Measurement and simulation of pollutant emissions from marine diesel

combustion engine and their reduction by water injection

Nader Larbi
a,
*
, Jamel Bessrour
b,1
a
03 rue Kacem Amine, 2040 Rads, Tunisia
b
Ecole Nationale dIngnieurs de Tunis, BP 37, 1002 Tunis Belvdre, Tunisia
a r t i c l e i n f o
Article history:
Received 13 March 2007
Received in revised form 14 February 2008
Accepted 20 August 2008
Keywords:
Engine combustion
Numerical modeling
NO
x
and SO
x
emissions
Detailed chemical kinetic
a b s t r a c t
Taking into account the complexity and cost involved to conduct an experimental investment, the
recourse to a tool of simulation, which in turn entails access to information by measurement, offers an
effective and fast alternative to deal with the problem of pollutant emissions from internal combustion
engines. An analytical model based on detailed chemical kinetics employed to calculate the pollutant
emissions of a marine diesel engine gave results, in general, satisfactory compared to experimentally
measured results. Especially the NO emission contents are found higher than the standards limiting val-
ues set out by the International Maritime Organisation (IMO). Thus, this study is undertaken in order to
reduce as much as possible these emissions.
The reduction of pollutant emissions is apprehended with water injection.
2010 Elsevier Ltd. All rights reserved.
1. Introduction
Energy produced by powerful diesel engines is largely used in
marine propulsion, because of their favourable reliability and their
signicant output. However, the increasingly constraining legisla-
tions, aiming to limiting the pollutant emissions from the exhaust
gas produced by these engines, tend to perform their supremacy
[1,2].
Many studies were undertaken, on the experimental and ana-
lytical level as well, in order to study the mechanisms which
govern the generation of the various produced pollutants [3
5]. The analysis of the pollutant emissions and their reduction
in the exhaust gas of the medium speed turbocharged
marine diesel engine, constitutes the principal objective of this
study.
With advanced research and in spite of the computers high
capacity, it is still impossible, in combustion, to treat at the same
time complex kinetics and industrial turbulent ow conditions.
The majority of the studies treated the detailed chemistry of com-
bustion in association with ideal ow reactor conditions [5,6]. The
various simulations carried out in this work are based on the com-
puter code marketed by Reaction Design, the CHEMKIN package,
which is developed by the Sandia laboratories.
Comparisons of the numerical predictions with the experimen-
tal results carried out on a real unit at use aboard a car ferry ship
made it possible to analyze the validity of the numerical results.
2. Experimental study
The unit selected for measurements is a power generating unit
in operating on a car ferry ship during its docking in the harbour.
The diesel engine unit is a WARTSILA NSD type 6R32 LNE having
characteristic given in Table 1 with a nominal power of 2460 kW.
The measured values taken in exhaust gas were carried out using
a Testo350 (Precision 5% m.v.) gas analyzer which made it possi-
ble to timely record with accuracy the % O
2
, % CO
2
, NO, SO
2
, tem-
perature and pressure.
The measurements were carried out for various powers: 640,
900, 1370 and 1470 kW.
In order to determine the emitted quantities in grams per cycle
per cylinder (g/cycle/cylinder) and in (g/kW h) it is necessary to
dene the volume (V
m
) of the exhaust gas at the measurement
point at atmospheric pressure (P
a
) and for exhaust gas temperature
(T
m
)
V
m

m
T
R
g
T
m
P
a
1
where the volume is calculated from the ideal gas equation of state.
The mass conservation law enables us to deduce the mass with
the admission condition
m
T
m
air
m
comb
where it is assumed m
air
15k
a
m
comb
2
0965-9978/$ - see front matter 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.advengsoft.2010.01.004
* Corresponding author. Tel.: +216 97 520 477.
E-mail addresses: naderlarbi@yahoo.fr (N. Larbi), jamel.bessrour@enit.rnu.tn (J.
Bessrour).
1
Tel.: +216 71 874 700.
Advances in Engineering Software 41 (2010) 898906
Contents lists available at ScienceDirect
Advances in Engineering Software
j our nal homepage: www. el sevi er . com/ l ocat e/ advengsof t
The volume of exhaust gas at the measurement point is then
V
i
V
cyl
T
i
T
adm
P
adm
P
a
1
1
15k
a

