us department of energy & bosch gasoline systems ... · pdf fileus department of energy...

Advanced Flex-Fuel Systems

1 GS/ENS-NA System and Advanced Engineering | March 2nd 2011 | © 2011 Robert Bosch LLC and affiliates. All rights reserved.

Gasoline Systems

US Department of Energy & Bosch Gasoline Systems Collaborative Research on Advanced Flex Fuel Systems

Hakan YilmazDirector

System and Advanced EngineeringGasoline Systems, North America

Robert Bosch LLC


2Gasoline Systems

AGENDA:

• US Market and Powertrain Technology Overview

Fuel Economy CAFÉ 2016 Engine Technologies Turbo Charging & Gasoline Direct Injection

• Advance Flex Fuel System Project and Proposed Concepts

Project Overview Partnership and Targets Fuel Efficiency Engine HW and System Optimization Ethanol Detection In-Cylinder Pressure Sensor Utilization Model Based Controls Reduced Calibration Effort Emissions DI Start and Emission System Concept

• Questions

GS/ENS-NA System and Advanced Engineering | March 2nd 2011 | © 2011 Robert Bosch LLC and affiliates. All rights reserved.


3Gasoline Systems

EU CAFE CARB CHINA JAPAN

140

CAFE LT/MDV

CAFE PC

CO2[g/km]

130

95

1,09 t weight

2008CAFE = Corporate Average Fuel Economy PC = Passenger Cars, LT / LDT = Light Trucks (pick-ups, vans, SUVs), MD(P)V = Medium Duty (Passenger) Vehicles GHG = Greenhouse Gases NHTSA = National Highway Transportation and Safety Administration CARB = California Air Ressources Board mpg = miles per gallon China weight based limits (here for 1,09 tons curb weight) CAFE data NHTSA report October 2006 EU data for ACEA (Association des Constructeurs Européens d‘Automobiles) 6th EU Report 24.8.2006, for MY05 and 06 from T&E 2006 / 2007

2012

phase-in 2012-15 for 65/75/80/100% of fleet

229

183

mpg

22,324,9

27,931,2

37,241,6

55,762,3

l/100 km

10,59,4

8,47,5

6,35,7

4,23,8

GasolineDiesel

157 = 35,5 mpg



4Gasoline Systems

CO

2 re

duct

ion

EvolutionReference base: CC vehicle, 1400 kg, 100 kW, MT, G0=2.0L PFI, NEDC driving cycle

CC

COMMENTS

%-value is reduction potential for stand-alone measures.

Combination of mea-sures w/ negative synergies cause of part-redundancies

Turbocharger (T/C) w/o additional mea-sures leads to losses by lower compression.

Extreme downsizing (eDZ) includes effect on cylinder reduction 43

PFI=port fuel injection | DZ=downsizing | T/C = turbocharger | Thm=Thermal management | St/St=Start/Stop System

12%

dualVVT

Decos

ThMReCu

4% 1%

2%St/St

3%

4%

PFI improvement

22%DI

T/C

DZ30%

-2%

11%2%

DI turbo & downsizing

Dir

ect I

njec

tion

turb

o-ch

arge

d &

dow

nsiz

edG1

29%4%VVL

eDZ45%

5%

Extremedownsizing

DI t

urbo

extr

eme

dow

nsiz

ed

G2G0

Adv

ance

d P

FI

Synergistic CO2 Reduction Measures for Gasoline Engines



5Gasoline Systems

AGENDA:

• US Market and Powertrain Technology Overview

Fuel Economy CAFÉ 2016 Engine Technologies Turbo Charging & Gasoline Direct Injection

• Advance Flex Fuel System Project and Proposed Concepts

Project Overview Partnership and Targets Fuel Efficiency Engine HW and System Optimization Ethanol Detection In-Cylinder Pressure Sensor Utilization Model Based Controls Reduced Calibration Effort Emissions DI Start and Emission System Concept

• Questions



6Gasoline Systems

Ethanol

26.8Heat of Combustion(Calorific Value)

Air-Fuel Ratio

Evaporation Enthalpy

Boiling Temperature

Octane

Properties Unit

MJ/kg-fuel

-

kJ/kg-fuel

°C

RON

Gasoline

42.5

14.8

25…215

>91

380…500

9.0

845

78

111

MJ/kg-air 2.982.90

Dielectric Constant - 2.0 24.0

Benefits / Drawbacks

- Requires Higher Injection Quantity up to 35%- Cold start critical below -10°C (E85)- Produces more water, Retarded Cat warm up- More wall wetting- More oil contamination+ More Power / Higher Efficiency due to stronger air charge cooling+ High Knock Limit

