Download - Grid Postprocessing Cfd
-
7/28/2019 Grid Postprocessing Cfd
1/35
Grid Generation and Post-Processing for Computational
Tao Xing and Fred Stern
-
7/28/2019 Grid Postprocessing Cfd
2/35
lin
1. Introduction2. Choice of grid
2.1. Sim le eometries
2.2. Complex geometries3. Grid generation3.1. Conformal mapping3.2. Algebraic methods3.3. Differential equation methods
3.4. Commercial software3.5. Systematic gri generation or CFD UA
4. Post-processing4.1. UA (details in Introduction to CFD)
. .4.3. Calculation of integral variables4.4. Visualization
2
.
-
7/28/2019 Grid Postprocessing Cfd
3/35
-
7/28/2019 Grid Postprocessing Cfd
4/35
Choice of grid . . ,
usually follow the coordinate directions.
Complex geometries (Stepwise Approximation)
or curved boundaries by staircase-like steps.2. Problems:
(1). Number of grid points (or CVs) per grid line is not constant,
special arrays have to be created
(2). Steps at the boundary introduce errors into solutions
(3). Not recommended except local grid refinement near thewall is possible.
4An example of a grid using stepwise approximation of an Inclined boundary
-
7/28/2019 Grid Postprocessing Cfd
5/35
Choice of grid, contd Complex geometries (Overlapping Chimera grid)
1. Definition: Use of a set of grids to cover irregular solutiondomains
. van ages:
(1). Reduce substantially the time and efforts to generate a grid,especially for 3D configurations with increasing geometric
(2). It allows without additional difficulty calculation of flows
around moving bodies.
(1). The programming and coupling of the grids can be
complicated
.(3). Interpolation process may introduce errors or convergence
problems if the solution exhibits strong variation near the
5
.
-
7/28/2019 Grid Postprocessing Cfd
6/35
Choice of grid, contd
-
6
erent gr str ut on approac es
-
7/28/2019 Grid Postprocessing Cfd
7/35
Choice of grid, contd
7
-
7/28/2019 Grid Postprocessing Cfd
8/35
Choice of grid, contd Complex geometries (Boundary-Fitted Non-Orthogonal Grids)1. Types:
(1). Structured
(2). Block-structured
(3). Unstructured
2.Advantages:
(1). Can be adapted to any geometry
(2). Boundary conditions are easy to apply
(3). Grid spacing can be made smaller in regions of strong variable
variation.
3. Disadvantages:
1 . The transformed e uations contain more terms thereb
increasing both the difficulty of programming and the cost ofsolving the equations
(2). The grid non-orthogonality may cause unphysical solutions.
8
-
7/28/2019 Grid Postprocessing Cfd
9/35
Choice of grid, contd Complex geometries (Boundary-Fitted Non-Orthogonal Grids)
structuredBlock-structured
With matchin interface
Unstructured
Block-structured
Without matching interface
9
-
7/28/2019 Grid Postprocessing Cfd
10/35
Grid generation Conformal mapping: based on complex variable theory, which islimited to two dimensions.
Algebraic methods:. , ,
functions2. 2D: Orthogonal one-dimensional transformation, normalizing
,3. 3D: Stacked two-dimensional transformations, superelliptical
boundaries Differential e uation methods:
Step 1: Determine the grid point distribution on the boundariesof the physical space.
Step 2:Assume the interior grid point is specified by a differential
equa on a sa s es e gr po n s r u ons spec e onthe boundaries and yields an acceptable interior grid pointdistribution.
Commercial software Grid en Gambit etc.
10
-
7/28/2019 Grid Postprocessing Cfd
11/35
Grid generation (examples)
Orthogonal one-dimensional
transformation
11
Superelliptical transformations: (a)
symmetric; (b) centerbody; (c) asymmetric
-
7/28/2019 Grid Postprocessing Cfd
12/35
Grid generation (commercial software, gridgen)
Commercial software GRIDGEN will be used to illustrate
typical grid generation procedure
12
-
7/28/2019 Grid Postprocessing Cfd
13/35
Grid generation (Gridgen, 2D pipe)-
Creation of connectors: connectorscreate2 pointsconnectorsinput x,y,z of the two pointsDone.
Dimension of connectors: .
(0,0.5) (1,0.5) Repeat the procedure to create C2, C3,and C4
C2
C3
C4
(0,0) (1,0)C1
13
-
7/28/2019 Grid Postprocessing Cfd
14/35
Grid generation (Gridgen, 2D pipe, contd)
edgesSpecify edges one by oneDone.
