Supplement Postprocessing Sampling

Supplement
More on sampling
1
Sampling on point clouds, lines
and surfaces
2
Supplement – More on sampling
• OpenFOAM® provides the postProcess utility to sample field data for plotting.
• The sampling locations are specified in the dictionary sampleDict located in the case
system directory.
• You can give any name to the input dictionary, hereafter we are going to name it sampleDict.
• During the sampling, inside the case directory, a new directory named postProcessing will
be created. In this directory, the sampled values are stored.
• This utility can sample points, lines, and surfaces.
• Data can be written in many formats, including well-known plotting packages such as:
grace/xmgr, gnuplot and jPlot.
• The sampling can be executed by running the utility postProcess in the case directory and
according to the application syntax.
• A final word, this utility does not do the sampling while the solver is running. It does the
sampling after you finish the simulation.
3
• Let us do some sampling.

• For this we will use the 3D pipe case, which you will
find in the directory:
$PTOFC/advanced_postprocessing/pipe/
• In the case directory, you will find the README.FIRST file. In this file, you will find the general instructions of
how to run the case. In this file, you might also find some additional comments.
• You will also find a few additional files (or scripts) with the extension .sh, namely, run_all.sh,
run_mesh.sh, run_sampling.sh, run_solver.sh, and so on. These files can be used to run the case
automatically by typing in the terminal, for example, sh run_solver.
• We highly recommend to open the README.FIRST file and type the commands in the terminal, in this way
you will get used with the command line interface and OpenFOAM® commands.
• If you are already comfortable with OpenFOAM®, use the automatic scripts to run the cases.
4
Laminar flow in a straight pipe – Re = 600
Mesh and domain

5
Velocity magnitude at the outlet Pressure contours at the wall
6
Point and lines where we want to sample
7
Surface type and location Coarse and fine surfaces
8
We can also sample in arbitrary surface
9
What are we going to do?
• We will simulate a laminar flow in a straight pipe (Re = 600).
• We will use this case to introduce the sampling utility postProcess.
• We will introduce the utility topoSet used to do topological modifications on the mesh.
• We will use this utility to run functionObjects a-posteriori.
• We will compare the numerical solution with the analytical solution.
• To find the numerical solution we will use the solver pisoFoam.
• After finding the numerical solution we will do some sampling.
• At the end, we will do some plotting (using gnuplot or Python) and scientific visualization.
10
Running the case
• Let us run the simulation and do some sampling. In the terminal type:
1. $> foamCleanTutorials
2. $> blockMesh
3. $> checkMesh
4. $> topoSet
5. $> pisoFoam | tee log.solver
6. $> postProcess -func sampleDict1 -latestTime
7. $> postProcess -func sampleDict2 -latestTime
8. $> gnuplot gnuplot/gnuplot_script
9. $> paraFoam
• Do not erase the solution, we are going to use it in the next sections.
11
Running the case
• In step 4 we use the utility topoSet to do mesh topological manipulation. This utility will read
the dictionary topoSetDict located in the system directory. Later on, we will talk about what
are we doing in this step.
• In step 5 we run the simulation and save the log file.
• In step 6 we use the postProcess utility. By using the option –func we specify to do the
sampling according to the dictionary system/sampleDict1. We sample the latest saved
solution.
• In step 7 we use the postProcess utility. By using the option –func we specify to do the
sampling according to the read dictionary system/sampleDict2. We sample the latest saved
solution.
• In step 8 we use the gnuplot script gnuplot/gnuplot_script to plot the sampled fields.
Feel free to take a look at the script and to reuse it.
• Finally, in step 9 we visualize the solution.
12
The sampleDict dictionary
• Let us visit the sampleDict dictionaries.
• This dictionary is located in the directory system.
• The sampleDict file contains several entries to be set according to the user needs.
• You can set the following entries,
• The choice of the interpolationScheme.
• The format of the line data output.
• The format of the surface data output.
• The fields to be sample.
• The sub-dictionaries that controls each sampling operation.
• In these sub-dictionaries you can set the name, type and geometrical information of
the sampling operation.
• In this case, in the dictionary sampleDict1 we are sampling points and lines, and in the
dictionary sampleDict2 we are sampling surfaces.
13
The sampleDict1 dictionary
17 type sets; Sample sets (points and lines).
Format of the output file, raw format is a generic format that can be
19 setFormat raw;
read by many applications. The file is human readable (ascii
format).
21 interpolationScheme cellPointFace; Interpolation method at the solution level (location of the

