23
CHAPTER 5
GOVERNING EQUATIONS AND CFD METHODOLOGY
5.1 INTRODUCTION
CFD is concerned with replacing the governing partial differential equations of
fluid flow with manageable algebraic equations and to obtain final numerical description
of the complex flow field of interest. Numerical solution procedure requires the repetitive
manipulation of numbers. These facts are highlighted by Versteeg and Malalasekara
(1995).
5.2 GOVERNING EQUATIONS
The governing equations of fluid flow represent mathematical statements of the
conservation laws of physics.
i.
The mass of a fluid is conserved.
ii.
The rate of change of momentum equals the sum of the forces on a fluid
particle.
iii.
The rate of change of energy is equal to the sum of the rate of heat
addition and the rate of work done on a fluid particle.
5.2.1 Governing Conservation Equations
Continuity equation
The physical principle of the equation is conservation of mass. The mass balance
for the fluid element is that, the rate of increase of mass in fluid element is equal to the
net rate of flow of mass into fluid element. The unsteady, three dimensional mass
conservation or continuity equation at a point in a compressible fluid is given as,
24
u v w
=0
t
x
y
z
(5.1)
In vector notation,
div u = 0
t
(5.2)
where,
(/t)
- Rate of change of density.
Div (u)
- The net flow of mass out of the element across its boundaries
(Convective term)
For an incompressible fluid, the density ( ) is constant, hence
Div (u) = 0 (or)
u v w
=0
x y
z
(5.3)
Momentum equation
According to Newtons law, the rate of change of momentum of a fluid particle
equals the sum of the forces on the particle.
X-Component of momentum equation is,
Du ( p xx) yx zx
SMx
Dt
x
y
z
(5.4)
Y - Component of momentum equation is,
Dv xy ( p yy ) zy
SMy
Dt
x
y
z
Z - Component of momentum equation is,
(5.5)
25
SM(x,y,z) =
Dw xz yz ( p zz )
SMz
Dt
x
y
z
(5.6)
Source momentum per unit volume per unit time in x,y,z
directions respectively.
Energy equation
The energy equation is given as,
1
g t
g h
u~ j h Fh, j 1
x j
g t
~
u j p p u j ij u i si
x j
x j
x j
(5.7)
h c pT c opTo mm H m ht mm H m
(5.8)
5.3 TURBULENCE MODEL EQUATION
One of the most predominant turbulence models, the k- model, has been
implemented in most general purpose CFD codes and is considered the industry standard
model. It has proven to be stable and numerically robust and has a well established
regime of predictive capability. k is the turbulent kinetic energy and is defined as the
variance of the fluctuations in velocity. is the turbulence eddy dissipation (the rate at
which the velocity fluctuations dissipate). The Renormalisation group (RNG) version of
the k- model [Yakhot and Orszag (1986)] denoted as RNG k- model is employed in
high Reynolds number form in conjunction with law of the wall functions in the present
study.
26
5.4 COMBUSTION MODELS
5.4.1 Diesel Combustion Model
The analysis of the combustion process is carried out by using laminar and
turbulent characteristic-time (LATCT) combustion model of Magnussen based on the
eddy break-up concept. This combustion model is based on the works of Magnussen
[Magnussen and Hjertager (1976)], Abraham [Abraham et al. (1985)] and Patterson
[Patterson et al. (1994)].
The EBU model is based on the assumption that the reaction is a single-step
irreversible one, involving fuel (F), oxidant (O), product (P) and possible background
inert species. Moreover, the model assumes the reaction time scale is so small that the
rate-controlling mechanism is turbulent micro-mixing between burnt and unburnt gases,
relating the rate of combustion to the rate of dissipation of turbulent eddies. The micromixing time scale is k/.
In this model, which is applicable to both premixed and non-premixed systems,
the fuel is calculated from its own differential transport equation, incorporating the
reaction rate; an additional transport equation is solved for the fuel mixture fraction (sum
of burnt and unburnt fuel). According to the model, the fuel consumption rate is given
by:
Rf
mp
m
Aebu min m F , O , Bebu
k
sO
s p
where, So noMo / nFMF
Sp npMp / nFMF
(5.9)
(5.10)
mF, mo, and mp are the fuel, oxygen and product mass fractions respectively. Aebu and Bebu
are dimensionless empirical coefficients and k/ is the turbulent time scale. Due to the
above assumptions, it emerges that a certain degree of tuning of the model is required in
27
order to match experimental data, acting on the rate of fuel consumption by means of the
variation of turbulent conditions and Aebu and Bebu factors.
Since the burning rate is inversely proportional to turbulent time-scale, which
decreased near the walls, the standard EBU model predicts unrealistic flame accelerations
along the walls. To avoid this, the LATCT model is adopted. The turbulent time scale in
the rate of combustion is replaced by characteristic time, which is a combination of the
local laminar and turbulent time scale, in such a way that, at the limit of no combustion,
the laminar time scale is used and at a limit of complete combustion, the sum of the
laminar and the turbulent time scale is adopted, while in between a power law is used.
