0% found this document useful (0 votes)
2 views

Lecture-10 - 2D Velocity-Pressure Coupling

Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
2 views

Lecture-10 - 2D Velocity-Pressure Coupling

Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 16

MEE 514 – Computational Fluid Dynamics

1
Lecture 10
2D Velocity-Pressure Coupling
Dr Gilbert Accary
LECTURE PLAN
 VelocityPressure Coupling Problem
 Staggered Grid
 Discretized Momentum Equations
 SIMPLE Algorithm
• Velocity Correction
• Pressure Correction Equation
• Under-relaxation
 Implementation
 Example: Lids driven cavity with heat transfer

2
GOVERNING EQUATIONS

For a Newtonian, incompressible, 2D flow:


 X-momentum equation:

 Y-momentum equation:

 Continuity equation:

3
VELOCITY-PRESSURE COUPLING PROBLEM

 Non-linear quantities
 All three equations are intrinsically coupled
 Pressure appears in both momentum equations,
but there is no transport equation for pressure
Solution: Iterative strategy such as the SIMPLE,
SIMPLER or PISO algorithms
Ex: SIMPLE algorithm:
1. Use a guessed pressure field to determine velocity
2. Continuity equation is turned into a pressure correction
equation and solved to obtain a pressure correction
3. This pressure correction is used to update velocity
4
STAGGERED GRID (HARLOW AND WELCH, 1965)

 In the x-momentum eq.:


N
u-cell
n vn

uw ue
 In the y-momentum eq.: W P E
w e
y
s vs
v-cell
S
 No interpolation is required
to calculate velocities at the
scalar cell faces x

5
DISCRETIZATION OF THE MOMENTUM EQUATIONS

 X-momentum equation:
, , , i,J+1 I,J+1
u-cell
, , I, j+1

 Y-momentum equation: i-1,J I,J i+1,J

I-1,J i,J I+1,J


, , ,
I,j I+1, j
, , I-1, j
v-cell
i,J-1 I,J-1
 Coefficients , , , , and are
obtained using the velocity field
I, j-1
at the previous iteration

 Pressure field is determined by forcing the velocity 6


components to satisfy the continuity equation
SIMPLE ALGORITHM  PATANKAR AND SPALDING (1972)
(Semi-Implicit Method for Pressure-Linked Equation)

 Using a guessed pressure field P*, guessed velocity


component u* and v* are obtained
 X-momentum equation:
∗ ∗ ∗ ∗
, , , , ,

 Y-momentum equation:
∗ ∗ ∗ ∗
, , , , ,

 We define velocity and pressure correction u’, v’, P’


fields as the difference between the correct fields
(u,v,P) and the guessed one (u*,v*,P*)
7
u = u*+u’ v = v*+v’ P = P*+P’
SIMPLE ALGORITHM  VELOCITY CORRECTION

 Subtracting of correct and guessed discretized


momentum equations:
∗ ∗
, , ,
∗ ∗
, , , ,

∗ ∗
, , ,
∗ ∗
, , , ,

 Using the definition of the correction fields:


, , , ,

, , , ,

Main approximation in the algorithm 8


@ convergence u’ = 0 and v’ = 0
SIMPLE ALGORITHM  VELOCITY CORRECTION (CONT.)

 We obtain for the velocities corrections:

, , , , ,
,

, , , , ,
,

 Corrected velocities:
I, j+1

, , , , ,
I,J
i,J i+1,J

, , , , ,

Similar expressions can be written I,j


for , and ,
9
SIMPLE ALGORITHM  PRESSURE CORRECTION

I, j+1
 Velocity must satisfy continuity:
I,J
i,J i+1,J

 Integrated over a scalar CV: I,j

, , , ,

 Using the expressions of the corrected velocities:



, , , , ,


, , , , ,

We obtain after arranging an equation for pressure


correction
10
SIMPLE ALGORITHM  PRESSURE CORRECTION (CONT.)

 Pressure correction equation:


, , , , , ,
I,J+1
, , , , ,

 where:
, , I,J
I-1,J I+1,J
, ,

, ,

, , I,J-1

, , , , ,
∗ ∗ ∗ ∗
, , , , ,
11
SIMPLE ALGORITHM

12
UNDER-RELAXATION
 To remedy against divergence problem in the iterative
process, a relaxation coefficient  is introduced
 For a generic property  :

Instead we write:

new aP new bP
 Usually  < 1 (underrelaxation)
 Increased diagonal dominance
13
  > 1 (overrelaxation) is sometimes allowed to
accelerate an easy converging process
UNDER-RELAXATION (CONT.)
 In SIMPLE algorithm, pressure correction equation is
susceptible to divergence unless some under-relaxation
is used:

where p is the pressure under-relaxation factor (0<p<1).

 The velocities are also under-relaxed:


,
, , ,

new ,
,
new
, , ,

,
, , ,

new ,
,
new , 14
, ,
IMPLEMENTATION
Steady 2D incompressible flow with heat transfer
 X-momentum equation: q=0

 Y-momentum equation: Th

Vw Ve
 Continuity equation:
Tc
 Energy equation:
q=0

• We use a uniform grid 15


• We assume constant physical properties
EXAMPLE
Steady 2D incompressible flow with heat transfer

 Data: q=0
• Square cavity with L = 1 m
•  = 1 kg/m3,  = 0.1 kg/m.s Th

•  = 0.001 Pa.s
Vw Ve
• Th = 400K, Tc = 300K
• Ve = Vw = 10 m/s
Tc
 = 10,000

q=0

16

You might also like