interpolation points).
24 fields
25 (
26 U
27 p Fields to sample.
28 );
30 sets
31 (
33 s1 Name of the output file
34 {
35 type lineCellFace; Sample method from the solution to the line.
40 axis z;
41 start (0 0 0); Location of the sample line. We define start and end point, and the
42 end (0 0 0.22);
43 }
axis of the sampling.
44
45 s2 Name of the output file
46 {
47 type lineCellFace; Sample method from the solution to the line.
49 axis x;
50 start (-0.002 -0.002 0.2);
51 end ( 0.002 0.002 0.2);
Location of the sample line. We define start and end point, and the
52 } axis of the sampling.
53
Note:
Use the banana method to know all the options
available. 14
54
55 somePoints Name of the output file
56 {
57 type points; Sample a cloud of points
58 ordered true;
59 axis xyz;
61 points Location of the sample points.

62 (
63 (0 0 0.05)
64 (0 0 0.1)
65 (0 0 0.15)
66 (0 0 0.2) The sampled information is always saved in the directory
67 );
68 }
postProcessing/sampleDict1
71 );
As we are sampling the latest solution (0.1), the sampled data
will be located in the directory:
postProcessing/sampleDict1/0.1
The files s1_p.xy, s2_p.xy, s1_U.xy, s2_U.xy,

somePoints_p.xy, and somePoints_U.xy located in the
directory postProcessing/sampleDict1/0.1 contain the
sampled data. Feel free to open the output files using your
favorite text editor.
15
17 type surfaces; Sample surfaces.
Format of the output file, raw format is a generic format that

19 surfaceFormat raw; can be read by many applications. The file is human readable
(ascii format).
21 interpolationScheme cell; Interpolation method at the solution level (location of the

22
26 fields
interpolation points).
27 (
28 U
29 p
Fields to sample.
30 );
31
32 surfaces
33 (
34 surf1 Name of the object and output file
35 {
36 type plane;
37 planeType pointAndNormal; Surface sampling method. In this case we are using an
38 pointAndNormalDict infinite plane.
39 {
40 basePoint (0 0 0.1);
41 normalVector (0 0 1);
42 }
43 }
44
46 {
47 type cuttingPlane;
48 planeType pointAndNormal;
49 pointAndNormalDict
Surface sampling method. In this case we are using a cutting
50 { plane. The interpolate option means that we interpolate the
51 basePoint (0 0 0.1); cell centered values to the surface triangulation.
53 }
54
55 interpolate true;
56 } 16
57
59 {
60 type triSurfaceMesh; Surface sampling method. In this case we are using a STL
61 surface surface2.stl;
62 source cells; file, the file is always located in the directory
63 constant/triSurface.
65 }
66
68 {
69 type triSurfaceMesh; Surface sampling method. In this case we are using a STL
71 source insideCells; file, the file is always located in the directory
72 constant/triSurface.
73 interpolate false;
74 }
75 ); The sampled information is always saved in the directory
postProcessing/sampleDict2
As we are sampling the latest solution (0.1), the sampled data
will be located in the directory:
postProcessing/sampleDict2/0.1
The files p_surf1.raw, p_surf2.raw, p_surf3.raw,
p_surf4.raw, U_surf1.raw, U_surf2.raw,
U_surf3.raw, and U_surf4.raw, located in the directory
postProcessing/sampleDict2/0.1 contain the sampled
data. Feel free to open the output files using your favorite text
editor.
17
The output files
• The output format of the point sampling (cloud) is as follows:
Scalars
#POINT_COORDINATES (X Y Z) SCALAR_VALUE
0 0 0.05 13.310995
0 0 0.1 19.293817
…
Vectors
#POINT_COORDINATES (X Y Z) VECTOR_COMPONENTS (X Y Z)
0 0 0.05 0 0 2.807395
0 0 0.1 0 0 2.826176
…
18
The output files
• The output format of the line sampling is as follows:
Scalars
#AXIS_COORDINATE SCALAR_VALUE
0 18.594038
0.0015 18.249091
…
Vectors
#AXIS_COORDINATE VECTOR_COMPONENTS (X Y Z)
0 0 0 1.6152966
0.0015 0 0 1.8067536
…
19
The output files
• The output format of the surface sampling is as follows:
Scalars
#POINT_COORDINATES (X Y Z) SCALAR_VALUE
0 0 0.05 13.310995
0 0 0.1 19.293817
…
Vectors
#POINT_COORDINATES (X Y Z) VECTOR_COMPONENTS (X Y Z)
0 0 0.05 0 0 2.807395
0 0 0.1 0 0 2.826176
…
20
• To plot the sampled data using gnuplot we can proceed as follows:
1. gnuplot> set title 'Radial velocity at z=0.2 - Time 0.1 s'
2. gnuplot> set xlabel 'Diameter'
3. gnuplot> set ylabel 'U_z'
4. gnuplot> set grid
5. gnuplot> plot [][] 'postProcessing/sampleDict1/0.1/s2_U.xy' u 1:4 w p pt 7 title "Numerical

