Scribd
Upload
44 views

CH 9 Lecture 1

This document discusses the special features of solving the Navier-Stokes equations in computational fluid dynamics (CFD). It describes how the Navier-Stokes equations can be written as a set of coupled partial differential equations along with the continuity and energy equations. It also discusses two arrangements for variable discretization - co-located and staggered grids. The key challenge in solving the equations for incompressible flow is the lack of a pressure equation; the document derives the pressure Poisson equation used to solve for pressure in such cases.

Uploaded by

Saumya Sinha
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
44 views

CH 9 Lecture 1

This document discusses the special features of solving the Navier-Stokes equations in computational fluid dynamics (CFD). It describes how the Navier-Stokes equations can be written as a set of coupled partial differential equations along with the continuity and energy equations. It also discusses two arrangements for variable discretization - co-located and staggered grids. The key challenge in solving the equations for incompressible flow is the lack of a pressure equation; the document derives the pressure Poisson equation used to solve for pressure in such cases.

Uploaded by

Saumya Sinha
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 3

COMPUTATIONAL FLUID DYNAMICS: N-S EQUATIONS: Special Features

Dr K M Singh, Indian Institute of Technology Roorkee NPTEL 32.1


Lecture 32
NAVIER-STOKES EQUATIONS: SPECIAL FEATURES

32.1 INTRODUCTION
Numerical simulation of the flow problems involves solution of Navier-Stokes equation. This
vector equation is essentially a set of three coupled partial differential equations, and must be
solved along with the continuity equation and energy equation. All these equations are
coupled PDEs, and their collection is often referred to as Navier-Stokes equations in CFD
literature. One remarkable feature of these equations is the similarity of their form: each of
these equations can be recast in the form of the generalized transport equation which contains
a time derivative term, a convective term, a diffusive terms and a source term. For
compressible flows, the continuity equation, momentum equation and energy equation
represent the transport equations for density, velocity and temperature respectively.
Therefore, the discretization methods discussed earlier can be applied to each equation, and
collection of the resulting discretized nonlinear equations can solved using a sequential
iterative scheme. Thus, algorithm for numerical solution of unsteady subsonic compressible
flow problem is relatively straightforward in the sense that it is an extension of the algorithms
discussed in the preceding lectures for solution of generic transport equation.

Governing equations for incompressible flows are also similar in form to the generic
transport equation. However, there is a small problem due to non-existence of a separate
equation for pressure. In this case, there is no equation of state relating pressure, temperature
and density, and continuity equation reduces to a kinematic constraint on velocity field. This
requires special attention which we briefly review in this section lecture.
32.2 SPECIAL FEATURES OF NAVIER-STOKES EQUATIONS
Navier-Stokes equations are governing equations for a vector field (velocity). This allows
more freedom in choice of the grid used in numerical simulation. Depending on the choice of
discretization scheme (finite difference, finite volume or finite element method), there are
two possible choices for arrangement of the problem variables on grid nodes: (a) collocated
arrangement and (b) staggered arrangement. Each of these choices has its own advantages
and disadvantages which are as follows:

a) Co-located arrangement
If all the variables are stored at the same set of grid point, then the grid is called co-
located grid (Figure 32.1a). This choice simplifies the programming and allows the
use of same restriction and prolongation operators for each variable when multigrid
methods are employed. Its major disadvantage is lack of pressure-velocity coupling
which may lead to oscillations in pressure field (i.e. the chequer-board pattern for
pressure field).

b) Staggered arrangement
In staggered arrangement (Figure 32.1b), pressure/scalar nodes lie at the centroids of
grid cells, and velocity nodes are located at the centre of respective cell faces in case
of FDM/FVM. Separate set of elements are employed in case of FEM. Disadvantage
of this approach lies in elaborate book-keeping and requirement of separate set of
C
D



COMPUTA
Dr K M Sin


Figure
numer
scalar



Steady
Stokes
associat
as an un
solution
problem

32.3 C
Solution
indepen
comput
incomp
conserv
difficult
equation
momen
of a Poi





ATIONAL
ngh, Indian
multigrid o
the strong c
arrangemen
(a) Co-lo
e 32.1 Finit
rical solution
variables.
Another sp
state Navie
equations a
ted with so
nsteady pro
n of this tran
m.
COMPUT
n of incom
ndent equati
te density
ressible flo
vation is ess
ty is to con
n (i.e. to e
ntum and co
isson equati
To derive th
(
p
c
V =

c

FLUID DY
n Institute o
operators fo
coupling be
nt has been
cated arrang
e volume gr
n of Navier
ecial featur
er-Stokes e
are paraboli
lution of pu
oblem even
nsient Navi
TATION O
mpressible
ion for pres
and press
ows, continu
sentially a k
nstruct the
enforce ma
ontinuity equ
ion, and hen
he pressure
( )
(
v
t