3
Fig. 1 presents the preliminary experimental results obtained on the
chosen unit. As a consequence of the increase in power we note:
- A greater quantity of exhaust gas increasing the mode of the
turbocharger and consequently the pressure of overfeeding air.
- An increase in overfeeding air temperature.
- A decrease in % O
2
with an increase % CO
2
.
- An increase in NO emission.
- An increase in SO
2
emission.
According to these notes, we can conclude that there is an
improvement of combustion for the passage from the low power
to the half of full power output of the generating unit. The cycle
temperature increases causing a higher formation of thermal NO.
In addition, the admitted quantity of admission air is larger and
consequently supporting the formation of NO in greater quantity.
This last is higher than the standards prescribed by IMO. This is
the matter of concern of our study to reduce the NO emissions of
marine diesel engines.
3. Numerical study
It is often difcult to carry out a straight away parametric study
on an real unit since the combustion study is generally complex
with simultaneous presence of several physical and chemical phe-
nomena which vary from a model to another according to the aero
thermo chemical conditions of the application. For that, it is neces-
sary to be able mathematically to describe the aero thermo chem-
ical phenomena which controls the various processes present in a
real combustion chamber.
In spite of the great capability of computers, it is still impossi-
ble, in matter of combustion, to treat at the same time complex
kinetics and industrial turbulent ow conditions. For this reason,
one uses the modeling based on ideal chemical reactors with sim-
plied ows. These reactors are modeled by using several hun-
dreds of reactions [5,6].
In a diesel engine having a very good quality of fuel atomizing
and a room with high turbulence, homogeneous mixture can be
expected.
Due to the high turbulence induced by admission air through
deectors and injection system (high pressure: 450600 bar)
[8,9] in addition to the corrugated shape of the piston and head
of a diesel combustion engine that allows a good distribution in
Nomenclature
A preexponential factor
a, b and c constants (a = 0.98, b = 0.71 and c = 0.01)
C modeling parameters
C
p
thermal capacity (J kg
1
K
1
)
C
11
, C
12
and C
2
modeling parameters (C
11
= 3.28, C
12
= 0.308 and
C
2
= 0.324)
D the engine bore diameter for heat transfer (cm)
G ratio of connecting rod to crank-arm radius
E
A
activate energy (J)
FM molar fraction
FM
k
molar fraction of the kth species
h heat transfer coefcient (W m
2
K
1
)
h
k
specic enthalpy of the kth species (J kg
1
)
ICEN Internal Combustion Model Engine
IMO International Maritime Organisation
K
f
forward rate coefcient
_ m mass ow rate (kg s
1
)
M the average molecular mass (kg mol
1
)
M
k
molar mass of the kth species (g)
m
T
total mass of the reactants (g)
m
comb
mass of the fuel (g)
m
air
mass of the air (g)
m
k
mass of the kth species (g)
Nu
h
Nusselt number
P
a
atmospheric pressure (bar)
P
adm
admission air pressure (bar)
P
i
initial pressure inside the cylinder
P
motored
the motored cylinder pressure
Pr Prandtl number
PSR perfectly stirred reactor
Q heat loss (J s
1
)
Q
wall
wall reactor heat loss (J s
1
)
R equivalence ratio
R
g
universal gas constant (J kg
1
K
1
)
Re Reynolds number
S the surface area for heat transfer (cm
2
)
S
p
mean piston speed
T temperature (K)
T
adm
admission air temperature (K)
T
i
initial temperature inside the cylinder
T
m
exhaust gas temperature (K)
T
wall
chamber wall temperature (K)
V reactor volume (m
3
)
V
o
clearance volume (m
3
)
V
air
volume of the ambient air (m
3
)
V
cyl
volume swept by the piston (m
3
)
V
d
displacement volume (m
3
)
V
H
2
O
volume of water (m
3
)
V
i
the initial volume inside the cylinder (m
3
)
V
m
exhaust gas volume (m
3
)
V
swirl
Swirl velocity
Y
k
mass fraction of the kth species
Z percentage in volume of water mixed with admission
air
k gas conductivity (W m
1
K
1
)
k
a
air excess
a crank angle ()
b temperature exponent in the rate coefcient
l gas viscosity (kg m
1
s
1
)
q mass density of the mixture (kg/m
3
)
w the gas velocity
s residence time in the reactor (s)
_ x
k
molar rate of production of the kth species
inlet condition
Table 1
Characteristics of the diesel engine of the power generating unit.
Firing order 1-5-3-6-2-4
Speed, tr/mn 750
Piston speed, m/s 8.48.8
Cylinder bore, cm 32
Piston stroke, cm 35
Connecting rod length, cm 74.5
Compression ratio 12
Effective mean pressure, bar 28.5
Maximal ring pressure, bar 165
Clearance volume, cm
3
2557.7
Injection pressure, bar 450
N. Larbi, J. Bessrour / Advances in Engineering Software 41 (2010) 898906 899
combustion chamber [1013] we assume that the combustion pro-
cess can be modeled by a perfectly stirred reactor PSR.
In a PSR, the mixing in the reactor chamber is assumed intense
and, thus, it is expected that the temperature and composition in
the device are uniform through the reactor volume. A description
of the process taking place in the PSR is obtained by linking the con-
servation of mass and energy process to the generation of chemical
species within the reactor volume [4,6].
The species conservation equation is:
_ mY
k
Y

