Renewable and Sustainable Energy Reviews 19 (2013) 517–530

Contents lists available at SciVerse ScienceDirect

Renewable and Sustainable Energy Reviews

journal homepage: www.elsevier.com/locate/rser

Numerical modeling of compression ignition engine: A review

Suneel Kumar a,n, Manish Kumar Chauhan b, Varun a
a
Mechanical Engineering Department, National Institute of Technology, Hamirpur 177001, India
b
Mechanical & Industrial Engineering Department, Indian Institute of Technology, Roorkee 247667, India

a r t i c l e i n f o abstract

Article history: Diesel engine modeling draws the greater attention due to its higher efficiency as compared to spark
Received 4 June 2011 ignition (SI) engine. Still it is very challenging job to model diesel engine due to its complex combustion
Received in revised form phenomena. The focus of the present study is to review the different available model used for modeling
12 November 2012
of CI engines. The modeling of CI engine is divided into single zone, multizone and multi-dimensional
Accepted 19 November 2012
model. Which further subdivided in many submodel i.e. heat transfer, ignition delay period, droplet
Available online 20 December 2012
evaporation, intake and exhaust flow, chemical kinematics and soot formation model. A comparative
Keywords: study has also been carried out with experimental validation to show the compatibility with different
Numerical simulation modeling approach. Some optimum mathematical input parameter has been suggested by the analysis
Multizone model
of different modeling approach to minimize the NOx emission and soot formation to make diesel engine
Heat transfer
more eco-friendly.
CI engines
NOx & 2012 Elsevier Ltd. All rights reserved.

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 517
2. Engine submodels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 518
2.1. Intake and exhaust manifold submodel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 518
2.2. Combustion submodel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 519
2.2.1. Diffusion and premixed combustion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 519
2.3. Fuel spray submodel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 520
2.3.1. Break up time calculation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 520
2.3.2. Droplet diameter after break up . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 520
2.3.3. Fuel spray dynamics. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 520
2.4. Ignition delay period . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 520
2.5. Heat transfer submodels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 521
2.6. Pollutant formation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 521
2.7. Soot formation submodel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 522
3. Results and discussions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 522
4. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 528
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 528

1. Introduction the crude oil and its high rising price makes a challenging job
for the engine manufacturers to manufacture or model less
One of the major pollution contributors in today’s environ- polluting and more efficient engine in order to meet its future
ment is the internal combustion (IC) engine which has been demands [1,2].
either spark or compression ignition (CI) engine. Due to scarcity of Engine simulation has been extensively used to improve the
engine performance. Compression ignition direct injection (CIDI)
diesel engines have been widely used in heavy-duty vehicle,
n
Corresponding author. Tel.: þ91 988 2721651. marine transportation and now have been increasingly being
E-mail address: [email protected] (S. Kumar). used in light duty vehicles, particularly in Europe and Japan [3].

1364-0321/$ - see front matter & 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.rser.2012.11.043
518 S. Kumar et al. / Renewable and Sustainable Energy Reviews 19 (2013) 517–530

Nomenclature R universal gas constant (J mol  1 K  1)

Re Reynolds number
A pre-exponential factor T temperature inside the cylinder (K)
An minimum area of nozzle (m) Tw temperature of the cylinder wall (K)
Ac cylinder area (m2) Ta activation temperature (K)
Ae effective flow area (m2) Tg temperature of gas (K)
a1 empirical constant up piston speed (m/s)
cd coefficient of discharge V volume of the cylinder (m3)
D cylinder diameter (m)
EA activation energy Greek
FS geometric circulation area of inlet (m2)
K adiabatic exponent of working fluid in a cylinder Dp pressure drop (bar)
ki rate constant (cm3/mol-s) df change in crank angle (degree)
Ks adiabatic exponent of working fluid before inlet valve s surface tension (N/m)
m mass flow rate (kg/s) ms inlet valve flow coefficient
ms mass of soot ra density of liquid (kg/m3)
mf g masses of vaporized fuel y spray angle (rad)
msf rates of soot formation g ratio of the specific heats
P cylinder pressure (bar) lg gas thermal conductivity
Pr preparation rate (kg/Crank angle) ma viscosity of the ambient gas
Pa pressure at the inlet (bar)
Pm motored pressure (bar) Subscripts
P1 upstream and stagnation pressures (bar)
P2 downstream stagnation pressures (bar)
inj injection of fuel
pm average mean pressure
soc start of combustion
Po2 partial pressure of oxygen
f fuel
PS pressure of working fluid before inlet valve
Pr pressure of working fluid after outlet valve

Experimental work which is aimed at fuel economy and low development of equivalence ratio zones surrounding a central
pollutants emission for IC engine requires change in input para- liquid core.
meter which is highly demanding in terms of money and time. In multi-dimensional model, the instantaneous conservation
So, in order to overcome this drawback, an alternative simulation time average equation of mass, momentum, energy has been
of engine performance with the help of mathematical model and taken into calculation. Stochastic as well as computational fluid
powerful digital computers lowers the cost and time. In these dynamics (CFD) model are introduced to see the effect of gas
simulation models, the effect of various design structures like phase turbulence on the liquid droplets. However, the effect of
design of combustion chamber input parameters (intake pressure, droplets on turbulence in many parts has also been ignored.
injection timing, etc.) and operation changes (compression ratio, Cylinder is considered to be divided into number of zones and
speed, etc.) can be estimated in fast and non-expensive way behavior of each individual zone being studied.
provided that main mechanism are recognized and modeled In this review paper different modeling approach and their
perfectly to meet the experimental results [4,5]. comparison along theoretical and experimental results was
In case of single zone model cylinder temperature, pressure reviewed. It also investigates the engine parameter and their
and mass can be obtained from ordinary differential equations. effect on the performance of the engine major pollutant such as
These models do not take into account the air entrainment, NOx and soot formation. Classification of various diesel engine
vaporization of fuel droplet and spatial variation of mixture models has been presented in Fig. 1.
composition and temperature. Single zone models are advanta-
geous also because of their simplicity and their wide usage along
with empirical data within the engine industry to make design 2. Engine submodels
decisions. The measured pressure rise in an engine is used to tune
the model and helps in calculating the rate of heat release from Submodels included in the study of the smallest region inside
the engine cylinder [6]. the engine cylinder which helps in proper understanding of the
For the accurate prediction and study the behavior of exhaust actual phenomena of the engine. It includes the different engine
emission pollutants, forced the researchers to develop two submodels which also help to study complex behavior of the
zone combustion models [7–10]. It provides more accurate result combustion.
as compared to single zone. Multizone model overcomes the
drawbacks of single zone model taking into account both the 2.1. Intake and exhaust manifold submodel
spatial as well as temporal variation of temperature and concen-
tration, where the detailed analysis of fuel–air distribution Intake and exhaust manifold submodel deals with the mass
which permits calculation of exhaust gas composition within flow rate inside the cylinder either the air entering during the
the reasonable accuracy [11–14]. Multizone models separate intake stroke, mass of fuel during the compression stroke and the
the fuel spray into a large, finite number of zones. Zones are amount of gases flow through the exhaust stroke. The timing and
small packets of fuel that move through the combustion rate of fuel injection into the chamber affects the spray dynamics
chamber. A second type of multizone model follows the and combustion characteristics. If the upstream pressure of the
 S. Kumar et al. / Renewable and Sustainable Energy Reviews 19 (2013) 517–530 519

Diesel engine 2.2. Combustion submodel

Modeling
Combustion model determines the rate of combustion. White-
house and Way [17] proposed a semi-empirical combustion
model for the calculation of rate of combustion. The fuel is
Thermodynamic Multidimensional
model model injected inside the cylinder in the liquid form with conical jets
and forming the burning zone. However, before the fuel has to be
burned it should be first heated, evaporated as well as mixed with
Single zone Multi zone
Spray sufficient amount of air entrained from the air zone through
model
model model diffusion. This prepared fuel is ready for the chemical process and
may be burned according to the rate governed by a chemical
kinetics equation [18]. For the calculation of preparation rate the
Hiroyasu’s Kono’s Thin spray Thick spray equation used is given as
Whitehouse way
model model model model
model m
Pr ¼ Km1x x
f inj mf up P o2 ð5Þ

Megurdichian This expression was based on the assumption that the rate of
and Watson’s preparation is proportional to the surface area of all droplets
model
having uniform diameter, while Spalding [19] suggested that the
rate of reaction is proportional to the droplets non-uniform
Fig. 1. Classification of diesel engine model. diameter rather than uniform diameter. In order to compensate
the non-uniform droplets diameter distribution exponent makes
allowances for that fact. The last term in Eq. (5) shows the effect
Table 1
Mass flow rate equation for flow through exhaust valve.
of oxygen availability on mixing to find out the reaction rate, for
that purpose Arrhenius expression is used.
Sl. no. Mass flow rate equation Critical condition Z f
KPO2 Ered =T
 qffiffiffiffiffiffiffiffi R¼ pffiffiffi e ðPRÞdf ð6Þ
1. dme 1
m pPffiffiffiffi 21=ðk1Þ 2k Pe
r 2k=ðk1Þ
. . .Supercritical flow N T
df
¼ 6Z e F e RT k þ 11=ðk1Þ kþ1 P k þ 1k=ðk1Þ
0
rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
 2=k ffi
2. ðk þ 1Þ=k  Pe 2k=ðk1Þ
dme Pr Pr Z . . .Subcritical flow The combustion is further divided into two parts as diffusion
df
¼ 61Z me F e pPffiffiffiffi
RT
2k
k1 2=k  ðk þ 1Þ=k
P k þ 1k=ðk1Þ
P P
and premixed combustion.