solution", 2.8265544*(1-x**2/(0.0010606602 )**2) title "Analytical solution"
21
• To plot the sampled data using gnuplot we can proceed as follows:
1. gnuplot> set title 'Velocity profile at plane located at z = 0.1 - Time 0.1 s'
2. gnuplot> set xlabel 'X'
3. gnuplot> set ylabel 'Y'
4. gnuplot> set zlabel 'U_z'
5. gnuplot> set grid
6. gnuplot> splot [][][] 'postProcessing/sampleDict/0.1/U_surf1.raw' u 1:2:6 pt 7 ps 0.5
22
Creating faceSet and zoneSet
• To create sets and zones we use the utility topoSet.
• This utility reads the dictionary topoSetDict located in the system directory.
• faceSet/cellSet and faceZone/cellZone can be used to do modifications to the mesh or to
apply source terms.
• We can only do sampling operations on zoneSets made of a set of faces and/or cells,
therefore, if we have a faceSet/cellSet we need to convert it to a faceZone/cellZone.
• Creating an internal faceZone is particularly important if we are interested in computing the
mass flow in an internal surface, as sampledSurface does not work with surfaceScalarFields;
therefore, we need to use a faceZone.
• Alternatively, we can compute the mass flow in paraFoam/paraView or we can use the
areaNormaIntegrate operation on the sampleSurface.
• Let us create an internal faceZone and a cellZone and let us compute the mass flow and do
some sampling on these sets.
23
The topoSetDict dictionary
17 actions
18 (
19
20 { • In this step we are creating a new faceSet from a boxToFace
21 action new; source.
22 name internalfaces;
23 type faceSet; • The name of the new faceSet is internalFaces.
24 • The box source encloses the faces we want to tag.
25 source boxToFace;
26 sourceInfo
• We can visualize this set in paraFoam.
27 { • Remember, we can not sample on a faceSet, we need to
28 box (-1 -1 0.098) (1 1 0.1); convert it to a faceZone.
29 }
30 }
31
32 {
33 action new;
34 name internalfacepatch;
35 type faceZoneSet; • In this step we convert the faceSet internalFaces, to a
36 faceZone.
37 source setToFaceZone;
38 sourceInfo
• The name of new faceZone is internalfacepatch.
39 { • At this point, we can use this faceZone to do sampling.
40 faceSet internalfaces; • We can visualize this zone in paraFoam.
41 }
42 }
43
Note:
available.
24
The topoSetDict dictionary
44 {
45 action new; • In this step we are creating a new cellSet from a boxToCell
46 name internalcells; source.
47 type cellSet;
48 • The name of the new cellSet is internalCells.
49 source boxToCell; • The box source encloses the cells we want to tag.
50 sourceInfo
51 {
52 box (-1 -1 0.096) (1 1 0.1); • Remember, we can not sample on a cellSet, we need to
53 } convert it to a cellZone.
54 }
55
56 {
57 action new;
58 name internalcells;
59 type cellZoneSet;
60 • In this step we convert the cellSet internalcells, to a cellZone.
61 source setToCellZone;
62 sourceInfo • The name of new cellZone is internalcells.
63 { • At this point, we can use this cellZone to do sampling.
64 set internalcells;
65 }
66 }
67
68 );
Note:
available.
25
• Let us create an internal faceZone and a cellZone and let us compute the mass flow and do
some sampling on these sets.
1. $> topoSet
2. $> pisoFoam -postProcess -dict system/functionobject3 –latestTime
3. $> pisoFoam -postProcess -dict system/functionobject4 -latestTime
• In step 1 we use the utility topoSet to create the new faceZone and cellZone.
• In step 2 we run a functionObject a-posteriori. In this functionObject we compute:
• Mass flow in a sampledSurface.
• Mass flow in a faceZone.
• In step 3 we run a functionObject a-posteriori. In this functionObject we compute:
• Volume integral in a cellZone.
26
Visualizing the newly created sets and zones
Check the option Include Sets cellZone - In white

