Combustion Institute/Canadian Section (CI/CS) Spring Technical Meeting, 2007 May 13-16, 2007 Banff, Alberta, Canada
Combustion Timing Control of Heptane-fueled HCCI by Reformer Gas: Experiments and Chemical Kinetic Modeling Vahid Hosseini, M David Checkel Mechanical Engineering Department, University of Alberta, Edmonton, Alberta, Canada
project supported by Auto21 National Center of Excellence
Outline
Introduction Low octane HCCI combustion, Research target and methodology, Reformer gas Experimental setup Definitions Experimental Results Operating region, pressure trace characteristics, heat release analysis Modeling Heat release, temperature, intermediate species Conclusions
Outline
Introduction Low octane HCCI combustion, Research target and methodology, Reformer gas Experimental setup Definitions Experimental Results Operating region, pressure trace characteristics, heat release analysis Modeling Heat release, temperature, intermediate species Conclusions
Introduction HCCI combustion
(1)
Homogenous Charge Compression Ignition combustion is the autoignition of a highly diluted air/fuel mixture that leads to high efficiency and low NOx emissions
Introduction HCCI combustion engine: drawbacks
Combustion timing control Narrow operating range High emissions of HC and CO
(2)
Introduction HCCI combustion engine: drawbacks
Combustion timing control Narrow operating range High emissions of HC and CO
(2)
Introduction
(3)
Influential parameters on HCCI combustion timing Vary the Temperature Intake temperature and pressure
Vary the Chemistry Fuel Autoignition quality
Compression ratio Air/fuel ratio* EGR (internal, external)** *O2 concentration and specific heat (k) variation •** Five effects of EGR: charge heating, dilution, heat capacity, chemical , and stratification (SAE 2001-01-3607)
Introduction
(3)
Influential parameters on HCCI combustion timing Vary the Temperature Intake temperature and pressure
Vary the Chemistry Fuel Autoignition quality
Compression ratio Air/fuel ratio* EGR (internal, external)** Dual Fuel HCCI Engine
Introduction Examples of Dual Fuel HCCI Engine
A blend of low octane/high octane fuels is common
Iso-octane, n-Heptane (SAE 1999-01-3679) Ethanol, n-Heptane (SAE 2001-01-1896) DME, Methanol (SAE 2004-01-2993) DME, methanol syngas (SAE 2003-01-1824)
A blend of high octane fuel and hydrogen or syngas
Natural gas, hydrogen (SAE 2004-01-1972)
(4)
Introduction
(5)
Dual Fuel HCCI Engine
Advantages
Disadvantages
Possibility of cycle-by-cycle control
Carrying two fuels onboard
Wider operating window
Infrastructure
Beneficial for dual mode HCCI/SI engine
Introduction
(5)
Dual Fuel HCCI Engine
Advantages
Disadvantages
Possibility of a cycle-bycycle control
Carrying two fuels onboard
Wider operating window
Infrastructure
Beneficial for dual mode HCCI/SI engine Onboard partial reforming of base fuel to produce reformer gas (RG) for blending
Introduction
(6)
Reformer gas (RG)- Part 1
A mixture of light gases dominated by H2 and CO Can be produced onboard with a fuel processor from hydrocarbons or alcohols Various techniques: Partial oxidation Steam reforming Autothermal reforming
Introduction
(7)
Reformer gas (RG)- Part 2
Depending on technique, base fuel, catalysts, temperatures, etc … H2 concentration may vary Some other gases may present There is an efficiency loss incorporated with reforming
Autothermal reformer efficiency is 78%~84%*
*Docter and Lamm J. of Power Sources, 1999, 84, 194-200
Introduction
(8)
Application of Reformer gas in HCCI engine
Three fuel categories have been investigated
Natural gas because of industrial applications High octane fuels (PRF100, PRF80)* suitable for Gasoline / HCCI engines Low octane fuels (PRF0, PRF20)* suitable for Diesel / HCCI engines
*PRF = Primary Reference Fuel, iso-Octane/n-Heptane blend
Introduction
(9)
HCCI combustion engine fueled with n-Heptane
Low-octane HCCI is easier to achieve and more practical than high octane fuel Double-stage ignition process happens in the engine time scale Varying EGR mainly controls ignition timing, while varying λ controls combustion duration*
* Peng et al. SAE paper 2003-01-0747
Introduction
(10)
Research target, methodology
Control combustion timing of n-heptane fueled HCCI engine independent of EGR and λ. Use blended dual-fuel concept, taking advantage of double-stage combustion and reaction inhibition Experimental work on a research engine, combustion diagnosis, confirmation of results with a chemical kinetic model. Use the model to examine chemistry alteration.
Outline
Introduction Low octane HCCI combustion, Research target and methodology, Reformer gas Experimental setup Definitions Experimental Results Operating region, pressure trace characteristics, heat release analysis Modeling Heat release, temperature, intermediate species Conclusions
Experimental Setup
(11)
Engine and Instrumentation T
EGR
RG
Heater
P
Throttle
P
Fuel
Feedback to heater
Constant Tmixture
T
T
CFR Engine
Air Air Mass Flow Meter
Experimental Setup
(12)
Operating Condition
The engine was not motored Steady state constant speed of 700 RPM Wide open throttle, natural aspiration Simulated RG mixture: H2 (75%) and CO (25%) Other inert gases in RG compensated by EGR Electrical intake heater controlled to provide constant intake temperature AFTER fuel injection and EGR.