k
_ x
k
M
k
V 0 k 1; . . . ; K 4
The energy conservation equation is:
_ m
X
K
k1
Y
k
h
k
Y

k
h

k
Q 0 5
Admission air temperature
300
310
320
330
P ( kW)
K
Admission air pressure
1
1.5
2
2.5
3
640 900 1370 1470 P ( kW)
Bar
(a) Evolution of admission air temperature
(b) Evolution of admission air pressure
Exhausts gas temperature
300
350
400
450
500
P (kW)
K
O
2
500
700
900
1100
1300
1500
1700
1900
640 900 1370 1470 P (kW)
g/Kwh
4
6
8
10
12
14
16
g ; %
g/Kwh
% O2
g
(c) Evolution of exhausts gas temperature
(d) O2 Evolution
CO
2
500
550
600
650
700
750
800
850
640 900 1370 1470 P (kW)
g/Kwh
1
2
3
4
5
6
7
8
9
g ; %
g/Kwh
% CO2
g
CO
0
1
2
3
4
5
640 900 1370 1470 P(kW)
g/Kwh
0
0.01
0.02
0.03
g
g/Kwh
g
(e) CO2 Evolution
(f) CO Evolution
NO
0
10
20
30
40
50
60
P(kW)
g/Kwh
0
0.1
0.2
0.3
0.4
0.5
0.6
g
g/Kwh
g
SO
2
0
5
10
15
20
P(kW)
g/Kwh
0
0.05
0.1
0.15
0.2
g
g/Kwh
g
(g) NO Evolution
(h) SO2 Evolution
640 900 1370 1470
640 900 1370 1470
640 900 1370 1470 640 900 1370 1470
Fig. 1. Parameters measured with gas analyzers.
900 N. Larbi, J. Bessrour / Advances in Engineering Software 41 (2010) 898906
The nominal residence time is:
s
qV
_ m
6
where the mass density q is calculated from the ideal gas equation
of state
q
PM
R
g
T
7
Temperature
0
500
1000
1500
2000
2500
Angle ()
K
R=0.6
R=0.8
Pressure
0
50
100
150
200
250
Angle ()
Bar
R=0.6
R=0.8
(a) Temperature profile (b) Pressure profile
O
2
0
0.05
0.1
0.15
0.2
0.25
Angle ()
FM
R=0.6
R=0.8
H
2
O
0
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
0.09
Angle ()
FM
R=0.6
R=0.8
(c) O2 Profile
(d) H2O Profile
CO
2
0
0.02
0.04
0.06
0.08
0.1
0.12
Angle ()
FM
R=0.6
R=0.8
NO
0
0.005
0.01
0.015
0.02
0.025
Angle ()
FM
R=0.6
R=0.8
(e) CO2 Profile
(f) NO Profile
NO
2
0
0.00005
0.0001
0.00015
0.0002
0.00025
0.0003
0.00035
0.0004
Angle ()
FM
R=0.6
R=0.8
N
2
O
0
0.000002
0.000004
0.000006
0.000008
0.00001
0.000012
0.000014
0.000016
0.000018
0.00002
Angle ()
FM
R=0.6
R=0.8
(g) NO2 Profile
(h) N2O Profile
SO
2
0
0.0002
0.0004
0.0006
0.0008
0.001
0.0012
Angle ()
FM
R=0.6
R=0.8
SO
3
0
0.00001
0.00002
0.00003
0.00004
0.00005
0.00006
0.00007
0.00008
0.00009
Angle ()
FM
R=0.6
R=0.8
(i) SO2 Profile (j) SO3 Profile
180 240 300 360 420 480 540 180 240 300 360 420 480 540
180 240 300 360 420 480 540
180 240 300 360 420 480 540
180 240 300 360 420 480 540
180 240 300 360 420 480 540 180 240 300 360 420 480 540
180 240 300 360 420 480 540
180 240 300 360 420 480 540
180 240 300 360 420 480 540
Fig. 2. Evolutions of the temperature, the pressure and the pollutant emissions in the combustion chamber.
N. Larbi, J. Bessrour / Advances in Engineering Software 41 (2010) 898906 901
From this set of (K + 1) nonlinear algebraic equations, solutions
for the temperature and mass fractions are obtained. Even though
one seeks the solution to the steady-state equations, the computa-
tional algorithm often requires a partial solution of the related
transient problem. The analogous time-dependent equations for
mass conservation of each species are:
dY
k
dt

