Fluent - D

Fluent Inc.
5/16/2012 D1
Fluent Software Training
TRN-98-006
Modeling Turbulent Flows
Fluent Inc. 5/16/2012 D2
TRN-98-006
Unsteady, aperiodic motion in which all three velocity components
fluctuate mixing matter, momentum, and energy.
Decompose velocity into mean and fluctuating parts:
U
i
(t) U
i
+ u
i
(t)

Similar fluctuations for pressure, temperature, and species
concentration values.
What is Turbulence?
Time
U
i
(t)
U
i

u
i
(t)
TRN-98-006
Why Model Turbulence?
Direct numerical simulation of governing equations is only possible for
simple low-Re flows.
Instead, we solve Reynolds Averaged Navier-Stokes (RANS)
equations:

where (Reynolds stresses)

Time-averaged statistics of turbulent velocity fluctuations are modeled
using functions containing empirical constants and information about
the mean flow.
Large Eddy Simulation numerically resolves large eddies and models
small eddies.
(steady, incompressible flow
w/o body forces)
j i ij
u u R =
j
ij
j j
i
i k
i
k
x
R
x x
U
x
p
x
U
U
c
c
+
c c
c
+
c
c
=
c
c
2

TRN-98-006
Is the Flow Turbulent?
External Flows
Internal Flows
Natural Convection
5
10 5 >
x
Re along a surface
around an obstacle
where
UL
Re
L
where
Other factors such as free-stream
turbulence, surface conditions, and
disturbances may cause earlier
transition to turbulent flow.
L = x, D, D
h
, etc.
,300 2 >
h
D
Re
10 8
10 10 > Ra
o
|
3
TL g
Ra
A
20,000 >
D
Re
TRN-98-006
How Complex is the Flow?
Extra strain rates
Streamline curvature
Lateral divergence
Acceleration or deceleration
Swirl
Recirculation (or separation)
Secondary flow
3D perturbations
Transpiration (blowing/suction)
Free-stream turbulence
Interacting shear layers
TRN-98-006
Choices to be Made
Turbulence Model
&
Near-Wall Treatment
Flow
Physics
Accuracy
Required
Computational
Resources
Turnaround
Time
Constraints
Computational
Grid
TRN-98-006
Zero-Equation Models

One-Equation Models
Spalart-Allmaras
Two-Equation Models
Standard k-c
RNG k-c
Realizable k-c
Reynolds-Stress Model

Large-Eddy Simulation

Direct Numerical Simulation
Turbulence Modeling Approaches
Include
More
Physics
Increase
Computational
Cost
Per Iteration
Available
in FLUENT 5
RANS-based
models
TRN-98-006
RANS equations require closure for Reynolds stresses.

Turbulent viscosity is indirectly solved for from single transport
equation of modified viscosity for One-Equation model.
For Two-Equation models, turbulent viscosity correlated with turbulent
kinetic energy (TKE) and the dissipation rate of TKE.

Transport equations for turbulent kinetic energy and dissipation rate are
solved so that turbulent viscosity can be computed for RANS equations.
Reynolds Stress Terms in RANS-based Models
Turbulent
Kinetic Energy:
Dissipation Rate of
Turbulent Kinetic Energy:
c

2
k
C
t
Turbulent Viscosity:
Boussinesq Hypothesis:
(isotropic stresses)
|
|
.
|
\
|
c
c
+
c
c
+ = =
i
j
j
i
t ij j i ij
x
U
x
U
k u u R o
3
2
2 /
i i
u u k
|
|
.
|
\
|
c
c
+
c
c
c
c
i
j
j
i
j
i
x
u
x
u
x
u
v c
TRN-98-006
Turbulent viscosity is determined from:

is determined from the modified viscosity transport equation:

The additional variables are functions of the modified turbulent
viscosity and velocity gradients.
One Equation Model: Spalart-Allmaras
( )
2
1
2
2
~
1
~ ~ ~
~
1
~
~
~
d
f c
x
c
x x
S c
Dt
D
w w
j
b
j j
b
v
v
v
o
v
v
v