injector nozzle is known and assumed that flow through the

nozzle is incompressible, one dimensional and quasi-steady, then 2.2.1. Diffusion and premixed combustion
the mass flow rate of fuel injected inside the cylinder through the Diesel combustion is mainly considered as diffusion combustion
nozzle is given by [2] however it consist of premixed and diffusion phase, controlled by
qffiffiffiffiffiffiffiffiffiffiffiffi a fuel–air mixing process. Premixed phase of combustion occurs
m ¼ C d An 2rl DP ð1Þ before the start of diffusion combustion. It plays a significant role
in combustion process. Premixed combustion greatly influenced
In this type of mass flow rate, the intake and exhaust tanks by the engine speed, load and injection process, mainly the
(submodel) are treated as plenums with known temperatures and quantity of fuel injected in cylinder during the ignition delay.
pressures. Earlier these models do not provide any information The diffusion phase is governed by the turbulent mixing of air and
whether the flow is subcritical or supercritical at the inlet or fuel. The in-cylinder turbulence resulting from air motion and
outlet. They can be solved with the help of conservation of mass also fuel jet induced turbulence, which is predominant in modern
equation: in order to calculate the mass flow rate the equation is high pressure injection system is used in modern engine [20–22].
given by [15] In the recent years different burnt rate approaches are devel-
oped for premixed as well as diffusion combustion. Firstly, Vibe
rffiffiffiffiffiffiffiffiffiffiffi
j1 formula [23] was proposed for gasoline engines with totally
m ¼ Ae P1 ð2Þ
R1 T 1 premixed cylinder charge. This Vibe formula is also applicable
for diesel engine and widely used for the diesel engine combus-
" ! !#
tion modeling. This formula has only few drawbacks that it do not
2g P2 2=g P 2 ðg þ 1Þ=g
j1 ¼ 2=g
 ð3Þ taking into account the consideration of injection rate. Another
g1 P1 P 1 ðg þ 1Þ=g empirical approach was proposed by Schreiner [24] that includes
the entire heat release history for combustion from polygonal
These equations are further modified by taking into account
sections followed by a hyperbolic function for the burn-out phase.
the critical behavior of the fluid at intake manifold and exhaust
Austen and Lyn [25] described the individual burn rate as
manifold. The intake process is assumed as a subcritical flow and
mixing controlled process by dividing the injected fuel in a
the intake flow rate is given as [16]
number of packages. This work is further extended by Constien
v"ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
!#ffi [26] who used a multizone approach for droplet evaporation,
u
dms 1 u 2ks P2=ks Pðks þ 1Þ=ks
¼ m Fs t  ð4Þ formation and mixing and explained it in detail. The burn rate
df 6Z s ks 1 Ps 2=ks P s ðks þ 1Þ=ks formulation proposed by Hohlbaum [27] using the polygonal-
hyperbolic burn rate history was further extended by a spray
During the exhaust process, the flow at outlet valve is considered submodel based on the fact of gas jet theory. Weisser and
as a supercritical flow due to higher pressure difference between Boulouchos [28] firstly described the diffusion combustion as a
cylinder and exhaust port. When the pressure at the outlet valve single stage mixing controlled process. Barba [29] carried out this
decreases, then the behavior of flow changes to subcritical flow. formulation by adding a premixed combustion part and also
The calculation equation is given in Table 1. considered the pilot injection.
520 S. Kumar et al. / Renewable and Sustainable Energy Reviews 19 (2013) 517–530

2.3. Fuel spray submodel is divided into a number of zones only for multizone model.
Each zone control volume is treated as an open system where
2.3.1. Break up time calculation mass and energy equations are solved for individual zone.
For the calculation of break-up time, Hiroyasu equation [13] In order to avoid the computational efficiency in case of
has been most commonly used. In the quasi-dimensional com- multidimensional model, multizone model describe better the
bustion model of diesel engine modeling which is based on the spray penetration with the help of empirical correlations instead
theory of phase-divided spray mixing model, it is considered that of solving the full momentum equation. By the implementation of
the fuel droplets subjected to atomization period, gasification this method quasi-dimensional model provides the fastest and
period and eventually become gaseous after completion of break non-expensive means of generating the spatial information required
up time. The break up time is calculated according to Im and Huh to predict pollutant emission.
[15] is given as Since the multizone model based on empirical correlations to
rl do study the spray evolution, the fidelity of the spray penetration
t b ¼ 28:65 pffiffiffiffiffiffiffiffiffiffiffiffiffi ð7Þ model is extremely important for accuracy. For accurate predic-
ra Dp
tion of spray penetration Hiroyasu and Arai [36] proposed the
Reitz and Diwakar [30,31] included droplet break up for dense following correlations before and after break up. Hiroyasu and
spray model calculations. They used KIVA-2 software code for two Arai [36] correlation used is listed as follows:
breakup regimes: (i) bag break up (ii) stripping break up. They sffiffiffiffiffiffiffiffiffi
shows that the droplet break up has strong influence of the spray 2Dp
S ¼ 0:39 , 0 r t r tb ð13Þ
penetration, vaporization and mixing in high pressure sprays. The rc
size of spray droplets are controlled by coalescence and break up. sffiffiffiffiffiffiffi
They also include that droplet break up dominates in hollow-cone Dppffiffiffiffiffiffiffi
S ¼ 2:95 d0 t , tb r t ð14Þ
sprays because coalescences are minimized by expanding the ra
spray geometry. Reitz and Diwakar [31] concluded that injected
droplets having a diameter equal to the nozzle exit diameter. These correlations have been widely implemented in multi-
For the validation of computed length, droplet size, spray penetration zone models by different researchers [37–40]. These equations
and droplet velocity data compared with the available measured data are valid only if the fuel elements are on the spray centerline.
in high pressure spray. If the fuel elements in the radial direction away from the centerline
then the equation changes to
2.3.2. Droplet diameter after break up X r ¼ X exp ½4:403  103 ðL1Þ2  ð15Þ
In theory of combustion it is assumed that droplet after break
where L (¼2, 3, 4, 5,y), denotes the number of fuel elements
up having the same initial diameter which is equal to the Sauter
injected per injection interval.
Mean Diameter (SMD) having neglecting droplets size distribu-
tion and the details of the atomization process. The following
Eqs. (8)–(10) which are commonly used for the calculation of 2.4. Ignition delay period
SMD in multizone models are given by Hiroyasu et al. [32] to
compensate the variation of droplet size from one fuel parcel to Ignition delay period is defined as the time interval between
the next allowance is made depending on the operating condition the actual dynamic injection point and ignition. The ignition delay
[33]. period correlated to the cylinder pressure and temperature
! ! through the following semi-empirical equation given as [41].
LS
d32 ml0:54 rl0:18
¼ 4:12Rei0:12 wei 0:75 ð8Þ t d ¼ a1 pam3 ea2 =tm ð16Þ
dn ma0:54 ra0:18
! ! Ikegami et al. [42] and Tsao et al. [43] determined the
HS
d32 m0:37 r0:47 induction period of diesel sprays in a rapid compression machine
¼ 0:38Rei0:25 wei 0:32 l l
ð9Þ
dn m0:37
a r0:47
a
and reciprocating engine. Gerrish and Ayer [44], Spadaccini [45],
Walsh and Cheng [46] and Scharnweber and Hoppie [47] con-
!
d32 d d
LS HS cluded that ignition delay period decreases by the preheating of
¼ MAX 32 , 32 ð10Þ fuel before injection. Parker et al. [48] studied the effects of
dn dn dn
preheating of both the fuel and compressed air in a constant
Ingebo [34] developed a new theory for liquid jet breakups and volume bomb to supercritical temperatures of pure n-dodecane
shows that water jets in swirling and non-swirling flows, based which shows that ignition delay period can be expressed as
on the acceleration waves. A correlation was developed for the
td ¼ A  eðT a =T Þ ð17Þ
ratio between the orifice diameter and SMD, and it can be
expressed as Parker et al. [48] shows that preheating of the fuel before
injection has a significant role in shortening of ignition delay
dinj
¼ C ðW E RE Þ1=4 ð11Þ period by diminishing the vaporizing and mixing times and
SMD
concluded that induction period is insensitive to the combustion
Eq. (11) obtained is compared with the results of aerody- chamber pressure.
namics theory of liquid jet break up proposed by Reitz and Bracco Hardenberg and Hase [49] proposed an empirical formula for
[35] which predicts that SMD of liquid droplet is given by the direct injection diesel engines to predict accurately ignition
4ps  3 delay as a function of the fuel characteristics like cetane number,
SMD ¼ B  ð12Þ
rU 2inj  2 cylinder pressure, and temperature. Dent and Mehta [50] shows
that result of this formula carried out good agreement with
experimental data over a wide range of engine conditions.
2.3.3. Fuel spray dynamics However, the pressure and temperature used in their correlation
In case of multidimensional models entire combustion cham- are identified as the corresponding conditions, when the ignition
ber is taken as the computational domain and the fuel spray delay period starts at the end of TDC and can be estimated by
 S. Kumar et al. / Renewable and Sustainable Energy Reviews 19 (2013) 517–530 521

Table 2 Table 3
Heat transfer coefficient correlation given by different authors [2]. Calculation of rate constant [74].