As previously mentioned, EBU model is based on a single step reaction.
n-dodecane (C12H26) is used as fuel with the reaction mechanism as follows.
C12H26 + 18.5O2 -> 12 CO2+13H2O
(5.11)
5.4.2 Ignition Model
The ignition model used is shell auto-ignition model developed for the study of
knock in gasoline engines [Halstead et al. (1975)], appropriately modified to predict
ignition for fuels of higher molecular weight typical of diesel fuels [Theobald and Cheng
(1987)].
5.5 SPRAY SUB-MODELS
Concerning spray modelling, Huh atomisation model [Huh, K.Y., and Gosman,
A.D (1991)] is used to represent the fuel jet primary break-up at the nozzle exit; the
model is based on the assumption that atomisation is governed by two main mechanisms;
the gas inertia and the internal turbulent stresses generated within the nozzle. The model
28
estimates the initial perturbations from the analysis of the flow through the hole and then
uses established wave growth theory to evaluate the atomisation process.
As far as secondary break-up is concerned, Reitz-Diwakar model [Reitz, R.D.,
and Diwakar, R (1986)] is adopted. The model is based on the assumption that droplets
may become unstable under the action of the interfacial forces generated by their motion
relative to the combustion phase. According to the model, droplets that break-up under
the effect of aerodynamic forces can be subdivided into bag-breakup and stripping
break-up. The first depends on the non-uniform pressure surrounding the droplets, while
the latter is a process in which liquid is stripped from the droplet surface by the
continuous phase. These break-up regimes change according to the Weber number.
Additional sub models are used for the spray simulation, accounting for both
droplet-droplet interaction and droplets-wall impingement. The former characterises
droplets history by three kind of events; coalescence, separation and bouncing, and is
based on the model proposed by ORourke; [O Rourke, P.J (1981)], the latter based on
the model proposed by Bai [Bai, C., and Gosman, A.D (1995)] recognises different
regimes (stick, spread, rebound and splash) according to: incident droplet velocity
relative to the wall, incidence angle, droplet size and properties as well as wall
temperature, surface roughness, film thickness and near-wall gas conditions. Turbulent
dispersion and condensation are activated.
5.6 NOX MODEL
In the course of the combustion process in diesel engines, nitrous oxides,
consisting of nitrogen-dioxide NO2 and nitrogen-monoxide NO, are formed in areas of
high-temperature. Usually considered together as NOX, nitrogen monoxide is by far
dominating the net amount of nitrous oxides emitted from engines. There are two
possible sources of nitrous oxide formation in engines, namely thermal and prompt NO.
In diesel engines, however more than 90 % of the NO emission stems from the thermal
29
NO formation process. As its name suggests, it is strongly temperature dependent. It is
produced by the reaction of atmospheric nitrogen with oxygen at elevated temperatures.
Hence only thermal NO formation which can be described by the extended Zeldovich
mechanism is accounted for in the present study. The time scale for NO X reactions is
larger than the time scales for the turbulent mixing process and the combustion of
hydrocarbons that control the heat-releasing reactions. Hence, the computation of NO X
can be decoupled from the main reacting flow field predictions. For thermal nitric oxide,
the principal reactions are generally recognised by the following three extended
Zeldovich mechanisms:
K1
N 2 O NO N
(5.12)
K 1
K2
N O2 NO O
(5.13)
K 2
K3
N OH NO H
(5.14)
K 3
5.7 SOOT MODEL
The formation and emission of carbonaceous particles is a process that is often
observed during the combustion of hydrocarbons. It is widely accepted that the formation
of soot is a complex process, which consists a lot of complex reactions (fuel pyrolysis
and oxidation, formation of polycyclic an aromatic hydrocarbons; the inception of first
particles etc). An approach is based on the laminar flamelet concept, in which all scalar
quantities are related to the mixture fraction and scalar dissipation rate, whereas the
species mass fractions are unique functions in the mixture fraction-scalar dissipation
space, the soot mass fractions are not. However, the rate of soot formation can be
correlated with local conditions in diffusion flames or in partially premixed counter flow
twin flames. This is the method proposed by Mauss et al. (1998).
30
In this method, an additional transport equation for the soot mass fraction is
solved. Source terms for the soot volume fraction are taken from a flamelet library using
a presumed probability density function and integrating over mixture fraction space. In
order to save computer storage and CPU time, the flamelet library of sources is
constructed using a multi-parameter fitting procedure resulting in simple algebraic
equations and a proper set of parameters. The transport equation for soot mass fraction is
given by:
Ys x
t
u j Ys
t Ys
x j Prt , s x j
(5.15)
where Ys is the soot fraction. The prandtl number for soot is assumed to be 1.4 and
the soot density s = 1860 kg/m3. The source term for the soot volume fraction is given
by:
= , ParIncep + f (, SurfGrowth - ,fragment - ,oxid )
(5.16)
where f = Ys / s is the mean soot volume fraction. The mean source terms are
obtained from the flamelet library mentioned earlier.