to visualize the sets
Check the option Include Zones

to visualize the zones
faceZone - In red
Select the sets you want to visualize

(faceSet, faceZone, cellSet or cellZone)
27
The functionobject3 dictionary
17 functions
18 {
19
21 surface1_massflow
22 {
23 type surfaceRegion;
24 functionObjectLibs ("libfieldFunctionObjects.so");
25 enabled true;
26
27 writeControl timeStep;
28 writeInterval 1;
29
30 log true;
31 writeFields false;
32
33 regionType sampledSurface;
34 Name dummy;
35
36 sampledSurfaceDict
37 {
38 type sampledTriSurfaceMesh;
40 source cells;
41 interpolate false;
42 }
43
54
55
operation
fields
areaNormalIntegrate;
• Compute mass flow using areaNormalIntegrate
56 ( operation with the field U.
57 U
58
59
);
• Using the operation sum with the field phi, will not
60 } work because phi is a surfaceScalarField.
28
65 {
68 enabled true;
69
71 writeInterval 1;
72
73 log true;
75
76 regionType faceZone;
77 name internalfacepatch;
78
79 operation sum;
80 fields
81 (
82 phi
83 );
84
85 }
• Compute mass flow using sum operation with the

field phi.
• In this case it works because we are sampling on
a faceZone.
• The faceZone are internal faces of the mesh.
29
91 {
94 enabled true;
95
97 writeInterval 1;
98
99 log true;
101
103 name dummy;
104
106 {
109 source cells;
In this case we are using a finer surface therefore we
111 } enable the option interpolate.
112
113 operation areaNormalIntegrate;
114 fields
115
116
(
U
Compute mass flow using areaNormalIntegrate
117 ); operation with the field U.
118
119 }
30
124 {
127 enabled true;
128
130 writeInterval 1;
131
132 log true;
134
136 name dummy;
137
139 {
142 source insideCells;
144 }
145
146 operation areaNormalIntegrate;
147 fields
148 (
149 U Compute mass flow using areaNormalIntegrate
150
151
);
operation with the field U.
152 }
31
156 plane1_massflow
157 {
160 enabled true;
161
165
166 log true;
168
170 name dummy;
171
173 {
174 type plane;
175
176
planeType pointAndNormal;
pointAndNormalDict
We sample in a plane.
177 {
178 basePoint (0 0 0.1);
180 }
181 }
182
184 operation areaNormalIntegrate; Compute mass flow using areaNormalIntegrate
185
186
fields
(
operation with the field U.
187 //phi
188 U
189 );
190 }
191
193 }
32
17 functions
18 {
19
22 cells_fo1
23 {
24 type volRegion;
26 enabled true;
27
29 writeInterval 1;
30
31 log true;
32
34
35 regionType cellZone;
36 name internalcells;
37
38 operation volIntegrate;
39
40 fields
41 (
42 U Compute volIntegrate of the field U.
43 );
44
45 }
97
98 }
99
Remember, we can do cellSource and faceSource