Fuel Properties – Gasoline vs. Ethanol



7Gasoline Systems

Advanced FFV (Project size 3.6M USD) Project Targets

Timeline Partners

• US Department of Energy

• Robert Bosch LLC

• Ricardo, Inc

• University of Michigan, Ann Arbor

• Targets10+% fuel efficiency improvement with E85Adv. controls to reduce FFV development effortEthanol learning via in-cylinder pressure sensor

0 3 6 9 12 15 18 21 24 27 30 33 3610/07 01/08 04/08 07/08 10/08 01/09 04/09 07/09 10/09 01/10 04/10 07/10 10/10

Phase 1

Phase 3

T-2

T-6

Phase 2

Phase 4

T-3T-4

T-1

T-7

T-10

T-12

T-5

T-8T-9

T-11

T-12

K-Off M-1 M-2 M-3 M-4

1970 1980 1990 2000 2020 2040 2060

• ACEA-Selbstverpflichtung

2008: 140g/km (-25%)

Power / Comfort / Safety

Emissions (HC,CO,NOx, PM)

CO2-Emissions

Resources

From Fossil to renewable Fuels

Technological Challenges



8Gasoline Systems

Engine Optimization for Gasoline & Ethanol DI System Concept for Emissions

Ethanol Detection via PS-C Model Based Controls via PS-C

Throttle Intake

manifoldCylinder

Exhaustmanifold to turbine and

wastegate

cylW

CompressorIntercoolerBoost

manifold

Compressor and intermediate blocks

N

measurements for cylinder air flow

estimation

Cylinder flow estimatorwith MAF sensor bias estimation

eaf cylW

sAFR

Conversion to stoichiometric

AFR

1

AFR FF controller

ffW fbW 1PIPI

sks

1

+

-

AFR FB controller

injW

e

eff

Residue calculation

,Si Spcyl cylp p

r

Air flow

MAF

Nor

mal

izat

ion

Filte

ring

Cyc

le A

vera

ging

cylP



9Gasoline Systems

Advanced Flex Fuel Engine Configuration

Goal: Optimized engine design for improved combustion performance Hardware structural robustness for all fuel blends Minimized fuel economy penalty due to increased ethanol content Enhanced performance through exploitation of fuel properties

Barriers: Engine hardware peak cylinder pressure capability restricts E85 performance volumetric compression optimized for gasoline only additional NVH measures necessary for E85 combustion

Engine management system Fuel system pressure not sufficient for high ethanol content Injection dynamic flow rate optimized for gasoline Boost / VVT systems require optimization for ethanol



10Gasoline Systems

Larger Fuel Pump Revised Fuel Spray Late Intake Valve Closing High Compression Ratio Revised Piston Bowl 130 Bar Pmax 0

2

4

6

8

10

12

0 50 100 150 200 250 300 350 400CA Deg

Valv

e Li

ft (m

m)

Std Plus 20 Deg

Std Plus 40 Deg




11Gasoline Systems

Approach: Engine Design Optimization Increased compression ratio: 9.25:1 10.7:1

Ecotec LNF Design

Flex-Fuel Optimized Design

Flex-Fuel Optimized Design




12Gasoline Systems

Approach: Engine Design Optimization Increased maximum cylinder pressure: 100bar 130bar

Cylinder Head

Head Gasket

Connecting Rod

Rod Bearings Crank Damper Washer

Piston Pin




13Gasoline Systems

Approach: Engine Design Optimization Late Intake Valve Closing (LIVC) strategy

intake cam profile with extended open duration of 37degCA intake cam phaser with increased authority of 60degCA

Intake Cam Phaser

Intake Camshaft

Intake Cam Profile0

2

4

6

8

10

12

0 90 180 270 360 450 540 630 720

Crank Angle [degCA]

Valv

e Li

ft [m

m]

Exhaust Cam

Intake Cam

New Intake Cam




14Gasoline Systems

Approach: Fuel System Adaptation Increased operating pressure in the direct injection system: 150bar 200bar Increased capability in the fuel pump: 3-lobe, 0.9cc/rev 4-lobe, 1.1cc/rev Decreased fuel rail volume: 135cc 66cc Optimized spray pattern