S. No. Author name Heat transfer coefficient equations Reaction K i,r (cm3/mol s) Author

1. Eichelberg hce ¼ 7:8  1011  up0:333  p0:5  T 0:5 1 1:8  104  expð38,400=T Þ Baulch et al. (1991)
2. Hohenberg  0:8 GRI-MECH 3.0 (2000)
hch ¼ c1  V 0:8  p0:8  T 0:4  up0:333 þ c2 0:544  104  T 0:1  expð38,020=T Þ
3 Heywood [2]
c1 ¼ 13  10 , c2 ¼ 1:4 0:76  104  expð38,000=T Þ

2. 9
6:4  10  expð3150=T Þ Baulch et al.(1991)
3. Woschni hcw ¼ 3:26  103  D0:2  p0:8  T 0:55  up0:8
9:0  109  expð3280=T Þ GRI-MECH 3.0 (2000)
1:48  108  T 1:5  expð2860=T Þ Pattas (1973)

using a polytropic model for the compression stroke. Belardini 3. 6:4  1013 Baulch et al.(1991)
et al. [51] included that ignition delay time is a function of the 3:36  1013  expð195=T Þ GRI-MECH 3.0 (2000)
compression ratio, rpm and swirl in direct injection diesel 4:1  1013 Heywood [2]
engines. It has also been concluded that there is no such correla-
tion that fits experimental data. More fundamental equation
proposed by Hardenberg and Hase [49] to calculate the ignition literature. This equation considers net heat transfer as the summation
delay period is written as of both radiative and convective heat transfer rate. The net heat
n transfer is given by [58]
td ¼
6
! dQ lg  b    
    ¼a Re T W T g þ c T 4W T 4g ð21Þ
  1 1 21:2 0:63 dt d
 0:36 þ 0:22Sr exp EA  þ ð18Þ
RT 17190 P12:4 The total heat transfer in the two zones at each angle step is
distributed in proportion to their kilo moles and absolute tem-
Considering the effect of change of pressure and temperature
peratures [59]. The heat transfer coefficient correlations given by
in combustion zone, the following condition is specified [52]:
different authors have been given in Table 2.
Z ti
1
tig ¼ df, t ig Z 1 ð19Þ
0 t 2.6. Pollutant formation
Wu [53] studied the effect of engine speed on the ignition
delay period and concluded that as the engine speed is increased Basically nitrogen oxide and soot formation are two major
fuel burning rate in the premixed combustion phase decreased pollutants from diesel engine emissions. Nitrogen oxide and
which cause increases in the ignition delay period. Wu also added particulate matter emissions from diesel engines are increasing
that there is no correlation available to find the direct relation rapidly due to usage of high speed engines. Due to the advantage
between ignition delay and premixed burning rate. of direct injection diesel engine leads research to minimize the
emission aspect of high speed engines. The main challenging job
2.5. Heat transfer submodels for the designing of diesel engine is the reduction of NOx and soot
simultaneously. So, a lot of research has been carried out on
The heat transfer correlations of Annand [54] (including both design of combustion and fuel injection system to minimize the
the convective and radiations terms) Hohenberg, Woschni, Nussle pollutant emission. In order to control these emissions proper
and Eichelberg have been most commonly used for the calcula- combination is required between combustion chamber config-
tion of instantaneous average heat transfer coefficients. Keribar urations, fuel injection system nozzle diameter and optimum fuel
and Morel [55] proposed exceptional model for the heat transfer injection timing.
calculation which includes the heat transfer through cylinder wall For the objective of calculating the mass of combustion
as well taking into consideration the heat transfer through products of each zone, the complete chemical equilibrium system
cylinder head, piston crown, cylinder liner and valves. From the proposed by Vickland et al. [60] is used. It is considered that at
transient heat flux measurement in diesel engine it is concluded any instant of time for each zone 11 species are present. These
that due to spatial non-uniformities of the fluid flow and species are O2, N2, CO2, H2O, H, H2, N, NO, O, OH, CO. By using this
combustion cause different temperature histories at different scheme a set of 11 equations is formed, by which the concentra-
point in the cylinder head [56]. tion of each zone’s (11 species) has been obtained. By solving
Further study has been carried out for the heat transfer these equations, mass of fuel burned, volume temperature and
through engine to coolant and found that 20–35% of the fuel mass of air entrained in each zone has been obtained [61].
energy is transfered to coolant [57]. Primary heat transfer Chemical equilibrium based upon the assumption, so it cannot
mechanism takes place through convection from the cylinder correctly predict the concentration of NO and this drawback is
gases to the surrounding areas. Also heat transfer due to radiation overcome by chemical kinetics formation scheme proposed by
is taken into consideration because temperature inside the Lavoie et al. [62].
cylinder during combustion is very much higher. Heat transfer Besides that a lot of work has been carried out on the property
through convection is given by Newton law of cooling: of diesel fuel to see the effect on combustion and exhaust
emission of variable fuel property. To reduce the effect of acid
dQ  
¼ hc Ac T cyl T wall ð20Þ rain, sulfur content in the diesel fuel has been reduced. Other fuel
dt properties which are related to the improvement of engine
The convection heat transfer coefficient (hc) can be calculated performance and emissions include cetane number, density of
by using different empirical approaches such as Woschni, fuel, viscosity, distillation temperature, oxygen and aromatic
Annand, Hohenberg and Eichelberg. Annand’s formula has been content has also been studied by different researchers [63–68].
found to be more fundamental equation which can be used to Nakatani et al. [69] used the diesel particulate NOx reduction
calculate the rate of heat transfer. This equation seems to be more (DPNR) catalyst and found that reduction in the NOx and
fundamental than other alternative formulae available in the particulates matter simultaneously. It has been noticed that more
522 S. Kumar et al. / Renewable and Sustainable Energy Reviews 19 (2013) 517–530

than 80% of NOx and PM has been reduced. Hosoya et al. [70] occurred. Sinnamon and Cole [80] used sampling to study the
proposed model with selective catalytic convertor and diesel pollutant emission from a prechamber stratified charge engine.
particulate filter (DPF) to reduce the NOx and particulate matter They concluded that prechamber induces turbulence which helps
and reported that by using selective catalytic reduction (SCR) NOx to increases the fuel–air mixing rates and reduces the combustion
emissions reduced by 72%. The above used model with the interval. They also show that with the increment in the degree of
combination of DPF has been used and found a significant change stratification beyond a certain limit resulting in higher formation
in the reduction of particulate matter upto 93%. of NO and CO emission even if lean air–fuel ratio has been used.
NOx is strongly affected by higher temperature inside the Anna [81] proposed a thermodynamic model to predict the
combustion chamber during the premixed and diffusion combus- soot formation and oxidation rate of diesel engine. This model has
tion. By lowering the combustion temperature inside the cylinder been based upon the one step chemical reaction and D2 law and
leads to reduction of NOx emission but in other hand lowering also included droplet interference and forced convection effect. It
temperature causes reduction of thermal efficiency which is not has been assumed that during the vaporization and combustion
desirable. Supercharging strongly affect the combustion and processes liquid fuel droplets maintain their spherical symmetry
reduces the injection rate which results in reduction of NOx throughout.
emission. Uchida et al. [71] studied the effect of supercharging Hiroyasu et al. [13] proposed a model for the calculations of
and concluded that favorable result are obtained in the terms of net soot formation. According to this model, there is a soot
fuel consumption and exhaust emission. formation oxidation rate, which consider the available fuel mass
The extension of Zeldovich’s model with the inclusion of OH and oxygen of each zone (by including its partial pressure)
[72] scheme is generally accepted as a very successful method for respectively. So, net soot formation rate is calculated by subtract-
predicting NO concentration. The value of temperature, volume ing the soot oxidation rate from the formation one. By integrating
and equilibrium concentration for each zone is known and by each zone separately over time, the total value of soot formation
using these values the concentration of NO in each zone has been inside the combustion chamber is calculated by adding the
calculated by integrating the equation referring to the change of corresponding values for all the zones. Hiroyasu et al. [13]
NO concentration at any instant of time. By the addition of NO proposed soot formation model is implemented in multizone
concentration for all zones at a certain time, the total [NO] models to calculate the soot formation. The soot formation rate
concentration inside the combustion chamber [73] has been is calculated by assuming that vaporize fuel having a first-order
obtained. Chemical reactions are given below: reaction which is given by the equation:
k1 dms
Oþ N2 2NO þ N ð22Þ ¼ m:sf m:sc ð25Þ
dt
k2
N þ O2 2NO þ N ð23Þ m:sf ¼ Af mf g P 0:5 eðEsf =RT Þ ð26Þ

k3
N þ OH2NOþ H ð24Þ Po2 1:8 ðEsc =RT Þ
m:sc ¼ Ac ms P e ð27Þ
P
Rate of reaction k1, k2, k3 are calculated with the help of
Lipkea and DeJoode [82] modified the previous model and
Table 3.
formed a new set of equation to calculate the net soot formation.
According to this model soot formation and soot oxidation rate is
2.7. Soot formation submodel
calculated as follows:

Soot is basically produced when nondiluted spray produces dmsf

¼ Asf dmf0:8 p0:5 eðEsf =Rmol T Þ ð28Þ
fuel rich vapor regions and undergoes chemical reaction. It has dt
also been formed on liquid film along the cylinder walls. Flower  
dmsc pO2 n ðEsc =Rmol T Þ
[75] shows that with the increase in pressure volumetric rate of ¼ Asc dmsn p e ð29Þ
dt p
soot formation in laminar diffusion flames increases. Rah et al.
[76] proposed a correlation of NO and soot formation with the Therefore, the net soot formation is expressed as follows:
ignition of liquid fuel droplet. It also included that soot formation dms dmsf dmsc
can be suppressed by delaying the ignition, vaporizing as much as ¼  ð30Þ
dt dt dt
fuel possible prior to the ignition. Rah et al. [76] also proposed a
The net soot formation rate is proposed by Khan–Hiroyasu–
possible method of controlling the NOx and soot emission by
Belardini expression and soot oxidation rate given by Nagle–
using of low oxygen concentration in oxidation to delay ignition
Strickland Constable expression is given as [73]:
and high fuel–air ratio to reduce NOx.
Uyehara [77] shows that in diesel engine maximum amount of dms dmsf dmso
¼  ð31Þ
the soot occurs in the temperature range of 2000–4000 K. It also dt dt dt
shows that less than this temperature, soot formation rate is
negligible. Kagami et al. [78] studies the effect of fuel Cetane dmsf
¼ Af eðT f =T Þ mf v ð32Þ
number, viscosity and distillation range on the performance of dt
gaseous and smoke emission, knocking, fuel consumption, start- dmso 6M c
ability and cold performance of diesel engine. The result shows ¼ mR ð33Þ
dt rs Ds s so
that the fuel with low Cetane number produces higher emission
and also shows that the fuel consumption is high as compared to Single zone, two zone and multizone models are proposed by
fuel which is having very high Cetane number. the different author as listed in Table 4.
Tateishi et al. [79] proposed a new combustion model for
indirect diesel injection (IDI) diesel engines and show that this 3. Results and discussions
model reduces the fuel consumption, emission and noise. Engine
noise is due to rise in very high peak pressure at the early stage In this literature review different modeling approach is used to
of combustion, when there rapid heat release rate has been understand the behavior of diesel engine modeling and on the
Table 4
Comparative study of different models of CI engine.

Author name/country and year Model used Engine spe. (B/S/CR) Parameter studied HTC/exp Submodel used Results
rpm val.

Ziarati/UK (1990) [83] Single zone model (215/241/11/) na Delay Period, droplet Annand/ Combustion, heat transfer Model is capable to run under heavy fuel and diesel fuel
penetration yes and spray model Shows good agreement of cylinder pressure and
temperature with experimental result.

Ishida/Japan (1994) [84] Two zone model (100/105/ na ) 3500 NOx emission, pressure, na/yes Heat release model, Ignition NOx emission is reduced by retarding the time by decreasing
temperature and ignition delay model, NOx emission the maximum combustion temperature and pressure in
delay model expansion stroke.
NOx reduces by using small nozzle diameter is caused by
decreasing the combustion pressure.

S. Kumar et al. / Renewable and Sustainable Energy Reviews 19 (2013) 517–530

Rakopolous/Greece (1994) [85] Multi zone model (86/80/18.5/) Insulation effect on K-EModel/ Fuel spray, evaporation, ignition No remarkable change in engine efficiency has been seen
1200,1500,3100 performance and no delay, heat transfer, chemical but increases in the exhaust enthalpy helps to improve
exhaust emission on combustion model overall engine efficiency by using power turbine
diesel engine Increment in the NO and soot formation occurs by the
insulation effect that is not desirable.

Belardini/Italy (1996) [86] 3-D model (199/95/17/)na NOx and soot formation, na/yes NOx, soot formation, ignition The model over predicts both acetylene and soot volume
heat delay, fraction. However, it is noted that better results can be
obtained with a careful optimization of the model constants.

Yong/Pohang (2000) [15] Multi-zone model (123/155/17.1/) IMEP, pressure traces, Woschni/ Intake and exhaust manifold, In cylinder flow model is combined with Hiroyasu model
1200,1400,1600,1800 NOx and soot emissions, yes NOX, Ignition delay, Air both model shows results consistent with the general
fuel injection entrainment, In cylinder flow operating characteristics of a DI diesel engine.
timing,swirl ratio effect and Fuel evaporation The calculated imep, pressure traces, brake specific NOx and
model soot emissions studied by the present model shows good
agreement with the test bed measurements at full and part
load conditions for a range of the engine speeds.

Jung/Michigan (2001) [87] Quasi – (130/160/15)2100 Pressure, temperature, Assian– Ignition delay model, heat Extended Zeldovich mechanism accurately predicted the NO
dimensional ignition delay, Heywood/ transfer model, air entrainment and soot formation. At least five zones are recommended to
multi-zone model combustion, NO and soot yes model, spray evaporation model determine the NO and soot in radial direction.
formation Modified Hiroyasu correlation for spray tip penetration
show good agreement with experimental data in a range of
injection pressure.

Papagiannakis/Greece (2004) [88] Single zone model (85.73/82.55/17.6) Pressure, heat release, na/no HCCI model, heat transfer Exhaust emission and performance of engine study have
1000–3000 brake specific fuel model, pollutant formation been carried out at different speeds and loads for diesel and
consumption, NOx, CO, model dual fuel (combination of diesel and gaseous fuel).
HC and soot emission Dual fuel combustion process has lower peak cylinder
temperature in comparison to conventional diesel.
NOx and soot emission is low and HC and CO emission is
high in dual fuel operation compared to normal diesel
operation.

Rakopolous/Greece (2004)[58] Two zone model (82/88.90/na)4500 The growth of the fuel Annand/ Fuel Preparation,combustion Two zone models proves to very efficient for the
spray zone, jet mixing, yes Model, NOx, Soot Formation determination of engine performance and exhaust emission
C–H–O system model by concerning the effect of two major parameter of load and
injection timing.
Theoretical result show good agreement with exp. value.

Zheng/China (2005)[89] Multi-dimensional (95/115/20) Pressure, temperature, na/yes Injection time, natural gas composition and initial

523
model 1200,1600 ignition timing, natural temperature have a significant effect on ignition,
 524
Table 4 (continued )

Author name/country and year Model used Engine spe. (B/S/CR) Parameter studied HTC/exp Submodel used Results
rpm val.

gas composition, NO Combustion model, turbulence combustion and emission.

emission model, mixture formation Increased amount of ethane in natural gas provide advance
model, No formation model in ignition and increment in NO emission.
Auto-ignition of natural gas occurs at 1300 K of mixture
temperature and NO emission reduce when fuel injection
timing delay.

Tauzia/France (2006)[90] Phenomenological (123.8/165/16.5) Premixed combustion, na/yes/cfd- Premixed combustion model, Model studied at injection timing -161 to þ121ATDC shows
Model 1900–3000 pressure, temperature kiva 3v Ignition delay model good agreement with experimental and CFD analysis from a
quantitative and qualitative point of view for ignition delay

S. Kumar et al. / Renewable and Sustainable Energy Reviews 19 (2013) 517–530

and in-cylinder pressure evolution during combustion.

Xingcai/Chima (2006) [91] Multi-zone model (98/105/18.5)1800 IMEP, thermal efficiency, na/yes HCCI combustion model, Energy input, ignition timing, maximum gas pressure is
pressure, heat release pollutant model deteriorated with increase of ethanol addition.
rate, ignition timing, For all fuels, CO emission is very high at 1.5-2.5 bar of
emission characteristics engine load.
When IMEP is larger than 3 bar ignition timing lies in -51CA
ATDC-TDC.

Chmela/Austria (2007) [92] Zero Dimensional na Premixed combustion na/no Ignition delay,Premixed and An improved model for diffusion combustion is presented
ROHR Model based on Arrhenius and diffusion combustion,Reaction which shows a sequential approach.
Mangussen equation. Rate combustion Model Ignition delay and premixed combustion model solved by
Arrhenius and Magnussen approach shows better
description of the effects of considerably transient ROI
shapes by using a more detailed consideration of the
processes in the fuel jet.