sampling only on cellZone/faceZone.
33
Exercises
• Where is located the source code of the utility postProcess?
• Try to do the sampling in parallel? Does it run? What about the output file? Is it the same?
• How many options are there available to do sampling in a line?
• How many options are there available to do sampling in a surface?
• Compute the mass flow at the inlet and outlet patches using the operation sum and areaNormalIntegrate. Do
you get the same output?
• Compute the mass flow at the inlet and outlet patches using paraFoam/ParaView and compare with the output
of the postProcess utility. Do you get the same results?
• Do point, line, and surface sampling using paraFoam/ParaView and compare with the output of the
postProcess utility. Do you get the same results?
34
Sampling on points (probing)
35
• OpenFOAM® provides the postProcess utility to probe field data for plotting.
• The probing locations are specified in the dictionary probesDict located in the case system
directory.
• You can give any name to the input dictionary, hereafter we are going to name it probesDict.
• During the probing, inside the case directory a new directory named postProcessing, will be
created. In this directory, the sampled values are stored.
• This utility can sample only points.
• Data can be written in many formats, including well-known plotting packages such as:
grace/xmgr, gnuplot and jPlot.
• The probing can be executed by running the utility postProcess in the case directory and
according to the application syntax.
• A final word, this utility does not do the sampling while the solver is running. It does the
sampling after you finish the simulation.
36
Probes location
37
• We hope you did not erase the previous solution because we will use it to play around with the
probeLocations. In the terminal type:
1. $> postProcess -func probesDict
• This will probe all the saved solutions at the specified locations. It will save time vs. quantity of
interest.
• The sampled information is always saved in the directory (same name as the input file),
• postProcessing/probesDict
• As we started to sample from time 0, the sample data is saved in the directory
• postProcessing/probesDict/0
• The files p, and U, located in the directory postProcessing/probesDict/0 contain the

sampled data. Feel free to open them using your favorite text editor.
38
The probesDict dictionary
17 type probes;
18
20 fields
21 (
22 p
23 U Fields to sample. The output files will have the name of the fields.
24 );
25
27 probeLocations
28 (
29 (0 0 0.025)
30 (0 0 0.050)
31 (0 0 0.075)
32 (0 0 0.10) Probe locations.
33 (0 0 0.125)
34 (0 0 0.150)
35 (0 0 0.175)
36 (0 0 0.20)
37 );
39
The output files
• The output format of the probing is as follows:
Scalars
# Probe 0 (0 0 0.025)
# Probe 1 (0 0 0.05)
# Probe 2 (0 0 0.075)
# Probe 3 (0 0 0.1)
# Probe 0 1 2 3
# Time
0 0 0 0 0
0.005 19.1928 16.9497 14.2011 11.7580
0.01 16.6152 14.5294 12.1733 10.0789
…
…
…
40
The output files
• The output format of the probing is as follows:
Vectors
# Probe 0 (0 0 0.025)
# Probe 1 (0 0 0.05)
# Probe 2 (0 0 0.075)
# Probe 3 (0 0 0.1)
# Probe 0 1 2 3
# Time
0 (0 0 0) (0 0 0) (0 0 0) (0 0 0)
0.005 (0 0 2.1927) (0 0 2.1927) (0 0 2.1927) (0 0 2.1927)
0.01 (0 0 2.5334) (0 0 2.5334) (0 0 2.5334) (0 0 2.5334)
…
…
…
41
Exercises
• Try to do the sampling in parallel? Does it run? What about the output file? Is it the same?
• Do the same sampling using paraFoam/paraview and compare with the output of the postProcess utility. Do
you get the same results?
• Compute the descriptive statistics of each column of the output files using gnuplot. Be careful with the
parentheses of the vector files.
(Hint: you can use sed within gnuplot)
42
On-the-fly postprocessing
functionObjects and the
postProcess utility
43
• It is possible to perform data extraction/manipulation operations while the simulation is running
by using the functionObjects.
• functionObjects are small pieces of code executed at a regular interval without explicitly being
linked to the application.
• When using functionObjects, files of sampled data can be written for plotting and post
processing.
• functionObjects are specified in the controlDict dictionary and executed every time step or
pre-defined intervals.
• All functionObjects are runtime modifiable.
• All the information related to the functionObject is saved in the directory postProcessing or in
the solution directory.
• It is also possible to execute functionObject after simulation is over, we will call this running
functionObject a-posteriori.
44
• Let us do some on-the-fly postprocessing.