(
(
|
|
.
|
\
|
c
c
+
c
c
+
c
c
+ =
( )
( )
(
+
=
3
1
3
3
/
~
/
~
~
v
v v
v v
v
c
t
v
~
Generation Diffusion
Destruction
TRN-98-006
One-Equation Model: Spalart-Allmaras
Designed specifically for aerospace applications involving wall-
bounded flows.
Boundary layers with adverse pressure gradients
turbomachinery
Can use coarse or fine mesh at wall
Designed to be used with fine mesh as a low-Re model, i.e., throughout
the viscous-affected region.
Sufficiently robust for relatively crude simulations on coarse meshes.
TRN-98-006
Two Equation Model: Standard k-c Model
Turbulent Kinetic Energy
Dissipation Rate
c c c
o o
2 1 ,
, , C C
k
are empirical constants
(equations written for steady, incompressible flow w/o body forces)
Convection
Destruction
c o
)
`
c
c
c
c
+
c
c
|
|
.
|
\
|
c
c
+
c
c
=
c
c

i
k t
i i
j
j
i
i
j
t
i
i
x
k
x x
U
x
U
x
U
x
k
U ) (
Destruction
Convection

|
|
.
|
\
|
)
`
c
c
c
c
+
c
c
|
|
.
|
\
|
c
c
+
c
c
|
.
|
\
|
=
c
c
k
C
x x x
U
x
U
x
U
k
C
x
U
i
t
i i
j
j
i
i
j
t
i
i
2
2 1
) (
c
c
o
c c
c c c
TRN-98-006
Two Equation Model: Standard k-c Model
Baseline model (Two-equation)
Most widely used model in industry
Strength and weaknesses well documented
Semi-empirical
k equation derived by subtracting the instantaneous mechanical energy
equation from its time-averaged value
c equation formed from physical reasoning
Valid only for fully turbulent flows
Reasonable accuracy for wide range of turbulent flows
industrial flows
heat transfer
TRN-98-006
Two Equation Model: Realizable k-c
Distinctions from Standard k-c model:
Alternative formulation for turbulent viscosity
where is now variable

(A
0
, A
s
, and U* are functions of velocity gradients)
Ensures positivity of normal stresses;
Ensures Schwarzs inequality;

New transport equation for dissipation rate, c:

c

2
k
C
t

c
k U
A A
C
s o
*
1
+
=
0 u
2
i
>
2
j
2
i
2
j i
u u ) u u ( s
b
j
t
j
G c
k
c
k
c S c
x x Dt
D
c c
c
c
vc
c
c
c
o
3 1
2
2 1
+
+
+
(
(
c
c
|
|
.
|
\
|
+
c
c
=
Generation Diffusion Destruction Buoyancy
TRN-98-006
Shares the same turbulent kinetic energy equation as Standard k-c
Superior performance for flows involving:
planar and round jets
boundary layers under strong adverse pressure gradients, separation
rotation, recirculation
strong streamline curvature
Two Equation Model: Realizable k-c
TRN-98-006
Two Equation Model: RNG k-c
Turbulent Kinetic Energy
Dissipation Rate
Convection Diffusion
Dissipation

c o
|
|
.
|
\
|
c
c
c
c
+ =
c
c

i
k
i
t
i
i
x
k
x
S
x
k
U
eff
2
Generation
|
|
.
|
\
|
c
c
+
c
c

j
i
i
j
ij ij ij
x
U
x
U
S S S S
2
1
, 2
where
are derived using RNG theory
c c c
o o
2 1 ,
, , C C
k
Additional term
related to mean strain
& turbulence quantities
Convection Generation Diffusion Destruction
R
k
C
x x
S
k
C
x
U
i i
t
i
i

|
|
.
|
\
|
|
|
.
|
\
|
c
c
c
c
+
|
.
|
\
|
=
c
c

2
2 eff
2
1
c
c
o
c c
c c c
TRN-98-006
Two Equation Model: RNG k-c
k-c equations are derived from the application of a rigorous statistical
technique (Renormalization Group Method) to the instantaneous Navier-
Stokes equations.
Similar in form to the standard k-c equations but includes:
additional term in c equation that improves analysis of rapidly strained flows
the effect of swirl on turbulence
analytical formula for turbulent Prandtl number
differential formula for effective viscosity
Improved predictions for:
high streamline curvature and strain rate
transitional flows
wall heat and mass transfer
TRN-98-006
Reynolds Stress Model
k
ijk
ij ij ij
k
j i
k
x
J
P
x
u u
U
c
c
+ u + =
c
c
c
Generation
k
i
k j
k
j
k i ij
x
U
u u
x
U
u u P
c
c
+
c
c
|
|
.
|
\
|
c
c
+
c
c
' u
i
j
j
i
ij
x
u
x
u
p
k
j
k
i
ij
x
u
x
u
c
c
c
c
c 2
Pressure-Strain
Redistribution
Dissipation
Turbulent
Diffusion
(modeled)
(related to c)
(modeled)
(computed)
Reynolds Stress
Transport Eqns.
Pressure/velocity
fluctuations
Turbulent
transport
) (
j ik i jk k j i ijk
u u p u u u J o o + ' + =
TRN-98-006
Reynolds Stress Model
RSM closes the Reynolds-Averaged Navier-Stokes equations by
solving additional transport equations for the Reynolds stresses.
Transport equations derived by Reynolds averaging the product of the
momentum equations with a fluctuating property
Closure also requires one equation for turbulent dissipation
Isotropic eddy viscosity assumption is avoided
Resulting equations contain terms that need to be modeled.
RSM has high potential for accurately predicting complex flows.
Accounts for streamline curvature, swirl, rotation and high strain rates
Cyclone flows, swirling combustor flows
Rotating flow passages, secondary flows

