Diesel HCCI Results at Caterpillar - Department of Energy · Diesel HCCI Results at Caterpillar Kevin Duffy, ... Caterpillar Engine Research ... Performance/Emissions Data

Diesel HCCI Results at Caterpillar

Kevin Duffy, Jonathan KilkennyAndrew Kieser, Eric Fluga

DOE Contracts DE-FC05-00OR22806, DE-FC05-97OR22605

Contract Monitors – Roland Gravel, John Fairbanks

DEER ConferenceNewport, RI

August 27, 2003

Caterpillar Engine ResearchDiesel & Emissions Technology

Outline

n Review challengesn Development processn Recent progress

•Multi-cylinder engine•Operating range expansion• Advanced controls• Fuels effects

n Summaryn Future plans


n Proper air / fuel mixing

n Limited load range

n Combustion phasing and control

HCCI - The Challenge

HCCI Engine(Homogeneous ChargeCompression Ignition)

Low temperature combustionultra low emissions


SystemSimulation

CAT3DCombustion

Model

SprayVisualization

Design

Optical Engine

Single CylinderTesting

FEA

HCCI Development at Caterpillar

Multi Cylinder Testing

This project was undertaken pursuant to an agreement with theUnited States in connection with settlement of disputed claims in anenforcement action under the Clean Air Act


Ultra-Low Emissions

NOx (g/hphr)

Part

icul

ate

(g/h

phr)

0 0.5 1 1.5 2 2.5

0.025

0.05

0.075

0.1

2010 Regulations

2004 Regulations

Recent HCCI Results





HCCI - The ChallengeHCCI Engine

(Homogeneous ChargeCompression Ignition)



Mixture Preparation Optimization

HCCI Tip Progression

0

0.5

1

1.5

2

2.5

0 0.2 0.4 0.6 0.8 1

NOx (g/hp*hr)

Sm

oke

(A

VL

)

Tip 1

Tip 2

Tip 3

Additional 90% Smoke Reduction


Multi-Cylinder Engine Results

0

0.5

1

1.5

2

Main Timing

AV

L S

mo

ke

Tip A

Tip B

0

1000

2000

3000

4000

5000

Main Timing

HC

(pp

m)

Tip A

Tip B

0

2000

4000

6000

8000

10000

Main Timing

CO

(pp

m)

Tip A

Tip B

Main Timing

BS

FC

Tip A

Tip B

1200 RPM, ~1/4 Load @ <0.04 g/hphr NOx









Light Load HCCI Operation

n Dec and Sjoberg (2003) examined low load HCCI operationwith fully premixed iso-octane and a GDI system

n Found as load was reduced (phi<0.2), CO increased rapidlydue to bulk gas quenching, combustion efficiency decreasedto <50%

n Suggested there is a fundamental low load limit for HCCIdriven by these bulk gas quenching phenomena


HCCI Idle Operation

5.67.2IMEPcov

11401580HC (ppm)

386NOx (ppm)

36105358CO (ppm)

BL+5deg retardBLTiming


Expanding HCCI Load Range

0200400600800

100012001400160018002000

600 900 1200 1500 1800 2100Speed (rpm)

BM

EP

(kP

a)475 HP C15 Full Load Curve









Controlling Cylinder-to-Cylinder Variability

n Charge air temperature variability, outside cylinders get less“fresh” charge

n Intake valve actuation available on current ACERTTM engines

Back-flow During ValveTiming Strategies Fresh Charge

Inferred Temperature Distribution


Cylinder to Cylinder Balancing

n Intake valve actuator trim strategies used tophase combustion properlyn Allows for higher power levels to be achieved

Untrimmed Trimmed


Combustion Control – System Interaction

InjectionTiming

BoostPressure

VGTPosition

BackPressure

ManifoldTemp

CompressionRatio

A/F Ratio

BSFC

EmissionsCylinderPressure

Rise Rate

DesiredSpeed/Power

ExhaustTemperature

Heat Rejection


Controls Development Process

0

0.5

1

1.5

2

2.5

3

-2 0 2 4 6 8 10

BSNOx (g/hphr)

AV

L S

mo

ke

1200 rpm

1500 rpm

1800 rpm

1500 rpm, 1/4 load

0

50

100

150

200

250

-20 -15 -10 -5 0 5 10 15 20Crank Angle (Deg)