Fuel Rail

Fuel Pump and its Follower, Adaptor




15Gasoline Systems

Conclusions: Fuel Efficiency Improvement Increased compression ratio + LIVC Advanced multi-variable calibration Optimized transmission shift pattern Optimized final drive ratio


406080

100120140160180200220240260280300320340360380400420440460

1000 1500 2000 2500 3000 3500 4000 4500 5000 5500 6000 6500

Engine Speed (rpm)

Torq

ue, M

(Nm

)

FFV - E85

FFV - E0

Production LNF - E0

2008 HHR SS (LNF)Published Data

E0

LNF

E85/E0

FFV

E85

FFV

E0Final Drive Ratio -- 4.05 2.90 2.90 2.900-30 mph s 3.07 2.96 3.380-60 mph s 6.2 6.75 6.13 7.160-100 mph s 15.1 16.00 13.87 16.821/4 mile s 14.8 15.18 14.63 15.611/4 mile mph 99 97.65 102.74 96.50Max. Power HP 260 256 300 234Max. Torque ft-lbs 260 260 307 264

Fuel

FTP Fuel Economy [mpg]

Delta

Highway Cycle Fuel Economy [mpg]

Delta

2.9 FDR* 4.05 FDR* 2.9 FDR* 4.05 FDR*

FFV E85 21.59 19.37 10.3% 32.36 28.47 12.0%

FFV E0 29.83 27.02 9.4% 44.62 39.62 11.2%

Fuel

FTP Fuel Economy [mpg]

Delta

Highway Cycle Fuel Economy [mpg]

Delta

2.9 FDR* 4.05 FDR* 2.9 FDR* 4.05 FDR*

FFV E85 21.59 19.37 10.3% 32.36 28.47 12.0%

FFV E0 29.83 27.02 9.4% 44.62 39.62 11.2%



16Gasoline Systems

Emission Concept

Goal: Achieve ULEV emission levels with all fuel blends E0..E85 Define potential path to reach SULEV Optimized combustion design for different fuels No additional after-treatment system complexity

Barriers: Cold starts with E85 significantly higher ethanol injection quantity resulting in increased HC delayed catalyst heating due to higher water concentration

Increased wall wetting and oil contamination with increased ethanol content increased HC emissions and smoke



17Gasoline Systems

Cold Start Strategy Platform

180 Bar

-20°C 3 sec 4.2 (same as E0

single injection)

Results Conclusion

• 2.0L Turbo DI (LNF)• ECU MED17 Platform• E85 Class III Fuel (E70)• 3 lobe HP pump

suction compressionTDC BDC TDC

HSPHomogeneous SPlit Start

Engine speedFuel rail pressure

Single injection

Triple Injection

100 Bar

-20°C1.8 sec5.3

SH1Single High Pressure Start

Start Pressure

Start TemperatureStart Time Start Fuel Factor

MY09 Chevy HHR Turbo



18Gasoline Systems

Results: Multiple Injection Strategy at Cold Start

1

2

3

4

5

6

7

8

9

10

11

12

13

14

Start Time[seconds]

Sta

rt E

nric

hmen

t Fac

tor

321

Triple Injection100 Bar

Single Injection100 Bar

Great

Pot

entia

ls fo

r Im

prov

emen

t

• Fuel: Gasoline• Start temperature: -20°C• Vehicle: MY2009 Chevy HHR• Engine: Ecotec LNF (2.0L Turbo DI)

Cold Start Strategy



19Gasoline Systems

Results: Multiple Injection Strategy at Cold Start

1

2

3

4

5

6

7

8

9

10

11

12

13

14

Start Time[seconds]

Sta

rt E

nric

hmen

t Fac

tor

321

• Fuel: E85 Class III• Start temperature: -20°C• Vehicle: MY2009 Chevy HHR• Engine: Ecotec LNF (2.0L Turbo DI)

Triple Injection

180 BarTriple Injection

100 Bar

Single Injection No Start !

180 Bar

Cold Start Strategy



20Gasoline Systems

Approach: Piston Bowl Design for HC Reduction• Increased bowl width for improved air/fuel mixture• Optimized fuel spray deflection angles• Smoother surface transitions & decreased crevice volumes

Approach: Injector Design and Injection Strategies• Spray targeting for fuel mixture preparation• High Pressure Stratified Start • Homogenous Split Injection for Catalyst Heating

Cum

ulat

ive

HC

em

issi

ons

2400

Principal sketch, FTP-cycle

0 2 10 20

Post Light-Off

Time [s] BOSCH

sidemounted

Start High-PressureStratified Start

suction compr.