Y
k
Y

s

_
xM
k
q
8
and the time-dependent energy conservation equation is:
C
p
dT
dt

1
s
X
K
k1
Y

k
h

k
h
k

X
K
k1
h
k
M
k
_ x
k
q

Q
qV
9
The net chemical production rate _ x
k
of each species results
from a competition between all the chemical reactions involving
that species. Each reaction proceeds according to the law of mass
action and the forward rate coefcients (k
f
) are in modied Arrhe-
nius form:
k
f
AT
b
exp
E
A
R
g
T

10
The simulation of the internal combustion with high turbulent
conditions of an engine is carried out by the CHEMKIN code using
450 elementary reactions mechanism between 77 species. Speci-
cally the model ICEN (Internal Combustion Engine Model) [7] has
been used to simulate the temporary behaviour of the engine
combustion.
The process of ignition of fuel used by the code is governed by
its self ignition temperature.
The relation between volume divided by clearance volume is:
V
cyl
V
0
1
C 1
2
G 1 cos a

G
2
sin
2
a
p h i
11
The convective heat transfer coefcient between the gas and
cylinder wall obtained from the generalized heat-transfer correla-
tion in terms of a Nusselt number.
Nu
h
aRe
b
Pr
c
12
Nu
h

hS
k
13
The heat loss is calculated at each step in time according to:
Q
wall
hST T
wall
14
The Woschni correlation [7] allows a more accurate estimation
of the average cylinder gas speed used in the denition of the Rey-
nolds number for the heat-transfer correlation. The velocity used in
the Reynolds number denition Eq. (15) is an estimation of the
average cylinder gas velocity, w, instead of the mean piston speed
Re
D