Outline
Introduction Low octane HCCI combustion, Research target and methodology, Reformer gas Experimental setup Definitions Experimental Results Operating region, pressure trace characteristics, heat release analysis Modeling Heat release, temperature, intermediate species Conclusions
Definitions
(13)
Engine operating parameters
λ was calculated based on the both fuel and reformer gas Tintake is the temperature of the mixture entering the cylinder (air, fuel, RG, EGR) & RG m RG mass fraction (%) = 100 × & RG + m & fuel m
Definitions
(14)
Example of RG replacement effect
N=700 RPM, EGR=0%, WOT, Heptane, λ=2.0 Heptane flow rate reduction 0% 23%
RG mass fraction 0% 30%
Air flow reduction 0% 3% Energy flow reduction 0% 1%
Definitions
(15)
Combustion Diagnosis Parameters Gross cumulative heat release HTRmax
Rate of heat release
GHRmax
D
90% GHRmax
/C A
NTC
RH R
=0 J
Combustion Duration (CD)
LTRmax
NTC -35
-25 -15 CAD, aTDC LT R tim
in g
-5 HT R
tim
10% GHRmax (SOC)
in g
-40
LTR: Low Temperature Reaction HTR: High Temperature Reaction NTC: Negative Temperature Coefficient SOC: Start of Combustion CD: Combustion Duration GHR: Gross Heat Release
-30
-20 CAD, aTDC
-10
0
Outline
Introduction Low octane HCCI combustion, Research target and methodology, Reformer gas Experimental setup Definitions Experimental Results Operating region, pressure trace characteristics, heat release analysis Modeling Heat release, temperature, intermediate species Conclusions
Experimental Results
(16)
RG effect on operating region boundaries RG addition at any fixed EGR rate pushed back the knock boundary toward richer mixture
1.6
n-Heptane base fueled CR=9.5 Intake Temperature =100 oC N=700 RPM
no RG added operating region
RG
2
λ
in c rea se
2.4
RG-enriched operating region
1.2
0
10
20 30 EGR (%)
40
50
Experimental Results
(17)
RG effect on pressure characteristics 48
Increasing RG fraction decreased maximum pressure and maximum rate of pressure rise
n-Heptane fueled CR=11.5 λ=1.24 ±0.02 EGR=40%
Pmax 44 6 (dP/dθ)max 40 4 36
2
32
0
5 10 15 RG mass fraction (%)
20
cyl. max. pressure (bar)
cyl. max. pressure rate (bar/CAD)
8
Experimental Results
(18)
*NTC region is period between first-stage and second-stage (main) ignition.
4 h ea
t re
leas
e ra
tio
2
24
TC
22
20
18 0
d
3
1 st/2 n
1
CR=11.5 λ=2.03 ±0.02 EGR=10%
26
0
Increasing RG prolonged NTC* and reduced 1st to 2nd rate of heat release ratio
N
negative temp. coef., NTC (CAD)
28
5 10 15 RG mass fraction (%)
20
25
LTRmax to HTRmax rate of heat release ratio (%)
5
RG effect on heat release characteristics
Experimental Results
(19)
RG effect on combustion characteristics
5 1 2 3 4 combustion duration, CD (CAD)
CR=11.5 λ=2.03 ±0.02 EGR=10%
C
-4
-8
-12
0
Increasing RG fraction retarded the main combustion and had no significant impact on combustion duration
CD
SO
start of combustion, SOC (CAD, aTDC)
0
0
5 10 15 RG mass fraction (%)
20
25
Experimental Results
(20)
RG effect on rate of heat release
Increasing RG fraction retarded the main combustion and had no significant impact on combustion duration
CR=11.5 λ=1.24 ±0.02 EGR=40%
120
80
40
0
-40
-20
0 CAD, aTDC
20
rate of heat release (J/CAD)
RG=0% RG=7% RG=10% RG=12% RG=20%
Experimental Results
(21)
RG effect on efficiencies
CR=11.5 λ=2.03 ±0.02 EGR=10%
100 e al m er h t
cy n e ci i f f
25
98 20 combustion efficiency
96
15 0
5 10 15 RG mass fraction (%)
20
thermal efficiency (%)
Increasing RG fraction slightly increased thermal efficiency and decreased combustion efficiency (This study does not take RG reforming efficiency and intake heater power into account )
combustion efficiency (%)
Outline
Introduction Low octane HCCI combustion, Research target and methodology, Reformer gas Experimental setup Definitions Experimental Results Operating region, pressure trace characteristics, heat release analysis Modeling Heat release, temperature, intermediate species Conclusions
(22)
Modeling Chemical kinetic model * ,**
Single Zone Semi-reduced mechanism of 290 reactions and 57 species for n-heptane *** SOC, LTR, and NTC timing is predicted precisely *Kongsereeparp et al, 2005, Fall Technical Conference of the ASME ICE Division, Ottawa, Canada
** Kongsereeparp et al ,SAE 2007-01-0205
25
20
Pressure (bar)
15 Experiment 10 ChemComb-SZM
5
0 rate of heat release -5 -60
***Mechanisms by Dr. V Golovichev,
Chalmers University of Technology, http://www.tfd.chalmers.se/~valeri/MECH/chem.inp_c7h16
-50
-40 -30 CAD, aTDC
-20
-10
Modeling
(23)
RG effect on rate of heat release RG = 0 % RG = 5 % RG = 10 % RG = 17 % RG = 23 %
With increasing RG replacement fraction, the model showed very similar effects to the engine experiments.