TDC BDC TDC BDC

combust.

injec.

Catalyst-Heating Homogenous-SPlit

suction compr.

TDC BDC TDC

1st inj

BDC

combust.

2nd inj

2400

Principal sketch, FTP-cycle

0 2 10 20

Post Light-Off

Time [s] BOSCH

Time [s] BOSCH

sidemounted

Start High-PressureStratified Start

suction compr.

TDC BDC TDC BDC

combust.

injec.

Catalyst-Heating Homogenous-SPlit

suction compr.

TDC BDC TDC

1st inj

BDC

combust.

2nd inj

Emission Concept



21Gasoline Systems

Conclusion: ULEV level Emissions reached with NMHC SULEV level Emissions reached with CO and NOx

Emission Concept

SULEV 20C FTP - E0 0.034 0.489 0.028 345SULEV 20C FTP - E85 0.029 0.246 0.029 349SULEV 20C FTP - E77 0.021 0.281 0.010 333

NMHC CO NOX

[g/mile] [g/mile] [g/mile]California ULEV Standard 0.055 2.10 0.070California SULEV Standard 0.010 1 0.02

Cat Test - Fuel NMHC CO NOX CO2[g/mile] [g/mile] [g/mile] [g/mile]



22Gasoline Systems

Ethanol Detection Strategy

Goal: Develop a systematic approach to achieve an accurate, robust, and fast

ethanol content estimation for engine performance optimization, even in the presence of component aging and sensor drifts

Barriers: Direct measurement from Ethanol sensor on fuel line additional cost and system complexity

Indirect estimation by Exhaust Gas Oxygen (EGO) sensor sensitive to mass airflow sensor drift and fuel system failures requires coordination of fuel system and ethanol content adaptation

Indirect estimation by Cylinder Pressure Sensor (PS-C): robustness and accuracy has not been addressed for combustion and heat

release based detection



23Gasoline Systems


Fuel System Fault Adaptation

des =1

Wcyl

AFRs

PuddleCompensation

Wff1 Wff FeedbackController

Cfb(s)

Wfb

WcylManifoldBreathingDynamics

Air ChargeEstimation

Tm MAF Winj

Conversion toEthanol Content

e

Engine EGO sensor = AFRs

AFR

Stoichiometric AFR Adaptation(Ethanol Content Adaptation)

Switching between the two adaptations (tank refill, etc.)

Approach: Ethanol Detection via A/F Ratio using EGO sensor• Ethanol content and fuel system fault adaptation depends on the same signal• Error accumulates in estimated ethanol content due to the switching approach



24Gasoline Systems


Approach: Latent Heat of Vaporization (LHV) Based Features• Injection Mode: switching between Single and Split injections• Extracted Features

Residue: Detection Feature:

0

0.5

1

Fuel

Inje

ctio

n P

ulse

-120 -100 -80 -60 -40 -20 0 20 40 60 800.8

0.9

1

1.1

1.2

1.3

1.4

1.5

Cyl

inde

r Pre

ssur

e [b

ar]

Crank Angle [deg aBDC]

E85 - SingleE85 - Split

CompressionIntake

Intake ValveOpen

IVC

Injection Mode

0 10 20 30 40 50 60 70 800.01

0.015

0.02

0.025

0.03

0.035

EtOH (%)

Det

ectio

n Fe

atur

e

MAF = 80 kg/hr,MAF = 100 kg/hr,MAF = 130 kg/hr,MAF = 150 kg/hr,MAF = 170 kg/hr,

• Principles: Charge cooling effects, captured by rLHV, increases with ethanol fuel content


2500RPM


25Gasoline Systems


Approach: Ethanol Detection using AFR and LHV Based Features MAF Sensor Error: Estimated MAF using Manifold Absolute Pressure (MAP) Sensor Fuel Injector Drift: Estimated EtOH using Cylinder Pressure Sensor

des =1

Wcyl

AFRs

PuddleCompensation

Wff1 Wff FeedbackController

Cfb(s)

Wfb

WcylManifoldBreathingDynamics

Air ChargeEstimation

Tm MAFWinj

EngineEGO Sensor

= AFRs

AFR

Stoichiometric AFR Adaptation(Ethanol Content Adaptation)