wq
l
15
To obtain the average cylinder gas velocity, Woschni proposed a
correlation that relates the gas velocity to the mean piston speed
and to the pressure rise due to combustion:
CO2
0
4
8
12
16
E900 N900 E1370 N1370 E1470 N1470
N: Numerical
g
0
100
200
300
400
500
600
700
800
E: Experimental g/kWh
g
g/kWh
Fig. 3. Evolution of numerical and experimental results for the CO
2
emission.
NO
0
0.3
0.6
0.9
1.2
N: Numerical
g
0
10
20
30
40
50
E: Experimental
g/kWh
g
g/kWh
E900 N900 E1370 N1370 E1470 N1470
Fig. 4. Evolution of numerical and experimental results for the CO
2
emission.
SO2
0
0.1
0.2
0.3
N: Numerical
g
0
2.5
5
7.5
10
12.5
E: Experimental g/kWh
g
g/kWh
E900 N900 E1370 N1370 E1470 N1470
Fig. 5. Evolution of numerical and experimental results for the CO
2
emission.
Fig. 6. Curve IMO limiting the NO
x
emissions according to the nominal speed of the
engines [14].
Fig. 7. Admission air manifold equipped with water injection system.
902 N. Larbi, J. Bessrour / Advances in Engineering Software 41 (2010) 898906
w C
11
C
12
V
swirl
S
p
" #
S
p
C
2
V
d
T
i
P
i
V
i
P P
motored
16
The chosen fuel is consisting of 86.6% in mass of carbon, 10.9%
in mass of hydrogen and 2.5% in mass of sulphur.
The numerical results are established according to pressure
which varies from 1.7 to 2.35 bar, an overfeeding air temperatures
which varies from 317 to 322 K and an equivalence ratio which
varies from 0.6 to 0.8 corresponding to the gures measured with
powers varying from 900 to 1470 kW and for a surrounding tem-
perature of 306 K.
The rst objective of the study is to analyze the inuence of the
equivalence ratio on the course of the combustion process and on
the pollutant emissions related to the real engine conditions (over-
feeding air temperature, overfeeding air pressure, engine speed, . . .).
The numerical conversion of the results FM (Molar Fraction)
into gram (m
k
) Eq. (17). for each species k (k varying from 1 to
77) is obtained while passing by the molar masses M
k
and the total
mass
Temperature
0
500
1000
1500
2000
2500
Angle ()
K
H2O=10%
H2O=0%
Pressure
0
50
100
150
200
250
Angle ()
Bar
H2O=10%
H2O=0%
(a) Temperature profile
e l i f o r p e r u s s e r P (b)
H
2
O
0
0.05
0.1
0.15
0.2
Angle ()
FM
H2O=10%
H2O=0%
O
2
0
0.05
0.1
0.15
0.2
0.25
Angle ()
FM
H2O=10%
H2O=0%
(c) H2O Profile
e l i f o r P 2 O (d)
CO
2
0
0.02
0.04
0.06
0.08
0.1
0.12
180 240 300 360 420 480 540
Angle ()
FM
H2O=10%
H2O=0%
NO
0
0.005
0.01
0.015
0.02
0.025
180 240 300 360 420 480 540
Angle ()
FM
H2O=10%
H2O=0%
(e) CO2 Profile
e l i f o r P O N (f)
NO
2
0
0.00002
0.00004
0.00006
0.00008
0.0001
0.00012
0.00014
0.00016
0.00018
180 240 300 360 420 480 540
Angle ()
FM
H2O=10%
H2O=0%
SO
2
0
0.0002
0.0004
0.0006
0.0008
0.001
0.0012
Angle ()
FM
H2O=10%
H2O=0%
(g) NO2 Profile
e l i f o r P 2 O S (h)
180 240 300 360 420 480 540
180 240 300 360 420 480 540
180 240 300 360 420 480 540
180 240 300 360 420 480 540
180 240 300 360 420 480 540
Fig. 8. Evolutions of the temperature, the pressure and the pollutant emissions.
N. Larbi, J. Bessrour / Advances in Engineering Software 41 (2010) 898906 903
X
77
k1
m
k
m
T
m
k
m
T
M
k
FM
k
P
77
k1
M
k
FM
k
in gram 17
Fig. 2 illustrates the inuence of the equivalence ratio on the
progress of the combustion and the pollutant emissions according
to the crank rotation angle. The sharp changes of the parameters
corresponds to the beginning of an early starting combustion, with
regard to the piston top dead center (0) and with a delay of igni-
tion in relation with the equivalence ratio.
For a reduction in the equivalence ratio (from R = 0.8 to 0.6) we
note that the excess of total air leads to:
COx (g/kWh)
20
120
220
320
420
520
620
720
0 5 10 15 20 25 H2O(%)
g
Fig. 10. CO
