© R
icard
o p
lc 2
005
RD. 05/52402.1 3
Fuel Economy Trends and Drivers – USA and Europe
Significant improvements since 1995:
– Increased diesel fleet (25% to 45% in last 5 years)
– Some improved gasoline engines
2002 & 2003 data shows levelling off
Diesels near saturation
OEMs currently face significant commercial challenges
Further improvements in gasoline powertrain efficiency will be required
that are cost effective and attractive to consumersSource: EPA Light-Duty Automotive Technology and Fuel Economy Trends: 1975 Through 2004
Overall fleet fuel economy deteriorating
Significant Truck market share increase (now 48%)
Significant increase in vehicle weight
Emissions legislation limiting dieselization
Increased pressure for improved fuel economy
– CO2 legislation? stricter CAFE? High Fuel Prices?
© R
icard
o p
lc 2
005
RD. 05/52402.1 4
Production Advanced Gasoline TechnologiesStratified Charge SIDI
Failed to deliver fuel economy promise
Work continues on central injection (2nd generation)
NOx challenge difficult to overcome in US
Homogeneous =1 SIDI
Fuel economy, emissions & performance benefits
Synergies with many other technologies
Less complicated/expensive than stratified, lowerrisk
Cylinder Deactivation
Fuel economy potential similar to Stratified SIDI
Relatively easy and inexpensive to implement
Particularly suitable for OHV engines
Easier to apply to V8 than V6 engines
Advanced NVH technologies required for full benefit
Fully Mechanical VVA
Unthrottled operation, throttled across intake valve
Potential friction reduction from reduced valve lift
Complex mechanical system
Only in production with BMW
Continuously Variable Cam Phasing
Most widely adopted advanced gasoline technology?
Fuel economy, emissions & performance benefits
Best benefits are from dual independent phasing
Phaser speed/controls can limit benefits
Baseline: V8 PFI Gasoline Engine WithoutEGR, Cost of $2,000, 500,000 units p.a.
Cost Benefit vs HSDI
Better
Worse
Stratifi
ed SIDI
Fu
ll V
VA
PFI
Ho
mo
ge
ne
ou
s S
IDI
DualVCP
Cylin
de
r D
ea
cti
va
tio
n
Emissions and performancebenefits must also be considered
© R
icard
o p
lc 2
005
RD. 05/52402.1 5
Contents
Fuel Economy Trends and Drivers – USA and Europe
Production Advanced Gasoline Technologies
Boosting for Downsizing Technologies
– Lean Boost Direct Injection
– EGR Boost Direct Injection
– 2-stroke / 4-stroke Switching
Optimization of Advanced Gasoline Combustion Systems
– Example: Optimization of a CAI enabled flexible valvetrain engine
Summary and Conclusions
© R
icard
o p
lc 2
005
RD. 05/52402.1 6
Lean Boost DI (LBDI)
Ignition retard
stability limit
Lean stabilitylimit
Increasing MAP
Naturally aspirated
0
20
40
60
80
100
120
140
160
180
200
1 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9Excess Air Factor
IME
P [
%]
Ignition retardstability limit
Lean stabilitylimit
Increasing MAP
Naturally aspirated
Control Octane requirement by
– direct injection
– lean operation at full load
– enables high compression ratio
Stratified operation at part load
Fuel economy approaching diesel
Lower cost and weight
Real world fuel economy benefit
© R
icard
o p
lc 2
005
RD. 05/52402.1 7
Lean Boost DI demonstrator has delivered downsizing benefits - Fullylean operation at all conditions delivers the fuel economy of a dieselwith gasoline NVH and lower cost
VCR
HCCI
SIDI LBDI
Camless EMV
Boosted DI
MPI Base1
1.5
2
2.5
3
3.5
1 1.05 1.1 1.15 1.2 1.25 1.3 1.35 1.4
Benefit Index
Ris
k In
de
x
VCR
HCCI
SIDI LBDI
Camless EMV
Boosted DI
MPI Base
1.1l LBDI maximises hardware value:
– Higher power than 1.6 l (+5%)
– Higher torque than 1.6 l (+20%)
– Higher economy NEDC >20% better
• 132 g/km EUDC CO2
– Matched or better acceleration
– Economy over wide operating range
LBDI delivers fuel economythrough downsizing, boosting &lean operation with highcompression ratio
Octane requirement controlled by:
– direct injection
– lean operation at full load
– stratified lean DI operation atpart load
© R
icard
o p
lc 2
005
RD. 05/52402.1 8
EGR Boost Direct Injection
EGR Cooler
SIDI V6 Engine
EGR Valve
EGR Boost
Fixed GeometryTurbocharger
Fixed GeometryTurbocharger
Intercooler
Intercooler
Throttle3-way Catalyst
EGR Cooler
SIDI V6 Engine
EGR Valve
EGR Boost
Fixed GeometryTurbocharger
Fixed GeometryTurbocharger
Intercooler
Intercooler
Throttle3-way Catalyst
Uses cooled EGR as a dilutant to reduceoctane requirement
Enables high CR boosted operation
Lambda=1 operation with Three-wayCatalyst
Cost penalty 30%
Fuel economy benefit 12%
Technical risk medium/high
© R
icard
o p
lc 2
005
RD. 05/52402.1 9
DI Gasoline 2 / 4 Stroke Switching Concept improves efficiencythrough extreme downsizing – 2 stroke operation provides enhancedlow speed torque
Very high low speedtorque - 220 Nm/litre
Ultimate fun to drive(torque and revs)
HCCI at part load ?