TRN-98-006
Large Eddy Simulation
Large eddies:
Mainly responsible for transport of momentum, energy, and other scalars,
directly affecting the mean fields.
Anisotropic, subjected to history effects, and flow-dependent, i.e., strongly
dependent on flow configuration, boundary conditions, and flow parameters.
Small eddies:
Tend to be more isotropic and less flow-dependent
More likely to be easier to model than large eddies.
LES directly computes (resolves) large eddies and models only small
eddies (Subgrid-Scale Modeling).
Large computational effort
Number of grid points, N
LES

Unsteady calculation
2
Re
t
u
TRN-98-006
Comparison of RANS Turbulence Models
Model Strengths Weaknesses
Spalart-
Allmaras
Economical (1-eq.); good track record
for mildly complex B.L. type of flows
Not very widely tested yet; lack of
submodels (e.g. combustion,
buoyancy)
STD k-c
Robust, economical, reasonably
accurate; long accumulated
performance data
Mediocre results for complex flows
involving severe pressure gradients,
strong streamline curvature, swirl
and rotation
RNG k-c
Good for moderately complex
behavior like jet impingement,
separating flows, swirling flows, and
secondary flows
Subjected to limitations due to
isotropic eddy viscosity
assumption
Realizable
k-c
Offers largely the same benefits as
RNG; resolves round-jet anomaly
Subjected to limitations due to
isotropic eddy viscosity
assumption
Reynolds
Stress
Model
Physically most complete model
(history, transport, and anisotropy of
turbulent stresses are all accounted
for)
Requires more cpu effort (2-3x);
limited near-wall modeling options;
tightly coupled momentum and
turbulence equations
TRN-98-006
Near-Wall Treatments
Most k-c and RSM turbulence
models will not predict correct
near-wall behavior if integrated
down to the wall.
Special near-wall treatment is
required.
Standard wall functions
Nonequilibrium wall functions
Two-layer zonal model
Boundary layer structure
TRN-98-006
Standard Wall Functions
t

/
2 / 1 4 / 1
w
P P
k C U
U
-
( )
>
(
+
<
=
-
-
-
) ( ln
1
Pr
) ( Pr
* *
* *
T t
T
y y P Ey
y y y
T
k
P P
y k C
y
2 / 1 4 / 1
-
q
k C c T T
T
P p P w
' '
2 / 1 4 / 1
) (
*

Mean Velocity
Temperature
where
where
and P is a function of the fluid
and turbulent Prandtl numbers.
thermal sublayer thickness
( )
- -
= Ey U ln
1
k
TRN-98-006
Nonequilibrium Wall Functions
Log-law is sensitized to pressure gradient for
better prediction of adverse pressure gradient
flows and separation.
Relaxed local equilibrium assumptions for
TKE in wall-neighboring cells.
Thermal law-of-wall unchanged
|
|
.
|
\
|
=
k
t

y
k C
E
k C
U
w
2 / 1 4 / 1 2 / 1 4 / 1
ln
1
/
~
(
+
|
.
|
\
|
=
- -
k k
y
k
y y
y
y
k
y
dx
dp
U U
v v
v
v
2
2 / 1 2 / 1
ln
2
1
~
where
TRN-98-006
Two-Layer Zonal Model
Used for low-Re flows or
flows with complex near-wall
phenomena.
Zones distinguished by a wall-
distance-based turbulent
Reynolds number
High-Re k-c models are used in the turbulent core region.
Only k equation is solved in the viscosity-affected region.
c is computed from the correlation for length scale.
Zoning is dynamic and solution adaptive.
y k
Re
y

200 >
y
Re
200 <
y
Re
TRN-98-006
Comparison of Near Wall Treatments
Strengths Weaknesses
Standard wall
Functions
Robust, economical,
reasonably accurate
Empirically based on simple
high-Re flows; poor for low-Re
effects, massive transpiration,
Vp, strong body forces, highly
3D flows
Nonequilibrium
wall functions
Accounts for Vp effects,
allows nonequilibrium:
-separation
-reattachment
-impingement
Poor for low-Re effects, massive
transpiration, severe Vp, strong
body forces, highly 3D flows
Two-layer zonal
model
Does not rely on empirical
law-of-the-wall relations,
good for complex flows,
applicable to low-Re flows
Requires finer mesh resolution
and therefore larger cpu and
memory resources
TRN-98-006
Computational Grid Guidelines
Wall Function
Approach
Two-Layer Zonal
Model Approach
First grid point in log-law region

