Energy: Horng-Wen Wu, Zhan-Yi Wu
Energy: Horng-Wen Wu, Zhan-Yi Wu
Energy: Horng-Wen Wu, Zhan-Yi Wu
Energy
journal homepage: www.elsevier.com/locate/energy
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 19 April 2012
Received in revised form
8 September 2012
Accepted 10 September 2012
Available online 6 October 2012
This study applies the L9 orthogonal array of the Taguchi method to nd out the best hydrogen injection
timing, hydrogen-energy-share ratio, and the percentage of exhaust gas circulation (EGR) in a single DI
diesel engine. The injection timing is controlled by an electronic control unit (ECU) and the quantity of
hydrogen is controlled by hydrogen ow controller. For various engine loads, the authors determine the
optimal operating factors for low BSFC (brake specic fuel consumption), NOX, and smoke. Moreover, net
heat-release rate involving variable specic heat ratio is computed from the experimental in-cylinder
pressure. In-cylinder pressure, net heat-release rate, A/F ratios, COV (coefcient of variations) of IMEP
(indicated mean effective pressure), NOX, and smoke using the optimum condition factors are compared
with those by original baseline diesel engine. The predictions made using Taguchis parameter design
technique agreed with the conrmation results on 95% condence interval. At 45% and 60% loads the
optimum factor combination compared with the original baseline diesel engine reduces 14.52% for BSFC,
60.5% for NOX and for 42.28% smoke and improves combustion performance such as peak in-cylinder
pressure and net heat-release rate. Adding hydrogen and EGR would not generate unstable combustion due to lower COV of IMEP.
2012 Elsevier Ltd. All rights reserved.
Keywords:
Port-injecting hydrogen
A diesel engine
Taguchi method
Injection timing
EGR percentage
Combustion performance
1. Introduction
A diesel engine is one of the most effective energy conversion
systems. It is widely applied to many power plants, such as a power
source for land vehicles, commercial marine vessels, stationary
power plants, and so on [1]. However, because of the ever stricter
emissions regulation, it is necessary to nd new ways of decreasing
exhaust emissions from engines and on alternative fuels for solving
these crises. Hydrogen is expected to be an ideal alternative for
fossil fuel systems [2] and also considered to meet energy, environment and sustainable development needs [3]. The following
descriptions are the characteristics of hydrogen, such as it gains
much higher ame speed and larger diffusion speed [4] so it can
benet the energy efciency and emissions. In addition, it is a long
term renewable, recyclable and non-polluting fuel since it is
without carbon [5]. The ammability limits of hydrogen vary from
an equivalence ratio of 0.1e7.1 [6], and the engine is hence operated
with a wide range of air/fuel ratio. The lower heating value of
hydrogen is 120 MJ/kg which is higher than that of diesel fuel
(42 MJ/kg). The auto-ignition temperature of hydrogen (858 K) is
412
using the EGR (exhaust gas circulation) (0e25%) with 7.5 lpm
hydrogen ow rate increased the brake thermal efciency, and
signicantly reduced NOX, but the exhaust gas emission
increases by 40% with 20% EGR compared with 0% EGR at full
load [16]. In addition, NOX increased without EGR with 7.5 lpm
hydrogen ow rate. Moreover, James et al. [17] evaluated a series
of commercially available natural gas fuel injectors, originally
designed for heavy-duty diesel application, for use with hydrogen
fuel in an electronic fuel-injected internal combustion engine.
They concluded that performance of the injectors evaluated was
within the necessary tolerances for hydrogen applications with
internal combustion engines. Saravanan et al. [18,19] studied on
port injection of hydrogen and varied injection timing, injection
duration and ow rates of hydrogen for various loads. Their results
showed that using port-injected hydrogen increased engine efciency with a greater reduction in emissions, especially the start of
injection at 5 ATDC. However, NOX increased while the hydrogen
ow rate increased at higher loads. Adam et al. [20] discussed the
performance and emission analysis of hydrogen fueled compression ignition engine with variable water injection timing. The
hydrogen was injected at the intake port at xed injection timing
from 0 CA to 40 CA and constant ow rate of 5 lpm with water
injected from 20 BTDC to 20 ATDC within injection duration of
20 CA and 40 CA. The results indicated that water injection timing
of 20 ATDC and duration of 20 CA had showed better engine
performance due to increased gross indicated work and indicated
thermal efciency.