LNT required for 2 strokemode only
(not required for typicalurban or motorwaydriving)
© R
icard
o p
lc 2
005
RD. 05/52402.1 10
Cost/FE Benefit Trade-off Advanced Gasoline Technologies
Baseline: V8 PFI Gasoline Engine Without EGR, Cost of $2,000, 500,000 units p.a.
Cost Benefit
vs HSDI
Better
Worse
Str
ati
fie
d S
IDI
Fu
ll V
VA
PFI
Homogeneous SIDIDualVCP
Cylin
de
r D
ea
cti
va
tio
n
Emissions and performancebenefits must also be considered
HSDI diesel
2s / 4s Switching SIDI
LBDI
EGR Boost
Downsizingtechnologies offerthe greatest fuel
economy potential
© R
icard
o p
lc 2
005
RD. 05/52402.1 11
Contents
Fuel Economy Trends and Drivers – USA and Europe
Production Advanced Gasoline Technologies
Boosting for Downsizing Technologies
– Lean Boost Direct Injection
– EGR Boost Direct Injection
– 2-stroke / 4-stroke Switching
Optimization of Advanced Gasoline Combustion Systems
– Example: Optimization of a CAI enabled flexible valvetrain engine
Summary and Conclusions
© R
icard
o p
lc 2
005
RD. 05/52402.1 12
Increased variability requires a mathematical optimisation technique– DoE provides an efficient, robust analytical solution
New powertrain configurations haveincreased challenges in optimisation:
– More variables
– More interactions
– More non-linear responses
– More emphasis on robustness
Design of Experiments can create a“virtual” engine or powertrain andreduce experimental work leading to:
– Shorter development times
– Better, more robust solutions
– Increased engineering understanding
DoE is now essential for many engine development and calibration tasks
Two requisites for successful DoE
– Good tools, especially for modeling and optimisation
– Intelligent implementation of DoE process
• Planning Design Testing Modeling Validation Optimisation
2
4
6
8 34
56
78
9101
0
2
3
4
5
6
EVL [mm]BMEP [bar]
BS
HC
[g
/kW
h]
2
4
6
8 34
56
78
9101
0
2
3
4
5
6
0
2
3
4
5
6
EVL [mm]BMEP [bar]
BS
HC
[g
/kW
h]
© R
icard
o p
lc 2
005
RD. 05/52402.1 13
Example – Optimisation of a Controlled Auto Ignition enabledgasoline engine using a flexible valve train providing variablelift & phasing on intake & exhaust
BMW Valvetronic systemapplied to intake andexhaust system on 4cylinder engine
© R
icard
o p
lc 2
005
RD. 05/52402.1 14
A Design-of-Experiments (DoE) approach using Stochastic ProcessModelling (SPM) and optimiser tools identified variable settings forbest fuel economy
Data sets obtained for inlet valve throttled, load ranges1000-3500 rev/min with inlet throttle wide open.