Sundarapandian/India (2007) [93] Multizone model (87.5/110/17)1500 Heat release, work done, Annand/ Combustion and heat release, Heat release is reduced by 4–8% for vegetable oil esters
and harmful pollutant yes heat transfer, engine compared to diesel fuel. Pressure of vegetable oils esters are
such as HC, CO, NOX and performance like specific fuel reduced about 4%, 5% and 7% respectively for Jatropha,
smoke. consumption, brake thermal mahua and neem oil ester compared to diesel.
efficiency But the negative impact has been seen on thermal efficiency
reduced by 3%, 4% and 5% for Jatropha, mahua and neem oil
ester compared to diesel.
CO reduced by 19% for Jatropha and 16% for the mahua and
neem oil ester compared to diesel.

Machrafi/France (2008) [94] Single zone model (82.55/114.5/4-14)na Heat release ratio, na/yes Combustion model ignition Heat release increase with increase in equivalence ratio
equivalence ratio, delay model, heat released which is affected by ignition delay.
ignition delay, model Compression ratio provides advancing in ignition delay and
compression ratio lower residual gas fraction.
Variation of inlet temperature is directly affected the
ignition delay but heat release distribution did not affected
the ignition delays.

Sahin/Turkey (2008) [95] Multi-zone model na Details of fuel spray, heat Annand/ Heat transfer, combustion Indicated thermal efficiency at 2500 rpm by present model
transfer, swirl yes model, spray model, NO is 0.4420 and Li model is 0.43 shows that in result only 3% of
model,engine emission model, error occurs.
performance Equivalence ratio changes between 1.086 to 1.1.49 while
actual ration like 0.650-0.80 so model is not adequate at low
equivalence ratio.
Semin/Malaysia (2008) [96] Single zone model (86/70/na)200–4000 MEP, mean piston speed, na/no Parametric model –MEP, SFC, From engine modeling it is concluded that highest brake
specific power output, power output, mean piston power is 4.314 kW and indicator power is 5.725 kW occurs
sfc intake valve mach speed. at 3000 rpm.
index. The engine modeling brake torque maximum is 14.3662
N m and indicated torque is 18.743 Nm occurs at 2800 rpm.
At 2800 rpm, the engine model has shown the lowest value
of the BSFC and the indicated specific fuel consumption.

Garcia/Spain (2009) [97] Single zone model (95/100/19) Pressure, fuel specific na/no Combustion model, pollutant HCCI combustion process provides low NOX and soot
1500–2400 consumption, start of model, HCCI combustion model emission than the diesel mode. HCCI with exhaust gas
combustion, NOX, CO, HC recirculation (EGR) has a great impact on combustion and
and soot emission emission.
When angular speed increases, start of combustion that
affected by inlet temperature at constant EGR rate, decrease.

S. Kumar et al. / Renewable and Sustainable Energy Reviews 19 (2013) 517–530

Huang/China (2009)[98] Multi-dimensional (115/115/17)1400 Pressure profiles, na/yes Turbulence model, combustion HCCI combustion process does not take place
CFD model, single combustion process, model, mixture formation simultaneously in combustion chamber core zone. High
zone model HC and CO emission model, HCCI combustion temperature reaction occurs in vicinity of piston surface.
Unburned fuel and CO increase when Di-methyl ether
(DME) equivalence ratio decreases.
Pressure profile of multi-dimensional CFD model is more
accurate than the single zone model because turbulence
flow and wall heat transfer is measured.

Kannan/India (2009) [99] Single zone mode (87.5/110/17.5)1500 Heat release rate, Hohenberg/ Ignition delay model, heat From simulation result it is concluded that 18% and 21.5% of
pressure, brake thermal yes transfer model, pollutant reduction in NO is achieved with 10% and 20% dilution of
efficiency, bsfc NO emission model. diesel with water emulsion respectively.
formation The brake thermal efficiency increases as reported by Abu –
Zaid by 3.5% by increasing the water emulsion by 20% over
the use of diesel for the engine speed range studied.

Komninos/Greece (2009)[100] Multizone model (120.65/140/17)na Mass transfer effect on Annand/ Heat transfer as well as Neglecting mass transfer prior to main heat release would
emission i.e. unburned yes pollutant formation model result in a severe underestimation of the HC accumulated in
HC and CO emission, the crevices during combustion (60% for the case Studied).
Neglecting mass transfer during combustion and expansion
would result to and underestimation of the final HC amount
by at least 14% for the case studied.

Kuleshov/Moscow (2009) [101] Multizone Model (88/85/17.6)1800 Heat release inside the Woschni/ combustion model, fuel spray The RK-model accounts for drop size interaction of free
(RK-model) cylinder and NOx yes modelgas exchange model, sprays with swirl effect, spray and wall impingement,
formation, inlet manifold ignition delay model evolution of near-wall flow formed by spray, hit of fuel on
pressure and EGR rate cylinder head surface and swirl intensity on heat
release rate.
This model helps in designing of piston bowl shape,
numbers of injector holes, diameters and directions in
match with swirl intensity to control emission and fuel
consumption.

Liu/China (2009) [102] Two Zone Model (102/120/18:1) Mass distribution, heat Woschni/ Heat transfer model, mass The compression ratio of 13 is great enough for the auto-
800–2000 transfer,mass exchange na exchange model, porous ignition in the porous medium engine, which is not so
between zones aheat medium model, chemical crucial like in the HCCI engine.
transfer in porous kinematics For the same intake pressure of 1 atm, the pressure in the
medium and studied the PM engine is relatively lower than that of a traditional
effect of PM on engine due to the voids in PM, which lowers the ignition
performance delay and higher engine output.
Due to higher average temperature in PM engine NO
emission is higher at excess air ratio of 1.6.

525
 526
Table 4 (continued )

Author name/country and year Model used Engine spe. (B/S/CR) Parameter studied HTC/exp Submodel used Results
rpm val.

Dhuchakallaya/UK (2010) [103] Spray model na Combustion, spray na/na Combustion model, Ignition Higher lift of length cause the increases of air entrainment in
dynamics and ignition model the premixed combustion zone resulting in high reaction rate.
delay Simulation results show good agreement with experimental
results at time of 4 ms of start of injection.

Gogoi/India (2010) [104] Single zone model (87.5/110/14– Engine performance- Annand/no Heat transfer, ignition delay Maximum brake power occurs at design speed and further
20)600–2000 IMPEP, BTE comparative model, gas exchange model increases of speed decrease brake power. For CR of 17.5 the
study of diesel with maximum peak power occurred at 1500 rpm for B20 and B40
biodiesel blends blends while for B60 it occurs at 1400 rpm.
Increased brake power was observed for B40 and B60 blends
as compared to diesel fuel.

S. Kumar et al. / Renewable and Sustainable Energy Reviews 19 (2013) 517–530

Blends B20, B40 and B60 shows an increase in thermal
efficiency due to increases in break power as compared to
diesel engine.

Prasad/India (2010) [105] Two Zone Model (11/127/16)1000– Pressure, heat release, Annand/ Preparation rate and reaction The cylinder pressure for low heat rejection biodiesel is
1500 heat transfer and yes rate model for the calculation of lowers than LHR diesel about 1% and higher about 4% and
performance heat transfer rate. 2.5% for TC biodiesel and TC diesel.
characteristics –SFC, BTE Brake thermal efficiency of LHR biodiesel engine is lower
than that of LHR diesel and higher than TC biodiesel and TC
diesel operation this is because of lower calorific value of
biodiesel as compared to diesel fuel.

Qi/China (2010) [106] Quasi-dimensional (135/150/17) 1500 Diesel engine Newton/ Spray mixing, combustion Simulation result shows good accordance with experimental
two zone model performance- sfc,effect yes model, ignition delay period, result proves the validation of phase divided spray
of air entrainment and heat transfer, pollutant mixing model.
spray mixing model formation, intake and exhaust The relative error between simulation and experimental
flow model result for the specific fuel consumption and NO emission
occur to be 2.8% and 9.1%.
By using this model computational time is 36 s to analyze
engine working process.

Rakopolous/Greece (2010) [107] Quasi-dimensional (85.73/82.55/17.61) Cylinder flow field and na/CFD heat transfer, fuel injection rate, Both quasi-dimensional and CFD model are compared for
500–2500 gas temperature model spray penetration, evaporation, the different piston bowl geometry changing from (d/D)
distribution. combustion and pollutant from 64% to 44% at different rotational speed.
formation It is concluded that both models predict the quite similar
cylinder pressure and temperature, traces whole closed
cycle of the engine.

Shankra/India (2010) [108] Multidimensional (130/150/15.5:1) CFD model proposed to Eckert/yes CFD based—combustion and It is observed that by advancing injection time cause results in
model (CFD 2000 study the effect of intake ignitions model, turbulence the increases pressure, temperature heat release rate,
model) pressure, fuel injection model, NOx and soot formation cumulative heat release and NOx emissions by 6.88%. If we
on the performance of DI and droplet breakup model, reverse the trend then NOx emissions by 6.85%.
engine Higher NOx (15.03% and 58.69% at 1.21 bar and 1.71 bar
respectively) and lower soot emission (8.82% and 51.47% at
1.21 bar and 1.71 bar) concluded that optimum injection time
and intake pressure is 121 bTDC and 1.21 bar.