At least ten points in the BL.
Better to use stretched quad/hex
cells for economy.
First grid point at y
+
~ 1.
At least ten grid points within
buffer & sublayers.
Better to use stretched quad/hex
cells for economy.
500 50 s s
+
y
TRN-98-006
Estimating Placement of First Grid Point
Estimate the skin friction coefficient based on correlations either
approximate or empirical:

Flat Plate-

Pipe Flow-

Compute the friction velocity:

Back out required distance from wall:
Wall functions Two-layer model

Use post-processing to confirm near-wall mesh resolution
2 . 0
Re 0359 . 0 2 /

~
L f
c
2 . 0
Re 039 . 0 2 /

~
D f
c
2 / /
f e w
c U u = t
t
y
1
= 250v/u
t
y
1
= v/ u
t
TRN-98-006
Setting Boundary Conditions
Characterize turbulence at inlets & outlets (potential backflow)
k-c models require k and c
Reynolds stress model requires R
ij
and c
Several options allow input using more familiar parameters
Turbulence intensity and length scale
length scale is related to size of large eddies that contain most of energy.
For boundary layer flows: l ~ 0.4o
99

For flows downstream of grids /perforated plates: l ~ opening size
Turbulence intensity and hydraulic diameter
Ideally suited for duct and pipe flows
Turbulence intensity and turbulent viscosity ratio
For external flows:
Input of k and c explicitly allowed (non-uniform profiles possible).
10 / 1 < <

t
TRN-98-006
GUI for Turbulence Models
Define Models Viscous...
Turbulence Model options
Near Wall Treatments
Inviscid, Laminar, or Turbulent
Additional Turbulence options
TRN-98-006
Example: Channel Flow with Conjugate Heat
Transfer
adiabatic wall
cold air
V = 50 fpm
T = 0 F
constant temperature wall T = 100 F
insulation
1 ft
1 ft
10 ft
P
Predict the temperature at point P in the solid insulation
TRN-98-006
Turbulence Modeling Approach
Check if turbulent Re
D
h
= 5,980
Developing turbulent flow at relatively low Reynolds number and
BLs on walls will give pressure gradient use RNG k-c with
nonequilibrium wall functions.
Develop strategy for the grid
Simple geometry quadrilateral cells
Expect large gradients in normal direction to horizontal walls
fine mesh near walls with first cell in log-law region.
Vary streamwise grid spacing so that BL growth is captured.
Use solution-based grid adaption to further resolve temperature
gradients.
TRN-98-006
Velocity
contours
Temperature
contours
BLs on upper & lower surfaces accelerate the core flow
Prediction of Momentum & Thermal
Boundary Layers
Important that thermal BL was accurately resolved as well
P
TRN-98-006
Example: Flow Around a Cylinder
wall
wall
1 ft
2 ft
2 ft
air
V = 4 fps
Compute drag coefficient of the cylinder
5 ft 14.5 ft
TRN-98-006
Check if turbulent Re
D
= 24,600

Flow over an object, unsteady vortex shedding is expected,
difficult to predict separation on downstream side, and close
proximity of side walls may influence flow around cylinder
use RNG k-c with 2-layer zonal model.

Develop strategy for the grid
Simple geometry & BLs quadrilateral cells.
Large gradients near surface of cylinder & 2-layer model
fine mesh near surface & first cell at y
+
= 1.
Use solution-based grid adaption to further resolve pressure
gradients.
Turbulence Modeling Approach
TRN-98-006
Grid for Flow Over a Cylinder
TRN-98-006
Prediction of Turbulent Vortex Shedding
Contours of effective viscosity
eff
= +
t
C
D
= 0.53 Strouhal Number = 0.297
U
D
St
t
where
TRN-98-006
Summary: Turbulence Modeling Guidelines
Successful turbulence modeling requires engineering judgement of:
Flow physics
Computer resources available
Project requirements
Accuracy
Turnaround time
Turbulence models & near-wall treatments that are available
Begin with standard k-c and change to RNG or Realizable k-c if
needed.
Use RSM for highly swirling flows.
Use wall functions unless low-Re flow and/or complex near-wall
physics are present.

Fluent - D

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Fluent - D

Uploaded by

Copyright:

Available Formats

Fluent Inc.

You might also like