– Control variables IVP, EVP, IVL, EVL, MAP
Source:
Variables
Inlet valve phasing (IVP)
Exhaust valve phasing (EVP)
Inlet valve lift (IVL)
Exhaust valve lift (EVL)
Manifold air pressure (MAP)
Responses Modelled
BMEP
Fuel consumption
HC
NOx
CO
COV (Net IMEP) (stability)
50% Mass Fraction Burned
Ignition to 10% MFB
10 – 90 % MFB
PMEP
Optimum spark timing (MBT)
Model qualityassessment
Model quality
assessment
Estimation of ranges ofeach variableeach variable
Estimation of ranges of
each variable
Design experimentDesign experimentDesign experimentDesign experiment
Test corner pointsof experiment on engine
Test corner points
of experiment on engine
Run test sequenceon engine
OK
Run test sequenceon engine
OK
ReRe--definedefine
corner pointscorner points
unsuitable
Re--define
corner points
Re define
corner points
unsuitable
Optimise
-with constraints
PASS
Optimise
-with constraints
PASS
Test engine to validate
optimum points
Test engine to validate
optimum points
EndEnd
OK
EndEnd
OK
Poor validation
Generate modelsGenerate models
Definition of variables under investigationDefinition of variables under investigationDefinition of variables under investigationDefinition of variables under investigationObjectivesObjectivesObjectivesObjectives
FAIL
Modelimprovement
© R
icard
o p
lc 2
005
RD. 05/52402.1 15
Comparison of model output and measured data shows goodcorrelation – Data shown at minimum fuel consumption
200
250
300
350
400
450
500
550
600
1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0
BMEP [bar]
BS
FC
[g
/kW
h]
DoE model - variable inlet and exhaust liftand phase - unconstrained optimum BSFC
Validation test results
200
250
300
350
400
450
500
550
600
1.0 2.0 3.0 4.0 5.0 6.0
BMEP [bar]
BS
FC
[g
/kW
h]
DoE model - variable inlet and exhaust liftand phase - unconstrained optimum BSFC
Validation test results
200
300
400
500
600
700
800
900
0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0
BMEP [bar]
BS
FC
[g
/kW
h]
DoE model - variable inlet and exhaust lift andphase - unconstrained optimum BSFC
Validation test results
1000 rev/min valve lift controlled
2000 rev/min valve lift controlled2000 rev/min throttled
3500 rev/min valve lift controlled
© R
icard
o p
lc 2
005
RD. 05/52402.1 16
Validated optimum settings show that CAI combustion providessignificant BSFC gains compared with normal SI operationResults shown for 2000 rev/min load rangeVCP=Variable Cam Phasing VVL=Variable Valve Lift
250
275
300
325
350
375
400
425
450
475
500
1.5 2.0 2.5 3.0 3.5 4.0 4.5
BMEP [bar]
BS
FC
[g
/kW
h]
SI - Dual VCP
SI - Dual VCP + Dual VVL
CAI - Dual VCP + Dual VVL
Baseline – fixed maximumlift and variable inlet andexhaust cam phasing
Variable inlet and exhaustphasing and lift – SI operation
Variable inlet and exhaustphasing and lift – CAI operation
© R
icard
o p
lc 2
005
RD. 05/52402.1 17
Typical pressure/volume characteristics for CAI operation showrecompression in exhaust stroke
Negative valve overlap createsrecompression cycle
Recompression largely reversible
Higher trapped mass due to high residuals –higher pressures
Lower losses across intake valve – reducedthrottling due to high residuals
© R
icard
o p
lc 2
005
RD. 05/52402.1 18
0
2
4
6
8
10
12
14
16
18
20
1.5 2.0 2.5 3.0 3.5 4.0 4.5BMEP [bar]
Fu
el C
on
su
mp
tio
n B
en
efi
t [%
]
Fuel Consumption benefitversus Dual VCP
CAI combustion provides significant BSFC gains compared withnormal SI operation mainly due to pumping work reductionResults shown for 2000 rev/min load rangeVCP=Variable Cam Phasing VVL=Variable Valve Lift
Dual VCP + Dual VVLwith SI combustion
Dual VCP + Dual VVLwith CAI combustion
-1.0
-0.9
-0.8
-0.7
-0.6
-0.5
-0.4
-0.3
-0.2
-0.1
0.0
1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0
BMEP [bar]P
ME
P [
bar]
SI - Dual VCP
SI - Dual VCP + Dual VVL
CAI - Dual VCP + Dual VVL