Shi/USA (2010) [109] Multidimensional (137.16/165.1/16.1) Pressure, injection na/yes Adaptive multi-grid chemistry At high load, high volatility fuel as gasoline and E10 have
model 1300 timing, specific fuel model, combustion model less sensitivity to fuel reactivity. So these are better for fuel
consumption, soot economy.
formation, NOx, CO, UHC For cleaner combustion, gasoline fuels are more efficient
emission than normal diesel fuel.
 Normal diesel is affected by first injection amount but
gasoline fuel is affected by second injection timing.

Som/USA (2010)
[110] Single zone model (50/200/na)
na na/yes Spray and combustion KH breakup model, spray and -KH-ACT model have better performance of parameters
characteristic, fuel combustion model, such as injection pressure, ambient density and fuel
penetration, aerodynamically induced temperature with vapour penetration for evaporating
temperature contours, breakup model, soot formation spray. -The inclusion of turbulence and cavitations
flame index enhances the primary breakup process, decreases in liquid
penetration and increases in radial dispersion. -Enhancing
spray breakup provides ignition length and flame
stabilization near nozzle exit.
Reiter/USA (2011) [111] Multi-zone model (106/127/17) 1000 Pressure, heat release na/ no Ignition delay model, In dual fuel engine, ignition delay is longer due to increased
rate, engine break combustion and heat release ammonia and resulting peak combustion pressure is low.
specific fuel model, pollutant formation BSFC of diesel is high below 40% of diesel fuel energy.
consumption, soot model, soot formation model NO emission is low by using less than 40% ammonia in dual

S. Kumar et al. / Renewable and Sustainable Energy Reviews 19 (2013) 517–530

formation fuel compared 100% of diesel.

Boretti/USA (2012)[112] Multi-zone model (131/158/18)1900 Cumulative heat release, na/no Combustion model with jet Diesel injection during compression and expansion to build
pressure, heat release, ignition and diesel ignition, up the pressure are best options to develop a pressure
pollutant emission stochastic reactor model (SRM), profile.
soot model Thermal efficiency of diesel injection and hydrogen jet
ignition combustion models is 10% larger than the diesel at
high loads.

Komninos/Greece (2012)
[113] Single zone, multi- na Chemical kinematics na HCCI combustion model, direct Present the effect of various configurations in single zone
zone and modeling, different injection HCCI, fully premixed model such as negative valve overlap, heavy EGR, fuel
multidimensional reaction mechanisms for HCCI, emission model reforming on biofuel HCCI combustion.
model biofuel combustion Study the multi-zone models for mixture thermal and fuel
simulation stratification which provide better and more realistic
estimates on combustion duration, peak combustion
pressure and emissions formation.
Multi-dimensional models provide higher spatial resolution
of the combustion chamber than the multi-zone model.

Visakhamoorthy/Canada (2012) [114] Multi-Zone model (72/73.6/22) 1500 Pressure, heat release Yes/ yes HCCI engine model, mixture The model represents good quality HRR and pressure
rate, Emissions, composition model, pollutant prediction curves after calibration.
predicted NOx levels model When engine operating point is close to misfire limit, model
is unable to handle cyclic variability i.e. to predict the
pressure.
Predicted NOx emission increase with small increase in
cylinder temperature.

HTC—heat transfer correlation, B—Bore of cylinder, Spe.—Specification, Exp.—Experimental, S—Stoke length, Val.—Validation, CR—Compression ratio.

527
528 S. Kumar et al. / Renewable and Sustainable Energy Reviews 19 (2013) 517–530

basis of these modeling approach, following result has been [3] /http://www.dieselnet.com/news/2008/02acea.phpS; 2009. p. 1–20.
obtained and discussed below: [4] Benson RS, Whitehouse ND. Internal combustion engine. Oxford: Pergamon
Press; 1979.
[5] Horlock JH, Winterbone DE. The thermodynamics and gas dynamics of
1. Insulation effect on the cylinder wall is studied by the multi- internal combustion engines. Oxford: Clarendon Press; 1986.
zone model and it is concluded that by the increasing the [6] Ramos JI. Internal combustion engine modeling. Hemisphere Publishing
Corporation, New York; 1989.
degree of insulation causes increment in the NOx emission and [7] Khan IM, Greeves G, Probert DM. Prediction of soot and nitric oxide
simultaneously increases soot formation. This increase in concentrations in Diesel engine exhaust. Air pollution control in transport
emission and soot formation is not admirable because of engines, Institution of Mechanical Engineers, Paper C142/71; 1971. p. 205–
17.
negative impact on the environment. However, the engine
[8] Whitehouse ND, Sareen BK. Prediction of heat release in a quiescent
efficiency is improved by providing more insulation on the chamber diesel engine allowing for fuel/air mixing. SAE 1974; Paper no.
cylinder wall. So, it is suggested that degree of insulation 740084.
should be properly selected. [9] Rakopoulos CD. Influence of ambient temperature and humidity on the
performance and emissions of nitric oxide and smoke of high speed diesel
2. Rate of heat release (ROHR) model proves very effective to engines in the Athens/Greece region. Energy Conversion and Management
accurately predict the different instantaneous parameter 1991;31(5):447–53.
which controls the diffusion combustion and also it takes very [10] Kouremenos DA, Rakopoulos CD, Hountalas DT. Computer simulation with
experimental validation of the exhaust nitric oxide and soot emissions in
less computational time.
divided chamber Diesel engines. In: Proceedings of the ASME-WA meeting,
3. Single zone approach by using the water diesel combustion 10(1). San Francisco, CA; AES 1989. p. 15–28.
emulsion has been successfully modeled to predict the engine [11] Shahed SM, Chiu WS, Lyn WT. A mathematical model of diesel combustion.
performance parameters. It has been found out that with the Combustion in engines, Institution of Mechanical Engineers, Paper C94/75;
1975. p. 119–28.
20% increases in the water emulsion brake thermal efficiency [12] Hodgetts D, Shroff HD. More on the formation of nitric oxide in a Diesel
(BTE) of the engine increases by 3.5% which is significant engine. Combustion in engines, Institution of Mechanical Engineers; 1975;
improvement for diesel engine. Similarly the NOx emission Paper C95/75. p. 129–38.
[13] Hiroyasu H, Kadota T, Arai M. Development and use of a spray combustion
reduced from 21% to 18.5% by increasing the water emulsion modeling to predict diesel engine efficiency and pollutant emissions.
from 10% to 20%. Bulletin of the Japan Society of Mechanical Engineers 1983;26(214):
4. CFD study is carried out for the detailed analysis on the 569–75.
[14] Kouremenos DA, Rakopolous CD, Karvounis E. Thermodynamic analysis of
multizone model to study the effect of intake pressure, injec-
direct injection diesel engines by multi-zone modelling. In: Proceedings of
tion timing on the NOx and soot formation. From the analysis it the ASME-WA meeting, vol. 3(3). Boston, MA: AES; 1987. p. 67–77.
is concluded that with the increase in the intake pressure NOx [15] Im YH, Huh KY. Phenomenological combustion modeling of a direct
emission increases while soot formation rate decreases. So, injection diesel engine with in cylinder flow effects. KSME International
Journal 2000;14(5):569–81.
from the three intakes pressure range 1.01, 1.21 and 1.71 bar [16] Zhu FJ, Wu J. Numerical simulation and its optimization of internal
the 1.21 bar gives more desirable optimum result which combustion engine working process. Beijing: National Defense Industry
compensate both type of emissions. Similarly out of 16, 12 Press; 1997.
[17] Whitehouse ND, Way RJB. Rate of heat release in diesel engines and its
and 8 bTDC injection timing the 12 CAD found to be providing
correlation with fuel injection data. Proceeding of the Institution of
more optimum result of pollutant emission. Mechanical Engineers 1969–1970;184:17–27.
5. Ignition delay model is studied at different cetane numbers [18] Whitehouse ND,Way RJB. A simple method for the calculation of heat
and variable speed by using Hardenberg equation and con- release rates in diesel engines based on the fuel injection rate. SAE 1971;
Paper no. 710134.
cluded that with the increases in the engine speed the ignition [19] Spalding DB. Some fundamentals of combustion. London: Butterworths; 1955.
delay period increase while with the increase of the cetane [20] Chmela FG, Orthaber GC. Rate of heat release prediction for direct injection
number. diesel engines based on purely mixing controlled combustion. SAE 1999;
Paper no. 1999-01-0186, p. 1–9.
[21] Chmela FG, Engelmayer M, Pirker G, Wimmer A. Prediction of turbulence
controlled combustion in diesel engines. In: Thiesel 2004 conference on
thermo and fluid dynamic processes in diesel engines; 2004.
[22] Barba C, Burkhardt C, Boulouchos K, Barged M. A phenomenological
4. Conclusion
combustion model for heat release prediction in high-speed DI diesel
engines with common-rail injection. SAE 2000; Paper 2000-01-2933.
In this present article single zone, multizone, and multi- [23] Vibe II. Combustion process and cycle of internal combustion engines.
dimensional models of diesel engine combustion has been Berlin: VEB Verlag Technik; 1970.
[24] Schreiner K. Studies on the heat release rate and heat transfer in high-speed
reviewed. Single zone model has been found to more effective heavy-duty diesel engines. MTZ 1993;54(11):554–63.
tool for immediate prediction of engine parameter but case by [25] Austen AE, Lyn W. Relation between fuel injection and heat release in a
case adjustment of the wiebe function parameter is required. direct injection engine and the nature of the combustion process. Proceed-
ing of the Institute of Mechanical Engineers; 1960.
Theses wiebe function parameter deals with engine geometry [26] Constien M. Determination of injection and combustion process of a direct
and operating condition. Multizone model account for air injection diesel engine. Thesis; 1991.
entrainment and inhomgeneities of the mixture and overcome [27] Hohlbaum B. Contribution to the theoretical study of nitric oxide formation
of high-speed heavy-duty diesel engines. Thesis. University, TH Karlsruhe;
the drawback of single zone model. Multi-dimensional models
1992.
account for both engine geometry temporal as well as spatial [28] Weisser G, Boulouchos Noemi K. A tool for a preliminary damage estimation
variation of the flow field. These accurately predict the spray of nitrogen oxide emission direct-injection diesel engines. In: Graz:
behavior and fuel spray dynamics model. These model 5th conference, the work process of the engine; 1995.
[29] Barba C. Development of combustion characteristics of combustion data for
included the CFD analysis which accurately predict the com-
the improved prognosis of the computer simulation of the combustion
bustion behavior and help to reduce the pollutant emission at process for passenger diesel engines. Thesis. ETH rich to economics; 2001.
great extent make diesel engine more reliable and fewer [30] Reitz RD, Diwakar R. Effect of drop breakup on fuel sprays. In: SAE
impact on the environment. international congress and exposition. SAE 1986; paper no. 860469.
[31] Reitz RD, Diwakar R. Structure of high pressure fuel sprays. SAE interna-
tional congress and exposition; 1987. Paper no. 870598.
References [32] Hiroyasu H, Arai M, Tabata M. Empirical equations for the Sauter mean
diameter of a diesel spray. SAE 1989. Paper no. 890464.
[33] Lafrance P. The breakup length of turbulent liquid jets. ASME Journal of
[1] Ferguson CR. Internal combustion engines. New York: Wiley; 1986. Fluids Engineering 1977;99:414–5.
[2] Heywood JB. Internal combustion engine fundamentals. New York: [34] Ingebo RD. Acceleration wave breakup of liquid jets with airstreams, in fluid
McGraw-Hill; 1988. mechanics of combustion systems. New York: ASME; 1981 pp. 101–5.
 S. Kumar et al. / Renewable and Sustainable Energy Reviews 19 (2013) 517–530 529