The experimental design using Taguchi method has been
applied in the industry for nding factors that are necessary for
achieving objectives. Lu et al. [21] used the Taguchi method to
determine the optimum design of natural-circulation solar-waterheater. Wu and Ku [22] used this method for experiments to
acquire the primary optimum setting of the operating parameters
in proton exchange membrane fuel cells (PEMFCs). As Nataraj et al.
[23,24] proved the Taguchi method to be a useful technique for the
simultaneous optimization of several engine parameters and also
for predicting the effect of various design parameters on diesel
emissions.
Furthermore, Saravanan et al. [25] investigated the combined
effect of fuel injection timing, EGR ratio, and fuel injection pressure
in controlling the NOX emission of a stationary diesel engine with
diesel by Taguchis L9 orthogonal arrays. Ganapathy et al. [26]
expressed the view that Taguchis approach based on thermodynamic model application is extended to study the suitability of
different types of biodiesel under research for optimization of
engine performance. Applying orthogonal arrays and reliable
quality Taguchi method reduces the number of experimental runs.
The main advantage of the Taguchi method for optimization is its
use of orthogonal arrays for design simplifying the task of planning
experiments greatly.
From the above papers cited, the use of hydrogen as a fuel in
a diesel engine has been studied; however, the authors found that
most of the experiments with adding hydrogen into the engine
discussed how the ow rate of hydrogen affected engine performance and emissions at a xed EGR ow rate by full-factorial
experiments. In addition, there are barely previous investigations
to study combustion effect and emission of a hydrogen/diesel dual
fuel engine by Taguchi method. Therefore, the authors use the
Energy-Share method [27] (it means the same total fuel energy
input) to investigate the engine emissions on a single-cylinder
diesel engine. Based on this method, brake specic energy
consumption (BSEC) is almost the same at a xed load and engine
speed. As a result, brake specic fuel consumption (BSFC) is discussed instead of BSEC to know how the hydrogen fuel affects the
BSFC of the diesel fuel. As far as the diesel engine, the NOX and
413
Table 1
The specications of the engine.
Bore stroke
Displacement volume
Compression ratio
Rated output
Rated speed
Injection pressure of diesel fuel injector
Start of injection
Injection pressure of hydrogen injector
94 90 mm
624 cm3
18
9.2 kW
2400 rpm
21.57e22.56 MPa
21.5 w 23.5 BTDC
250 kPa
Table 2
Measurement uncertainties.
Item
Uncertainty (%)
Pressure
Hydrogen ow rate
NOX
Smoke
0.5
1.5
1.0
2.0
414
EH2
EH2 Ediesel
(1)
where EH2 is the lower heating value (MJ/kg) of hydrogen multiplied by the mass rate of hydrogen (kg/min) and Ediesel the lower
heating value of diesel multiplied by the mass rate of diesel.
(EH2 Ediesel) is always constant at each load.
EGR is a NOX reduction technique used in most gasoline and
diesel engines. EGR works by recirculating a portion of an engines
exhaust gas back to the engine cylinders. The denition of EGR ratio
[29] in this study is given as:
EGR%
VEGR
V V V
_ f g=h
m
BPkW
(3)
dQ
1
dV
dp
pV dk
kp
V
J=CA
k1
dq
dq
dq
k 12 dq
(4)
Air=Fuel ratioA=F
_a
m
_f
m
(5)
COV
1
S=N ratio 10log
n
n
X
1
y2
i1 i
s
y
q
P
2
N 1$ N
i 1 yi y
(7)
(2)
EGR
airthe volume
f
where VEGR
is
of engines exhaust gas per cycle (m3/
cycle), Vair the volume of the intake air per cycle (m3/cycle), and Vf
the volume of the fuel per cycle (m3/cycle).