CR=11.5 Intake temp =373 K N=700 RPM
λ=2.02
rate of heat release, RHR (J/CAD)
250
200
150
100 RG mass fraction increase
50
0
-50 -35
-30
-25 -20 CAD,aTDC
-15
-10
Modeling
(24)
RG effect on end-of-compression and combustion temperatures 860
CR=11.5 Intake temp =373 K N=700 RPM λ=2.02
840 Temperature (K)
Increasing RG fraction raises end-of-compression temperature but lowers combustion temperature
n ctio fra ss e ma eas RG incr
820
800 RG = 0 % RG = 5 % RG = 10 % RG = 17 % RG = 23 %
780
760 -29
-28
-27 CAD, aTDC
-26
-25
Modeling
(25)
HCCI combustion of nheptane without RG
RHR (J/CAD)
O H2
C
OH appears at the main ignition stage and is present in combustion products. H2O2 and CH2O appear at the first-stage ignition and disappear at the main ignition stage.
O2 H2
RHR
OH, H2O2, and CH2O mol fraction (mol/mol)
Important intermediate species
OH
-35
-30
-25 -20 -15 CAD, aTDC
-10
-5
Modeling
0.004
(26) 0.002
H2O2 production rate
RG = 0 % RG = 5 % RG = 10 % RG = 17 % RG = 23 %
Increasing RG fraction -reduces H2O2 production rate at the 1st stage, -delays but amplifies H2O2 production at the 2nd stage of combustion
-27 d[H2O2]/dt (mole/cc-S)
0
0.08
0.06
CR=11.5 Intake temp =373 K N=700 RPM λ=2.02
-25
LTR 0.04
HTR
0.02
-26 CAD, aTDC
RG
ma
a cti r f ss
on
ase e r in c
0 -15
-14
-13 CAD, aTDC
-12
RG mass increase
0.08
ate r l o ta
1200
1600 T (K)
#123:
e
HO2+ HO2Ù H2O2 + O2 OH + OH (+M) Ù H2O2 (+M)
re de ac st tio ru n ct 12 io 3 n ra te
RG = 23 % RG = 17 % RG = 10 % RG = 5 % RG = 0 % 800
#122:
re pr act od io uc n 1 tio 22 n ra t
0 -0.08
t
-0.16
CR=11.5 Intake temp =373 K N=700 RPM λ=2.02
RHR
RG mass increase
d[H2O2]/dt (mole/cc-S)
H2O2 production / destruction by key reactions
(27)
0.16
Modeling
2000
0.002
c in as re
0.001
e
RG = 0 % RG = 5 % RG = 10 % RG = 17 % RG = 23 %
0
H2O2 mole fraction (molH2O2/moltotal)
n
tio ac fr
s
CR=11.5 Intake temp =373 K N=700 RPM λ=2.02
as m
Overall, total molar concentration of H2O2 is reduced, particularly during the period between 1st and 2nd stage ignition
G
(28)
R
H2O2 mole fraction
0.003
Modeling
-28
-24
-20 CAD, aTDC
-16
-12
Outline
Introduction Low octane HCCI combustion, Research target and methodology, Reformer gas Experimental setup Definitions Experimental Results Operating region, pressure trace characteristics, heat release analysis Modeling Heat release, temperature, intermediate species Conclusions
Conclusion
(29)
Experimental study showed that RG can effectively control ignition timing independent of λ and EGR, (ie. timing control at a essentially constant load). Increasing RG mass fraction:
Did not change LTR timing (1st stage ignition). Reduced LTRmax (magnitude of 1st stage ignition) Prolonged NTC (period between 1st stage & main ignition) Delayed HTR (time of main stage ignition)
Effect of RG on engine operating parameters was:
Expanded operating range toward richer mixture, pushing back knocking limit Higher thermal efficiency and lower combustion efficiency at any constant λ and EGR.
Conclusion
(30)
A Single-zone heat release model predicted SOC precisely and showed the same effect of RG on combustion timing as experiments. Increasing RG fraction raised end-of-compression temperature and decreased post-combustion temperature. Increasing RG fraction accelerated both production and destruction rates of critical intermediate species but total molar concentration was lower, particularly in the period between 1st stage and main ignition.
Thank you for your attention QUESTIONS? Vahid Hosseini
[email protected]
The contribution of the Auto21 National Center of Excellence to supporting this work is gratefully acknowledged.