[35] Reitz RD, Bracco FV. Mechanism of atomization of liquid jet. Physics of [67] Lange WW. The effect of fuel aromatic structure and content on particulate
Fluids 1982;25(10):1730–43. emissions in heavy-duty truck engines under transient operating condi-
[36] Hiroyasu H, Arai M. Fuel spray penetration and spray angle in diesel tions. SAE1991; paper no. 912425.
engines. Transactions of JSAE 1980;21:5–11. [68] Fukuda M, Tree DR, Foster DE, Suhre BR. The effect of fuel aromatic
[37] Hiroyasu H, Kadota T, Arai M. Development and use of a spray combustion structure and content on direct injection diesel engine particulates. SAE
modeling to predict diesel engine efficiency and pollutant emissions (Part 1992; paper no. 920110.
1 Combustion modeling). Bulletin of the JSME 1983;26(214):569–75. [69] Nakatani K, Hirota S, Takeshima S, Itoh K, Tanaka T. Simultaneous
[38] Kyriakides SC, Dent, JC, Mehta PS. Phenomenological diesel combustion PM and NOX reduction system for diesel engines. SAE 2002; paper no.
model including smoke and NO emission. SAE1986; Paper no. 860330. 2002-01-0957.
[39] Bazari ZA. DI diesel combustion and emission predictive capability for use [70] Hosoya M, Kawada Y, Sato S, Shimoda M. The study of NOX and PM
in cycle simulation. SAE1992; Paper no. 920462. reduction using urea selective catalytic reduction system for heavy duty
[40] Gao Z, Schreiber W. A Multizone analysis of soot and NOx emission in a DI diesel engine. SAE 2007; paper no. 2007-01-1576.
diesel engine as a function of engine load, wall temperature and intake air [71] Uchida N, Daisho Y, Saito T, Sugano, H, Combined effect of EGR and
O2 content. ASME 2000; Paper no. 2000-ICE-314. supercharging on diesel combustion and emission. SAE 1993; paper no
[41] Suppes G, Srinivasan B, Natarajan V. Auto ignition of biodiesel, methanol, 930601.
and a 50:50 blend in a simulated diesel engine environment. SAE 1995; [72] Kouremenous DA, Rakopolous CD, Hountalas DT. Computer simulation with
Paper no. 952758:10.4271/952758. experimental validation of the exhaust nitric oxide and soot emission in
[42] Ikegami M, Miwi K, Inada M. A study of ignition and combustion of diesel divided chamber diesel engines. AES 1989;10(1):15–28.
spray by means of rapid compression machine. Bulletin of the JSME [73] Jiang DM. Combustion and emission of internal combustion engine. Xi’an:
1981;24(195):1608–15. Xi’an JiaoTong University Press; 2001.
[43] Tsao KC, Myers PS, Uyehara OA. Gas temperature during compression in [74] Günter PM, Christian S, Gunnar S, Frank O. Simulation of combustion and
motored and fired diesel engine. SAE Transactions 1962;70:136–45. pollutant formation for engine-development. Berlin, Heidelberg: Springer-
[44] Gerrish HC, Ayer BE. Influence of oil temperature on the combustion Verlag; 2006.
in a prechamber compression ignition engine. NACA Technical note 1936; [75] Flower DL. The effect of elevated pressure on the rate of soot production in
no. 565. laminar diffusion flames. Combustion Science and Technology 1986;48:
[45] Spadaccini LJ. Auto-ignition characteristics of hydrocarbon fuels at elevated 31–43.
temperature and pressure. ASME Journal of Engineering for Power [76] Rah SC, Sarofim AF, Beer JM. Ignition and combustion of liquid fuel droplets,
1977;99(1):83–7. impact on pollution. Combustion Science and Technology 1986;48:273–84.
[46] Walsh G, Cheng WK. Effects of highly heated fuel on diesel combustion, SAE [77] Uyehara OA. Diesel combustion temperature on soot. International congress
international congress and exposition, 1985; paper no. 850088. and expositions. SAE1980; paper no. 800969.
[47] Scharnweber DH, Hoppie LO. Hyperbolic combustion in internal combus- [78] Kagami MAY, Date K, Maeda T. The influence of fuel properties on the
tion engine. SAE international congress and exposition, SAE1985; paper no. performance of Japanese automobile diesels. SAE international congress
850089. and expositions. SAE 1984; paper no. 841082.
[48] Parker TE, Forsha MD, Stewart HE, Hom. KS. Induction period for ignition of [79] Tateishi MI, Ishikwa H, Harada S. New combustion system of the IDI diesel
fuel sprays at high temperature and pressure. SAE international congress engine. SAE international congress and expositions. SAE1984; paper
and exposition. SAE1985; paper no. 850087. no.841081.
[49] Hardenberg HO, Hase FW. An empirical formula for computing the pressure [80] Sinnamon JF, Cole DE. The Influence of overall equivalence ratio and degree
rise and delay of a fuel from its cetane number and from the relevant of stratification on the fuel consumption and emission of prechamber
parameters of direct injection diesel engines. SAE international congress stratified—charge engine. SAE international congress and expositions. SAE
and exposition, SAE1979; paper no. 790493. 1979; paper no.790438.
[50] Dent DH, Mehta PS. Phenomenological combustion model for a quiescent [81] Anna DA. Combustion in diesel engines. Institution of Recherche on
chamber diesel engine. SAE international congress and exposition. SAE combustion, CNR, University of Nepals Italy; 1987.
1985; paper no. 850087. [82] Lipkea WH, DeJoode AD. Direct injection diesel engine soot modeling
[51] Belardini P, Bertoil C, Corcine DH, Police G. Ignition delay measurement in a formulation and results. SAE1994; paper no. 940670.
direct injection diesel engine. In: International conference on combustion [83] Ziarati R. Mathematical modeling and computer simulation of medium size
engineering MechE, vol. 2. London: Conference publication; 1983. p. 1–7. running on varying quality fuels. In: International Symposium on COMO-
[52] Xie MZ. Combustion calculation of internal combustion engine. Dalian DIA; 1990. p. 587–94.
University of Technology Press, Dalian; 2005. [84] Ishida M, Chen ZL, Ueki H, Sakaguchi D. Combustion analysis by two zone
[53] Wu KJ. Effects of compression ratio on combustion characteristics of a model in a DI diesel engine. In: International symposium on COMODIA;
direct injection diesel engine. SAE international fuels and lubricants meet- 1994.
ing and expositions, 1987. Paper no. 872056. [85] Rakopolous CD, Talkis GN, Tzanos EI. Analysis of combustion chamber
[54] Annand WJD. Heat transfer in the cylinders of reciprocating internal insulation effects on the performance and exhaust emissions of a DI diesel
combustion engines. Proceeding of the Institution of Mechanical Engineers engine using a multizone model. Pergamon 1995;15(7):691–706.
1963;177:983–90. [86] Belardini P, Bertoli C, Beatrice C, Anna DA, Giacomo ND. Application of a
[55] Keribar R, Morel T. Thermal shock calculation in I.C. Engine. SAE 1987; reduced kinetic model for soot formation and burnout in three dimensional
paper no. 870162. diesel engine combustion computations. In: Twenty-sixth symposium
[56] Alkidas AC, Cole RM. Transient heat flux measurement in a divide chamber (international) on combustion/the Combustion Institute; 1996. p. 2517–24.
diesel engine. ASME Journal of Heat Transfer 1985;107(2):439–44. [87] Jung D, Assanis ND. Multi-zone DI diesel spray combustion model for cycle
[57] Hernandez JJ. A combustion kinetic model for estimating diesel engine NOX simulation studies of engine performance and emissions. SAE Technical
emissions. Combustion Theory and Modelling 2006;10(4):639–57. Papers Series 2001-01-1246.
[58] Rakopoulos CD, Rakopoulos DC, Giakoumis EG, Kyritsis DC. Validation and [88] Papagiannakis RG, Hountalas DT. Combustion and exhaust emission char-
sensitivity analysis of a two zone diesel engine model for combustion and acteristics of a dual fuel compression ignition engine operated with pilot
emissions prediction. Energy Conversion and Management 2004;45:1471–95. diesel fuel and natural gas. Energy Conversion and Management 2004;45:
[59] Shahed SM, Chiu WS, Lyn WT. A mathematical model of diesel combustion. 2971–87.
Combustion in engines. Institution of Mechanical Engineers; 1975. Paper [89] Zheng QP, Zhang HM, Zhang DF. A computational study of combustion in
no. C94/75. p. 119–28. compression ignition natural gas engine with separated chamber. Fuel
[60] Vickland CW, Strange FM, Bell RA, Starkman ES. A consideration of high 2005;84:1515–23.
temperature thermodynamics internal combustion engines. Transactions of [90] Tauzia X, Maiboom A, Chesse P, Thouvenel N. A new phenomenological heat
the SAE 1962;70:785–93. release model for thermodynamic simulation of modern turbocharged
[61] Rakopolous CD, Hountalas DT, Tzanos EI, Talkis GN. A fast algorithm for heavy duty diesel engine. Applied Thermal Engineering 2006;26:1851–7.
calculation the composition of diesel combustion product using an eleven [91] Xingcai L, Yuchun H, Linlin Z, Zhen H. Experimental study on the auto-
species chemical equilibrium scheme. Advances in Engineering Software ignition and combustion characteristics in the homogeneous charge com-
1994;19:109–19. pression ignition (HCCI) combustion operation with ethanol/n-heptane
[62] Lavoie GA, Heywood JB, Keck JC. Experimental and theoretical study of blend fuels by port injection. Fuel 2006;85:2622–31.
nitric oxide formation in internal combustion engines. Combustion Science [92] Chmela FG, Priker GH, Wimmer A. zero dimensional ROHR simulation for DI
and Technology 1970:313–26. diesel engine. Energy Conversion and Management 2007;48:2942–50.
[63] Baranescu A. Influence of fuel sulfur on diesel particulate emissions. [93] Sundarapandian S, Devaradjane G. Performance and emission analysis of
SAE 1988; paper no. 881174. biodiesel operated CI engine. Journal of Engineering Computing and
[64] Bennethum JE, Winsor RE. Toward improved diesel fuel, SAE 1991; paper Architecture 2007;1:2.
no. 912325. [94] Machrafia H, Cavadiasa S. An experimental and numerical analysis of the
[65] Ulman TL, Mason RL, Montalvo DA. Effects of fuel aromatics, cetane influence of the inlet temperature, equivalence ratio and compression ratio
number, and cetane improver on emissions from a 1991 prototype heavy- on the HCCI auto-ignition process of primary reference fuels in an engine.
duty diesel engine. SAE 1990; paper no. 902171. Fuel Processing Technology 2008;89:1218–26.
[66] Sienicki EJ, Jass RE, McCathy CI, Slodwske WJ, Krodel AL. Diesel fuel [95] Sahin Z, Durgun O. Multi-zone combustion modeling for the prediction of
aromatic and cetane number effects on combustion and emissions from a diesel engine cycles and engine performance parameters. Applied Thermal
prototype 1991 diesel engine. SAE 1990; paper no. 902172. Engineering 2008;28:2245–56.
530 S. Kumar et al. / Renewable and Sustainable Energy Reviews 19 (2013) 517–530