BSFC [1] is the ratio of fuel mass ow rate of an engine to its
output power given by the following equation.
BSFCg=kW$h
(6)
Table 5
Contribution of factors for emissions at various loads.
No.
1
2
3
4
5
6
7
8
9
1
1
1
2
2
2
3
3
3
1
2
3
1
2
3
1
2
3
1
2
3
2
3
1
3
1
2
1
2
3
3
1
2
2
3
1
Table 4
Factors with levels.
A
B
C
415
Factors
Level 1
Level 2
Level 3
10 BTDC
5%
20%
TDC
10%
30%
10 ATDC
20%
40%
Contribution (%)
Factors
30% load
A
B
C
45% load
A
B
C
60% load
A
B
C
BSFC
NOX
Smoke
e
50.5
49.2
0.64
0.52
97.65
33.68
19.41
34.84
e
84.16
8
0.14
0.62
99.12
e
66.6
26.2
e
89.96
9.71
0.67
e
98.2
e
90.42
8.46
Table 6
Summary table for optimal conditions at various loads.
Factors
Levels
S/N ratio
for BSFC
49.51
49.35
48.76
49.59
49.21
48.82
S/N ratio
for NOX
Overall
optimum
A2
41.98
39.57
35.44
14.83
13.51
13.91
14.60
14.10
13.56
14.81
13.98
13.45
40.18
39.98
39.79
48.11
47.84
47.14
B3
C3
42.91
42.65
42.05
46.94
43.43
37.23
C3
A3
18.75
17.25
15.49
44.41
40.95
34.58
47.57
47.11
46.14
47.18
46.94
46.70
B3
A3
25.47
23.94
20.90
22.72
23.43
24.16
B3
C3
416
-48.4
S/N (dB)
-48.8
-49.2
-49.6
-50
-36
-40
-44
A1
A2
A3
B1
B2
B3
C1
C2
C3
A1
-12
A2
B2
B3
C1
C2
C3
B1
B2
B3
C1
C2
C3
B1
B2
B3
C1
24
S/N (dB)
S/N (dB)
B1
CO at 30% load
22
20
-15
18
-16
A1
A2
A3
B1
B2
B3
C1
C2
A1
C3
A2
A3
-11.6
-34
HC at 30% load
-12
S/N (dB)
-35
S/N (dB)
A3
26
-36
-37
-12.4
-12.8
-38
-13.2
A1
A2
A3
B1
B2
B3
C1
C2
C3
A1
A2
A3
C2
C3
-32
-46.8
S/N (dB)
-47.2
-47.6
-48
-44
-48.4
-48
A1
A2
A3
B1
B2
B3
C1
C2
C3
A1
-14
A2
A3
B1
B2
B3
C1
C2
B2
B3
C1
C2
B2
B3
C1
C2
C3
24
Smoke at 45% load
CO at 45% load
22
-16
S/N (dB)
S/N (dB)
-40
-18
20
18
16
14
-20
A1
A2
A3
B1
B2
B3
C1
C2
A1
C3
-35
A3
B1
C3
-13
HC at 45% load
-36
S/N (dB)
A2
-37
-38
-14
-15
-39
-16
-40
A1
A2
A3
B1
B2
B3
C1
C2
C3
A1
A2
A3
B1
C3
-45.6
S/N (dB)
-46.2
-46.8
-47.4
-42
-46
-48
-50
A1
A2
A3
B1
B2
B3
C1
C2
C3
A1
-20
A3
B1
B2
B3
C1
C2
C3
B2
B3
C1
C2
C3
CO at 60% load
20
-22
S/N (dB)
S/N (dB)
A2
25
Smoke at 60% load
-24
15
10
5
-26
A1
A2
A3
B1
B2
B3
C1
C2
A1
C3
-35
A2
A3
B1
-14.5
HC at 60% load
-15
S/N (dB)
-37
S/N (dB)
417
-39
-41
-15.5
-16
-16.5
-43
-17
A1
A2
A3
B1
B2
B3
C1
C2
C3
A1
A2
A3
B1
B2
B3
C1
C2
C3
v
#
"
u
u
1
1
CI tFa;1;fe Ve
Neff Re
(8)
and
Neff
N
1 TDOF
(9)
BSFC (g/kWh)
400
300
200
100
0
30%
45%
60%
Load(%)
Fig. 6. Comparison of BSFC between original baseline diesel engine and optimized
engine.