Redistribution of grid points: Boundary layer requires grid refinementnear the wall surface. select connectors (C2, mo y re str ute gr spac ng start+en w t
distribution function For turbulent flow, the first grid spacing near the wall, i.e. matching
oint could have different values when different turbulent modelsapplied (near wall or wall function).
r may e use or am nar ow Grid may be used for turbulent flow
14
-
7/28/2019 Grid Postprocessing Cfd
15/35
Grid generation (3D NACA12 foil)-
attack Creation of geometry: unlike the pipe, we have to import the database
for NACA12 into Gridgen and create connectors based on that (onlya o e geome ry s ape was mpor e ue o symme ry .
inputdatabaseimport the geometry dataconnectorcreateon DB entitiesdelete database
, ,
Half of airfoil surface
C2
C3
Half of airfoil surface
15
-
7/28/2019 Grid Postprocessing Cfd
16/35
Grid generation (3D NACA12 airfoil, contd) Redimensions of the four connectors and create domain Redistribute the grid distribution for all connectors. Especially
refine the grid near the airfoil surface and the leading andtrai ing e ges
16
-
7/28/2019 Grid Postprocessing Cfd
17/35
Grid generation (3D NACA12 airfoil, contd)
Duplicate the other half of the domain: uplicate the other half of the domain:respect to y=0Done.
Rotate the whole domain with angle of attack 60 degrees:domainmodifyrotateusing z-principle axisenter rotationangle:-60Done.
17
-
7/28/2019 Grid Postprocessing Cfd
18/35
Grid generation (3D NACA12 airfoil, contd)
translate distance and directionRun NDone. Split the 3D block to be four blocks:blockmodifysplitin
direction =?Done. pec y oun ary con ons an expor r an .
Block 1Block 1
Block 2
Block 4Block 2
Block 4
Block 3 Block 3
18
3D before split After split (2D view) After split (3D view)
-
7/28/2019 Grid Postprocessing Cfd
19/35
Systematic grid generation for CFD UA CFD UA analysis requires a series of meshes with uniform grid
refinement ratio, usually start from the fine mesh to generate coarsergrids.
, . .,is input into the tool and a series of three, 1D interpolation isperformed to yield a medium grid block with the desired non-integergrid refinement ratio.
n erpo a on s e same or a ree rec ons.Consider 1D line segment with and
points for the fine and medium grids, respectively.
1N ( )2 11 1 / GN N r= +
step 2: compute the medium grid distribution:
11 1 1i i ix x x
=
12 2 2i i ix x x
= +
where from the first step is interpolated at location
step 3: The medium grid distribution is scaled so that the fine and
2 1i iG
1ix 2ix
2ix
19
medium grid line segments are the same (i.e., )
step4: The procedure is repeated until it converges
2 12 1N N
x x=
-
7/28/2019 Grid Postprocessing Cfd
20/35
Post-Processing ,
factor, and uncertainties (for details, CFD Lecture 1, introductionto CFD).
MPI functions required to combine data from different blocks ifpara e compu a on use
Calculation of derived variables (vorticity, shear stress) Calculation of integral variables (forces, lift/drag coefficients)
,spectra
Visualization1. XY lots time/iterative histor of residuals and forces, wave
elevation)2. 2D contour plots (pressure, velocity, vorticity, eddy viscosity)3. 2D velocity vectors
4. 3D Iso-surface plots (pressure, vorticity magnitude, Q criterion)5. Streamlines, Pathlines, streaklines6. Animations
20
er ec n ques: as our er rans orm , ase averag ng
-
7/28/2019 Grid Postprocessing Cfd
21/35
Post-Processing (visualization, XY plots)
NACA12 with 60o angle of attack(CFDSHIP-IOWA, DES)
Wave profile of surface-piercing
ACA24, Re=1.52e6, Fr=0.37
21
(CFDSHIP-IOWA, DES)
-
7/28/2019 Grid Postprocessing Cfd
22/35
Post-Processing (visualization, Tecplot)
Different colors illustrate different blocks (6)
22
Re=10^5, DES, NACA12 with angle of attack 60 degrees
-
7/28/2019 Grid Postprocessing Cfd
23/35
Post-Processing (NACA12, 2D contour plots, vorticity)
e ne an compu e new var a e: a a er pec yequationsvorticity in x,y plane: v10computeOK.