[96] Semin, Bakar RA, Ismail AR. Investigation of diesel engine performance [105] Prasad BR, Porai PT, Shabir MF. Two-zone modeling of diesel/biodiesel
based on simulation. American Journal of Applied Sciences 2008;5(6): blended fuel operated ceramic coated direct injection diesel engine. Inter-
610–7. national Journal of Energy and Environment 2010;1(6):1039–56.
[97] Garcia MT, FJJE Aguilar, Lencero TS. Experimental study of the performances [106] Qi K, Feng L, Leng X. Simulation of quasi dimensional combustion model for
of a modified diesel engine operating in homogeneous charge compression predicting diesel engine performance. Applied Mathematical Modeling
ignition (HCCI) combustion mode versus the original diesel combustion 2010;35:930–40.
mode. Energy 2009;34:159–71. [107] Rakopolous CD, Kosmadakis GM, Pariotis EG. Investigation the piston bowl
[98] Huang C, Yao M, Lu X, Huang Z. Study of dimethyl ether homogeneous geometry and speed effects in a motored HSDI diesel engine using a CFD
charge compression ignition combustion process using a multi-dimensional against a quasi dimensional model. Energy Conversion and Management
computational fluid dynamics model. International Journal of Thermal 2010;51:470–84.
Sciences 2009;48:1814–22. [108] Shankra BJ, Ganeshan V. Effect of injection timing and intake pressure on
[99] Kannan K, Udayakumar M, Kovilvenni TN. Modeling of nitric oxide forma- the performance of a DI diesel engine a parametric approach. Energy
tion in single cylinder direct injection diesel engine using diesel–water Conversion and Management 2010;51:1835–48.
emulsion. American Journal of Applied Sciences 2009;6(7):1313–20. [109] Shi Y, Reitz RD. Optimization of a heavy-duty compression–ignition engine
[100] Komninos NP. Investigating the importance of mass transfer on the forma- fueled with diesel and gasoline-like fuels. Fuel 2010;89:3416–30.
tion of HCCI engine emissions using a multi-zone model. Applied Energy [110] Som S, Aggarwal SK. Effects of primary breakup modeling on spray and
2009;86:1335–43. combustion characteristics of compression ignition engines. Combustion
[101] Kuleshov AS. Multi-zone DI diesel spray combustion model for thermo- and Flame 2010;157:1179–93.
dynamic simulation of engine with PCCI and high EGR level. SAE Interna- [111] Reiter AJ, Kong SC. Combustion and emissions characteristics of compression-
tional 2009;2009:01–1956. ignition engine using dual ammonia–diesel fuel. Fuel 2011;90:87–97.
[102] Liu H, Xie M, Wu D. Simulation of a porous medium (PM) engine using a [112] Boretti AA. Stochastic reactor modelling of multi modes combustion with
two-zone combustion model. Applied Thermal Engineering 2009;29: diesel direct injection or hydrogen jet ignition start of combustion. Inter-
3189–97. national Journal of Hydrogen Energy 2012;37:13555–5563.
[103] Dhuchakallaya I, Watkins AP. Application of spray combustion simulation in [113] Komninos NP, Rakopoulos CD. Modeling HCCI combustion of biofuels:
DI diesel engine. Applied Energy 2010;87:1427–32. a review. Renewable and Sustainable Energy Reviews 2012;16:1588–610.
[104] Gogoi TK, Baruah DC. A cycle simulation model for predicting the perfor- [114] Visakhamoorthy S, Tzanetakis T, Haggith D, Sobiesiak A, Wena JZ. Numerical
mance of diesel engine fulled by diesel end biodiesel blends. Energy study of a homogeneous charge compression ignition (HCCI) engine fueled
2010;35:1317–23. with biogas. Applied Energy 2012;92:437–46.