x
mass in exhausts gas.
SOx (g/kWh)
0
2
4
6
8
10
12
0 5 10 15 20 25
H2O(%)
g
Fig. 12. SO
x
mass in exhausts gas.
0 5 10 15 20 25
O
2
0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
0 5 10 15 20 25 H2O (%)
g
0
50
100
150
200
250
g/kWh
g
g/kWh
H
2
O
3
5
7
9
11
13
15
17
H2O(%)
g
100
200
300
400
500
600
700
800
900
g/kWh
g
g/kWh
(a) O2 mass evolution (b) H2O mass evolution
NO
2
0
0.005
0.01
0.015
0.02
0.025
H2O(%)
g
0
0.2
0.4
0.6
0.8
1
1.2
g/kWh
g
g/kWh
(c) NO mass evolution (d) NO2 mass evolution
0 5 10 15 20 25
0 5 10 15 20 25
0 5 10 15 20 25
NO
0
0.2
0.4
0.6
0.8
1
H2O(%)
g
0
10
20
30
40
50
g/kWh
g
g/kWh
Fig. 9. O
2
, H
2
O and NO
x
evolutions.
NOx Composition
90%
92%
94%
96%
98%
100%
0 5 10 15 20 25 H2O (%)
% % N2O
%NO2
%NO
NOx (g/kWh)
0
5
10
15
20
25
30
35
40
45
50
0 5 10 15 20 25 H2O (%)
g
Fig. 11. NO
x
composition in exhausts gas.
904 N. Larbi, J. Bessrour / Advances in Engineering Software 41 (2010) 898906
- Dilution of exhaust gas decreasing its temperature and
pressure.
- Increasing the ignition delay.
- Decreasing the total time of combustion.
- Considerable rise in nitric oxide emission as of nitric dioxide.
- Considerable reduction in the sulphur dioxide emission.
- Considerable increase in sulphur trioxide.
- Decrease in carbon dioxide.
- Decrease of water.
We can conclude that there is an improvement of the combus-
tion process.
The considerable increase of additional air and consequently
that of atmospheric nitrogen promotes the generation of huge
quantity of NO
x
to the exhaust gas.
For a poor mixture the fuel jet of a smaller quantity undergoes a
faster oxidation which inuences a delay in ignition.
The various gures expressed in g/cycle/cylinder and g/kW h is
set in considering a low caloric value of 42,000 kJ/kg and a total
output of 0.4.
The contents CO
x
consist of mainly 99.9% to CO
2
. The contents
NO
x
consist of mainly 97.5% in NO and 2.5% in NO
2
for a poor mix-
ture. Consequently the NO
x
emission undergoes an increase similar
to major species, the nitric oxide.
The contents of SO
x
consist of mainly 95% of SO
2
and 5% of SO
3
for a poor mixture. Consequently the emission of SO
x
undergoes a
small decrease.
Figs. 35 show comparisons between computed and measured
CO
2
, NO and SO
2
emitted quantities in the exhaust gas as a function
of engine power.
Relatively good agreements are observed especially for emis-
sion expressed in g/kW h. One can conclude that the model of
the adopted calculation allows qualitative and quantitative results,
in general, satisfactory.
Nevertheless it should well be noted that the results reliability
depends on the adequate determination of the accuracy of readings
and the assumption adopted in the model.
Fig. 6 shows the IMO [14] curve for the limiting NO
x
emissions
according to the nominal speed of the engine. For a speed of
750 rpm, the NO
x
is limited to a value of 12.2 g/kW h. All the values
of NO (g/kW h) in Fig. 1g, and for the various powers, are higher
than the limiting value tolerated by the IMO.
4. Study of water injection
Therefore a subsequent study has been done to reduce this NO
emission [519] by water injection (Fig. 7). The objective of this
study is to analyze the inuence of water injection [17,18] in the
combustion chamber of the diesel engine on the different pollutant
emissions.
In order to study the inuence of water injection we choose the
case of the power of 1470 kW related to admission air having a
temperature of T
adm
= 322 K and a pressure of P
adm
= 2.35 bar.
The numerical results are established according to H
2
O mass in
reagent which varies from0 to 11.6 g, and for air mass which varies
from 74.38 to 55.8 g.
The volume of H
2
O mixed with the admission air is:
V
H
2
O