+ V
c
YNAMICS:
of Technolo
r each prob
etween the v
the most po
gement
rid arrangem
r-Stokes equ
e relates to
equations ar
ic in time (
urely ellipti
if the flow
er-Stokes p
OF PRES
Navier-Sto
ssure. In com
sure is co
uity equatio
kinematic c
pressure fie
ss conserva
uations to o
nce, is comm
Poisson eq
) vv b
:
ogy Roorke
blem variab
velocity and
opular in CF
ments for pr
uations. Bul
the mathem
re elliptic i
(and elliptic
ic PDEs, N
is steady, u
problem yiel
SSURE
okes equatio
mpressible
omputed fr
on does no
constraint o
eld so as to
ation). To
obtain an eq
monly refer
quation, let u
(
V
(


N-S E
ee
ble. Biggest
d pressure fi
FD analysis
(b) S
ressure and
llet represen
matical natu
in nature, w
c in space).
avier-Stoke
using a time
lds the solu
ons is com
flows, cont
rom equati
ot have a d
on the veloc
o guarantee
achieve thi
quation for
rred to as th
us rearrange

QUATION
NPTE
advantage
ield because
.
Staggered ar
velocity co
nts computa
ure of Navie
whereas the
Due to nu
es equations
e marching
tion to actu
mplicated b
tinuity equa
ion of sta
ominant va
city field. T
e satisfactio
is objective
pressure, w
e pressure P
e the mome
NS: Special
EL
of this app
e of which t
rrangement
omponents f
ational node
er-Stokes eq
e unsteady
umerical dif
s are mostly
scheme. Lo
ual steady st
by the lack
ation can be
ate. Howev
ariable. Thu
The way ou
n of the co
e, we comb
which is in t
Poisson equ
ntum as

l Features
32.2
proach is
this grid
for
e for
quations.
Navier-
fficulties
y solved
ong-time
tate flow
k of an
e used to
ver, for
us, mass
ut of this
ontinuity
bine the
the form
uation.
(32.1)



COMPUTATIONAL FLUID DYNAMICS: N-S EQUATIONS: Special Features
Dr K M Singh, Indian Institute of Technology Roorkee NPTEL 32.3
In Cartesian tensor notation, the preceding equation can be re-written as


( )
( )
i
i j ij i
i j
v p
v v b
x t x

t
(
c c c
= + +
(
c c c
(

(32.2)

Taking divergence of the preceding equation, we get


( )
( )
( )
i i
i j ij
i i i j i
v b p
v v
x x x t x x

t
(
c c | | c c c c
= + +
( |
c c c c c c
( \ .

(32.3)

For constant density (incompressible flow) and viscosity, viscous and unsteady terms
disappear by virtue of the continuity equation. Further, body force field is usually gradient of
a scalar function, and hence, its divergence also vanishes. Thus, Eq. (32.3) simplifies to


( )
i j
i i i j
v v
p
x x x x

(
c
| | c c c
= (
|
c c c c
( \ .

(32.4)

Equation (32.4) represents the desired Poisson equation for pressure for incompressible flows
which must be solved using an appropriate elliptic solver. Note that the Laplacian in pressure
equation (9.202) is the product of the divergence operator originating from the continuity
equation and the gradient operator of momentum equation. To maintain numerical
consistency, it is essential that the approximation of the pressure Poisson equation must be
defined as the product of the approximations of divergence and gradient operators employed
in discretization of continuity and momentum equation. Otherwise satisfaction of the
continuity equation cannot be guaranteed.

FURTHER READING
Ferziger, J. H. And Peri, M. (2003). Computational Methods for Fluid Dynamics. Springer.
Versteeg, H. K. and Malalasekera, W. M. G. (2007). Introduction to Computational Fluid
Dynamics: The Finite Volume Method. Second Edition (Indian Reprint) Pearson Education.

You might also like