Hea

t R

elea

se R

ate

Performance/Emissions Data

In-Cylinder Pressure Data

Neural Network “learns” the interactions

Engine Settings

Rapid Prototype Controller

Optimum SettingsSensors

Trai

ning

Dat

a fo

r N

eura

l Net

wor

k

Actuators

Injection Timing

Boost Pressure

VGT Position

Back Pressure

Manifold Temp

Compression Ratio

A/F Ratio

BSFC

EmissionsCylinder Pressure

Rise Rate

Desired Speed/ Power

Exhaust Temperature

Heat Rejection Injection Timing

Boost Pressure

VGT Position

Back Pressure

Manifold Temp

Compression Ratio

A/F Ratio

BSFC

EmissionsCylinder Pressure

Rise Rate

Desired Speed/ Power

Exhaust Temperature

Heat Rejection


Sample Neural Net ModelsPmax from EC parameters

CO from ICP+EC parameters PM from ICP+EC parameters

NOx from ICP


Two-Stage Combustion with Diesel Fuel

Rate of cool flame chemistry varies with load. Inductiontime affected by residual content, IVA, boost, etc.

Amount of cool flame activity affects main combustiontiming and rise rate.

Question: Can fuel properties be manipulated in such amanner to advantageously effect cool flame chemistry

and combustion phasing?


Study Fuels Effects on HCCI Combustion

n Teamed with major oil industry partnern Developed matrix of 12 fuels

• Diesel and gasoline like fuels• Vary ignition quality and volatility• In stock or easily blended fuels

n Test in single cylinder (~1 wk/fuel)n Preliminary (CAT only in-house) results presented

here for high and low cetane # fuels


Effect of Cetane Number

Higher CN leads to:-earlier cool flame and main ignition-40% increase in peak cylinder pres-2x increase in cyl pres rise rate

1200 rpm, ~1/4 load, equal fuel mass, timing=NOx takeoff

0

2

4

6

8

10

12

14

16

-50 -40 -30 -20 -10 0 10 20 30 40 50

CAD

PC

YL

(MP

a)

0

50

100

150

200

250

300

Hea

t Rel

ease

(kJ/

m3)

Higher CNLower CN

1200 rpm, ~1/4 load, equal fuel mass, advanced timing

0

2

4

6

8

10

12

14

16

-50 -40 -30 -20 -10 0 10 20 30 40 50

CAD

PC

YL

(M

Pa)

0

50

100

150

200

250

300

Hea

t Rel

ease

(kJ/

m3)

Higher CNLower CN


1200 rpm, ~1/4 load

-0.1-0.05

00.05

0.1

0.15

0.20.25

0.30.35

0.40.45

40 50 60 70 80 90Injection Timing (deg btdc)

BS

NO

x (g

/hph

r)

Higher CNLower CN

Emissions and BSFC Impact

Timing Retard

AVL smoke < 0.05 all points

1200 rpm, ~1/4 load

270

280

290

300

310

320

330

340

350

360

40 50 60 70 80 90Injection Timing (deg btdc)

BS

FC

(g/k

Wh

r)

Higher CNLower CN

Timing Retard

0

0.2


Highly Competitive BSFC

235

250

265

280

295

310

325

340

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6NOx (g/hp-hr)

2003 ACERT2007 Demo2010 HCCI

BS

FC

(g/k

W-h

r)

220

230

240

250

260

270

280

290

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0NOx (g/hp-hr)

2003 ACERT2007 Demo2010 HCCI

BS

FC

(g/k

W-h

r)


Conclusionsn HCCI demonstrated up to 1600 bar BMEP

• Single and multi-cylinder engines running• Highly competitive thermal efficiencies

n Stable HCCI combustion obtained at idle, light load conditionsn Meeting controls issue on all fronts

• Full control system characterization under development usingneural network models with in-cylinder pressure sensing

• Cylinder variability observed, VVT appears to be viabletrimming mechanism

n Fuels effects (CN) can have significant impact onperformance/emissions, higher CN not right direction


Next Steps

n Advanced control algorithm integration on multi-cylindern Expand load limits furthern HCCI fuels effects testingn Continue advanced fuel and air system

development

Diesel HCCI Results at Caterpillar - Department of Energy · Diesel HCCI Results at Caterpillar Kevin Duffy, ... Caterpillar Engine Research ... Performance/Emissions Data

Documents