Fig. 7. Comparison of in-cylinder pressure between original baseline diesel engine and
optimized engine.
418
5%
70
60
50
COV of IMEP(%)
40
4%
3%
2%
1%
0%
30
30%
45%
60%
Load(%)
20
Fig. 10. Comparison of COV of IMEP between original baseline diesel engine and
optimized engine.
10
300
0
360
380
400
420
440
460
-10
Crank angle (degree)
Fig. 8. Comparison of net heat-release rate between original baseline diesel engine
and optimized engine.
250
NOx(ppm)
340
200
150
100
50
0
30%
60%
Load(%)
Fig. 11. Comparison of NOX emission between original baseline diesel engine and
optimized engine.
60
25
baseline diesel engine
20
smoke(%)
50
A/F ratio
45%
40
30
15
10
5
0
20
30%
45%
60%
Load(%)
Fig. 9. Comparison of A/F ratios between original baseline diesel engine and optimized
engine.
30%
45%
60%
Load(%)
Fig. 12. Comparison of smoke emission between original baseline diesel engine and
optimized engine.
Similarly, in Fig. 12, the smoke emission with the optimized engine
reduces a lot compared with baseline diesel engine. At 45% load, the
reduction rate reaches to about 53%. The results indicate that for the
optimized engine there is signicant reduction in both NOX and
smoke emissions.
5. Conclusions
After obtaining the optimal operating factors for low BSFC, NOX,
and smoke of diesel/hydrogen mixtures with Taguchi method, and
comparing BSFC, NOX, smoke, and combustion performance between
the optimized engine and baseline diesel engine, the authors
conclude on the basis of the results and discussion as follows.
The Taguchi method was found to be a good method to nd
optimum combination for low BSFC, NOX, and smoke in a hydrogen/
diesel dual fuel engine. The predictions using Taguchis parameter
design technique agreed adequately with the conrmation results
on 95% condence interval. Using Taguchi method saves 67%
experiment run number in this research.
According to the analysis of S/N curves and ANOVA, hydrogenenergy-ratio affects BSFC and smoke more than other factors;
EGR affects NOX signicantly; hydrogen injections timing does not
have important effects on the objectives.
Compared with original baseline diesel engine, for BSFC, the
higher hydrogen ratio replaces much more amount of the diesel
fuel, and higher EGR reduces more oxygen content of air, so the best
combination can reduce BSFC at least 14.52% at various loads. For
NOX, because of the EGR technology, the best combination can
reduce more than 60.5% at various loads. For smoke, since there is
absence carbon of hydrogen in spite of EGR causing more smoke, the
best combination can still reduce at least 42.28% at various loads.
The best combination also improves combustion performance
as a result of the faster combustion in hydrogen than in diesel with
EGR. For in-cylinder pressure, it raises 13%e17% peak in-cylinder
pressure in the optimized engine. Net heat-release rate is higher
at the optimized engine than that at the baseline diesel engine at
various loads. For COV of IMEP, the best combination does not cause
unstable combustion.
Acknowledgment
The authors gratefully acknowledge the support of the National
Science Council of Taiwan, ROC, under grant NSC 98-2221-E-006259-MY3. Also, the authors would like to thank Dr. M Nataraj who
provided valuable papers.