PS-C Sensor

MAP

Latent Heat of Vaporization Adaptation(Ethanol Content Adaptation)



26Gasoline Systems


Results: Ethanol Detection using LHV and AFR Based Features Simulated Case: MAF sensor bias (5%) + fuel ethanol content change (0 50EtOH%) +

fuel injector drift (0 20%) under varying engine operation conditions

Current operation point is within the identified regime for generating LHV-based detection features

The generated LHV-based feature is reliable for ethanol content adaptation

Reset triggered

Fuel injector fault reported



27Gasoline Systems



4050

6070

800

20

40

60

80

-5

0

5

10

EtOH (%)Relative load (%)

EtO

H E

st. E

rror

(%

)

4050

6070

80 020

4060

80

-10

-5

0

5

10

15


EtO

H E

st. E

rror

(%

)

4050

6070

80 020

4060

80

-505

101520


EtO

H E

st. E

rror

(%

)

4050

6070

80 020

4060

80

0

10

20


EtO

H E

st. E

rror

(%

)

1500RPM 1750RPM

2000RPM 2500RPM

Steady-State Ethanol Detection

10% fuel injector drift occurred

Experimental Results: Steady-State Estimation: frequently

visited operation conditions ranging from 1500-2500RPM

Simulated Case: 10% fuel injector drift introduced via rapid prototyping at engine dynamometer detected by the LHV based feature


28Gasoline Systems

Control Strategy

Goal: Optimized engine controls for all fuel blends Robust and fast adaptation of engine control parameters Minimum additional calibration effort

Barriers: Current engine controllers adjust the spark and fuel injection timing, variable

valve timing and boost in a feed-forward manner

significant additional calibration efforts for different fuel blends deviation from optimal calibration due to interpolation inaccuracies lack of coordinated control of air path during speed/load transients



29Gasoline Systems

Control Strategy

Approach: Closed-loop combustion

control via spark angle using cylinder pressure feedback

Optimize set-point values of key engine control variables (e.g. spark angle, VVT) to adapt ethanol combustion

Variable Camshafts

Electronic Throttle

DI Fueling System(pf)

Intake Manifold(pim)

Exhaust Manifold(λ)

Compressor

Ignition System

Wastegate

Combustion Controller

Air Charge Controller

Operating Set-Point Demands

(uwg, θth, φVVT)*

(N, T*)

Induction Volume(pb)

Σ(Wcyl, pb)*

Σ

Comb. Feature Calculation

(CA50)

(CA50)*

(Ik)

Ethanol Detection

Turbo

Air/Fuel Controller

Etoh%

Σ

(λ)*

(λ)

Σ

Σ(φs)*

Direct Injection Spark Ignition

Engine(Ik, pcyl)

(pim, Wa, N, φVVT, ...)

(pcyl)

(pcyl)

(pb)

(COVIMEP, Ik, ...)

Air Charge Estimator

Engine Torque Struture

Pedal Command

(Wcyl)

(Wcyl )

(θth)

(uwg)

(Wa)

HFM Sensor

(uwg, θth, φVVT)

(COVIMEP,dpmax, ...)

(φVVT)

Fuel

Spark

Air

Signal

(pf, φinj)*

Component Controller



30Gasoline Systems

Control Strategy

Result: Engine Air/Fuel Path Dynamics + Combustion Modeling – Stochastic



31Gasoline Systems

Control Strategy

Approach: Closed-Loop Combustion Control Via Spark Advanced using Cylinder Pressure

Feedback MBT spark timing optimization for Ethanol Fuels using Extremum Seeking



32Gasoline Systems

SeriesSeries33--wayway

catalystcatalyst

Customized Engine Management (ECU) Customized Engine Management (ECU)

Model based controls Fuel property detection

TurboTurbo--chargingcharging

Standard turbo charger

Solenoid MHI Side/Central mount Split injection Dynamic flow rate

CustomizedCustomizedInjection Injection

InIn--Cylinder Pressure SensingCylinder Pressure Sensing Direct combustion feedback Ethanol detection

High energy coils Transient torque

Throttling and IgnitionThrottling and Ignition

2xVVT via Cam phasing Extended phase authority

Variable Valve ActuationVariable Valve Actuation

Stainless Steel High flow range

High Pressure PumpHigh Pressure Pump

Advanced Flex Fuel System Configuration



33Gasoline Systems

Thank You!


us department of energy & bosch gasoline systems ... · pdf fileus department of energy...

Documents