23
-
7/28/2019 Grid Postprocessing Cfd
24/35
Post-Processing (NACA12, 2D contour plot)
x rac s ce rom geome ry: a a x rac cefrom planez=0.5extract
24
-
7/28/2019 Grid Postprocessing Cfd
25/35
Post-Processing (NACA12, 2D contour plots)
2D contour plots on z=0.5 plane (vorticity and eddyviscosity)
Vorticity z Eddy viscosity
25
-
7/28/2019 Grid Postprocessing Cfd
26/35
Post-Processing (NACA12, 2D contour plots)
2D contour plots on z=0.5 plane (pressure andstreamwise velocity)
Pressure Streamwise velocity
26
-
7/28/2019 Grid Postprocessing Cfd
27/35
Post-Processing (2D velocity vectors)
2D velocity vectors on z=0.5 plane: turn off contourand activate vector, specify the vector variables.
Zoom in
27
-
7/28/2019 Grid Postprocessing Cfd
28/35
Post-Processing (3D Iso-surface plots, contd)
3D Iso-surface plots: pressure, p=constant 3D Iso-surface plots: vorticity magnitude
3D Iso-surface plots: 2 criterion
2 2 2
x y z = + +
Second eigenvalue of
3D Iso-surface plots: Q criterion
21
2p
( )ijijijij SSQ =2
12,, ijjiij uu =
2,, ijjiij uuS +=
28
-
7/28/2019 Grid Postprocessing Cfd
29/35
Post-Processing (3D Iso-surface plots)
3D Iso-surface plots: used to define the coherent vortical structures,including pressure, voriticity magnitude, Q criterion, 2, etc.
29
Iso-surface of vorticity magnitude
-
7/28/2019 Grid Postprocessing Cfd
30/35
Post-Processing (streamlines)
ream nes :
30
Streaklines and pathlines (not shown here)
-
7/28/2019 Grid Postprocessing Cfd
31/35
Post-Processing (Animations)
Animations (3D): animations can be created by saving CFDsolutions with or without skipping certain number of time stepsand playing the saved frames in a continuous sequence.
Animations are important tools to study time-dependent
developments of vortical/turbulent structures and their interactions
Q=0.4
31
-
7/28/2019 Grid Postprocessing Cfd
32/35
Other Post-Processing techniques
Fast Fourier Transform
1. A signal can be viewed from two different standpoints:
2. The time domain is the trace on an signal (forces,velocity, pressure, etc.) where the vertical deflection is the
,variable
3. The frequency domain is like the trace on a spectrum,
the vertical deflection is the signals amplitude at thatfrequency.
.
Fourier Transform
Phase averaging (next two slides)
32
-
7/28/2019 Grid Postprocessing Cfd
33/35
Other Post-Processing techniques (contd)
Phase averagingAssumption: the signal should have a coherent dominantfrequency. 1. a filter is first used to smooth the data and remove the high
frequency noise that can cause errors in determining the peaks.2. once the number of peaks determined, zero phase value
s ass gne a eac max mum va ue.3. Phase averaging is implemented using the triple decomposition.
( ) ( ) ( ) ( ) ( ) ( )' '
z t z t z t z t z t z t= + + = + 1N
( ) ( )0
lim 1 Tz t z t dt
T T=
( ) ( )
0n
z t z t nN
=
= +
random fluctuating component
organ ze osc at ng component
( )'z t( )z t mean component z t
33
s e me per o o e om nan requency
is the phase average associated with the coherent structures
( )z t
-
7/28/2019 Grid Postprocessing Cfd
34/35
Other Post-Processing techniques (contd)
FFT of wave elevation
time histories at one point
Original, phase averaged,and random fluctuations
of the wave elevation at
34
one point
-
7/28/2019 Grid Postprocessing Cfd
35/35
References and books
User Manual for GridGen User Manual for Tecplot Numerical recipes:
http://www.library.cornell.edu/nr/
Averaging of Time Resolved PIV measurementdata, Measurement of Science and Technology,
Volume 12, 2001, pp. 655-662. Joe D. Hoffman, Numerical Methods for
Engineers and Scientists, McGraw-Hill, Inc. 1992. . Dubief and F. Delca re On Coherent-vortex
Identification in Turbulence, Journal ofTurbulence, Vol. 1, 2000, pp. 1-20.
35