Appendix
Nomenclature
A/F
ANOVA
BP
CA
CI
COV
DOF
E
EGR
F
IMEP
k
_
m
air/fuel ratios
analysis of variance
brake power, kW
crank angle, degrees
condence interval
coefcient of variance, %
degree of freedom
energy, J
exhaust gas recirculation, %
F-ratio which is a test statistic for multiple independent
variables
indicated mean effective pressure, bar
specic heat capacity, J kg1 K1
mass ow rate, g h1
N
n
NOX
p
ppm
Q
R
Re
rpm
S/N
T
TDOF
419
q
s
Subscript
air
e
eff
f
H2
i
air
error
effective
fuel
hydrogen
species
V0
V
Ve
y
Y
References
[1] Heywood B. Internal combustion engine fundamentals. New York, U.S.A:
McGraw-Hill Book Company; 1988.
[2] Soid SN, Zainal ZA. Spray and combustion characterization for internal
combustion engines using optical measuring techniques e a review. Energy
2011;36:724e41.
[3] Li YL, Chen HS, Zhang XJ, Tan CQ, Ding YL. Renewable energy carriers:
hydrogen or liquid air/nitrogen? Appl Thermal Eng 2010;30:1985e90.
[4] Gomes Antunes JM, Mikalsen R, Roskilly AP. An investigation of hydrogenfuelled HCCI engine performance and operation. Int J Hydrogen Energy
2008;33:5826e8.
[5] Kawahara ETN, Piao Z, Fujita S. Hydrogen combustion and exhaust emissions
ignited with diesel oil in a dual fuel engine. SAE TECH paper 2001-01-3503;
2001. USA.
[6] White CM, Steeper RR, Lutz AE. The hydrogen-fueled internal combustion
engine: a technical review. Int J Hydrogen Energy 2006;31:1292e305.
[7] Hari Ganesh R, Subramanian V, Balasubramanian V, Mallikarjuna JM, Ramesh A,
Sharma RP. Hydrogen fueled spark ignition engine with electronic controlled
manifold injection e an experimental study. Renew Energy 2008;33:1324e33.
[8] Subramanian V, Mallikarjuna JM, Ramesh A. Effect of water injection and
spark timing on the NO emissions and combustion parameters of a hydrogen
fuelled spark ignition engine. Int J Hydrogen Energy 2007;32:1159e73.
[9] Subramanian V, Mallikarjuna JM, Ramesh A. Intake charge dilution effects on
control of NO emission in a hydrogen fueled SI engine. Int J Hydrogen Energy
2007;32:2043e56.
[10] Rakopoulosb CD, Scotta MA, Kyritsisa DC, Giakoumis EG. Availability analysis
of hydrogen/natural gas blends combustion in internal combustion engines.
Energy 2008;33:248e55.
[11] Wang S, Ji C, Zhang J, Zhang B. Comparison of the performance of a sparkignited gasoline engine blended with hydrogen and hydrogen-oxygen
mixtures. Energy 2011;36:5832e7.
[12] Yuksel F, Ceviz MA. Thermal balance of a four stroke SI engine operating on
hydrogen as a supplementary fuel. Energy 2003;28:1069e80.
[13] Mills Randell L, Zhao Guibing, Good William. Continuous hydrino thermal
power system. Appl Energy 2011;88:789e98.
[14] Bysveen M. Engine characteristics of emissions and performance using
mixtures of natural gas and hydrogen. Energy 2007;32:482e9.
[15] Saravanan N, Nagarajan G. An experimental investigation of hydrogenenriched air induction in a diesel engine system. Int J Hydrogen Energy
2008;33:1769e75.
[16] Saravanan N, Nagarajan G. Experimental investigation on a Di dual fuel engine
with hydrogen injection. Int J Energy Res 2009;33:295e308.
420