ZV
cyl
100
17
and V
air

100 ZV
cyl
100
18
Z, the percentage in volume of H
2
O mixed with admission air.
Fig. 8 illustrates the inuence of water injection during combus-
tion and pollutant emissions.
It represents the variations of the pressure, the temperature and
pollutant emissions in the cylinder as a function of the crank rota-
tion angle with and without the water injection. A substantial
reduction in NO and NO
2
emissions is noted.
Fig. 9 illustrates the inuence of water injection on NO
x
emis-
sions expressed in g/cycle/cylinder and g/kW h.
The NO emission (Fig. 11) is getting down and meets the IMO
standards (index requirement). We also note that the water injec-
tion decreases the temperature of the combustion zone by absorb-
ing the vaporisation of latent heat and thereby increasing his heat-
storage capacity.
This reduction in temperature implies a lowering of the NO con-
centration, which decreases in a quasi linear way according to the
rate of water injection.
The fall in the local temperature within the fuel jet tends to de-
lay the ignition and in turn the combustion with increase in sever-
ity (the quantity of fuel injected during the time allowing a
combustion, is raised).
These two effects are compensated, thus their inuence on the
temperature of the cycle and consequently on the emissions of NO
can be considered as minor.
The CO
2
emission (Fig. 10) and SO
2
emission (Fig. 12) remains
practically constant.
5. Conclusion
To address problems of global air pollution due to the pollutant
emission from fuel oil engine combustion, it is necessary to under-
stand the mechanisms by which pollutants are produced in com-
bustion processes. In the present work, an experimental and
numerical study is carried out on a unit being actually used aboard
a car ferry ship. A numerical model based on a detailed chemical
kinetics scheme is used to calculate the emissions of CO
2
, NO and
SO
2
in an internal combustion engine model having the same par-
ticulars of an actual unit.
In general the experimental and numerical results featured
same outcomes especially at high power and for excess air. In
studying the reduction of pollutant emissions of the diesel engine
by the use of water injection, substantial decrease in NO is ob-
served which allows the unit to meet IMO regulation.
References
[1] ANPE, PAM, FEM. Atelier sur la lutte contre la pollution due des activits
menes terre, Tunis; 2005.
[2] Tauzia X. Simulation de linuence des paramtres de fonctionnement des
moteurs Diesel suraliments semi-rapides sur les missions polluantes. Ecole
doctorale sciences pour lingnieur de Nantes; 1998.
[3] Schaal R. La Cintique Chimique Homogne. Presses Universitaires de France;
1971.
[4] Bowman CT. Kinetics of pollutant formation and destruction in combustion.
Prog Energy Combust Sci 1975;1:3345.
[5] Hamdi M, Ticha Hb, Sassi M. Simulation of pollutant emissions from a gas
turbine combustor. Combust Sci Technol 2004;176:81934.
[6] Gateau P, Franco P. Application des Modles Mcanistiques de Cintique
Chimique aux combustions industrielles. Rev lInst Fran Ptrol 1991;46(3):
40719.
[7] Reaction design. Input manual, Theory manual, Chemkin collection
RD01402; 2005.
[8] Baz I. Contribution la Caractrisation de la Cavitation dans les Injecteurs
Diesel Haute Pression. Thse de Doctorat lcole Centrale de Lyon; 2003.
[9] Takahashi H, Yanagisawa H, Shiga S, Karasawa T, Nakamura H. Analysis of
high-pressure diesel spray formation in the early stage of injection.
Atomisation Sprays 1997;7:3342.
[10] Jean C. Les Moteur Diesels marins. Institut maritime du Qubec, Cgep de
Rimouski. p. 1479.
[11] Haupais A. Combustion dans les moteurs Diesel. Techniques de lingnieur.
Trait mcanique et chaleur; 1992.
[12] Surugue J, Barrere M. Combustion. Librairie polytechnique Branger.
Dpartement Technique des presses de la cit; 1963. p. 111.
[13] Borghi R, Destriau M, De Soete G. La Combustion et les Flammes. Edition
Technip; 1995.
N. Larbi, J. Bessrour / Advances in Engineering Software 41 (2010) 898906 905
[14] Annexe VI MARPOL 73/78, Rgles Relatives la prevention de la pollution de
latmosphre par les navires et Code technique sur le contrle des missions
doxydes dazote provenant des moteurs diesel marins. OMI: Londres; 1998. p.
23.
[15] Remmels W, Velji A, Schmidt RM, Rauscher M. An experimental and
theoretical study of exhaust gas recirculation in diesel engines. CIMAC; 1995.
[16] Nadia H, abu Hamdeh. Effect of cooling the recirculated exhaust gases on
diesel engine emissions. Jordan: University of Science and Technology; 2003.
[17] Velji A, Remmels W, Schmidt RM, Water to reduce NO
x
emissions in diesel
engines a basic study. CIMAC; 1995.
[18] Sonoda K, Nakano K, Yamasita H, Nakayama N, Jinja Y. Research on
improvements in combustion of high water-content emulsied fuel. In:
Proceedings of the ISME, vol I. Yokohama; 1995. p. 4506.
[19] Yaguchi K, Yoshida T, Sato K, Kobayashi T, Ishii A. High NO
x
reduction
system mounted on 500 GT class vessel. In: Proceedings of the ISME,
Yokohama; 1995.
906 N. Larbi, J. Bessrour / Advances in Engineering Software 41 (2010) 898906