• Let us revisit the pipe case. Go to the directory:
$PTOFC/advanced_postprocessing/pipe
• In the case directory, you will find the README.FIRST file. In this file, you will find the general instructions of
how to run the case. In this file, you might also find some additional comments.
• You will also find a few additional files (or scripts) with the extension .sh, namely, run_all.sh,
run_mesh.sh, run_sampling.sh, run_solver.sh, and so on. These files can be used to run the case
automatically by typing in the terminal, for example, sh run_solver.
• We highly recommend to open the README.FIRST file and type the commands in the terminal, in this way
you will get used with the command line interface and OpenFOAM® commands.
• If you are already comfortable with OpenFOAM®, use the automatic scripts to run the cases.
45
Mesh and domain
46
What are we going to do?
• We will use this case to introduce functionObjects.

• We will use the utility postprocess to run functionObjects a-posteriori.
• We will also use the utility postprocess to compute some quantities in patches.
• We will compare the numerical solution with the analytical solution.
• We will do data cleaning and data analytics using shell scripting.
• To find the numerical solution we will use the solver pisoFoam.
• After finding the numerical solution we will do some sampling.

• At the end, we will do some plotting (using gnuplot or Python) and scientific visualization.
47
Running the case
• Let us run the simulation. In the terminal type:
1. $> foamCleanTutorials
2. $> blockMesh
3. $> checkMesh
4. $> topoSet
5. $> pisoFoam | tee log.solver
7. $> paraFoam
• In step 4 we use the utility topoSet to do mesh topological manipulation. This utility will read
the dictionary topoSetDict located in the system directory. Later on, we will talk about what
are we doing in this step.
• In step 5 we run the simulation and save the log file. In step 6 we use
pyFoamPlotWatcher.py to plot the residuals on-the-fly. As the job is running in background,
we can launch this utility in the same terminal tab.
• Finally, in step 7 we visualize the solution.
48
• Let us explore the case directory.
• You will notice that we now have a new directory named postProcessing.
• Inside this directory, you will find many subdirectories pointing to the functionObject used.
• By the way, we are saving a large amount of information.
• This is typical of unsteady simulations, and sometimes it can be too daunting to post-process
and analyze the data.
• To ease the pain of doing data analytics and post-processing, you can use shell scripting or
Python scripting.
• Hereafter, we are going to address how to use shell scripting.
49
• Inside the directory postProcessing, you will find the following sub-directories:
• cells_fo1 • pressureDrop
• cells_fo2 • pressureDrop.region1
• field_fo4 • pressureDrop.region2
• field_fo4.region1 • probesDict
• field_fo4.region2 • probes_fo1
• forces_object • probes_online
• inlet_average • sampleDict1
• inlet_massflow • sampleDict2
• inlet_massflow_posteriori • sets_fo2
• innerpatch_massflow • sets_online
• minmaxdomain • surface1_massflow
• outlet_areaNormalIntegrate • surface2_massflow
• outlet_massflow • surface3_massflow
• outlet_massflow_posteriori • surface4_massflow
• outlet_max • surfaces_fo3
• plane1_massflow
• Inside each sub-directory you will find the time directory 0, this is the time directory from which
we started to sample using functionObject. Inside this directory you will find the sampled
data.
• If you start to sample from time 50, you will find the time directory 50.
• If you stop the simulation and restart it, let us say from time 10, you will find the time directories
0 and 10.
• For line and surface sampling, you will find all the time directories corresponding to the saving
frequency. Inside each directory you will find the sampled data.
• Let us take a look at the general organization of a functionObject.
50
• The functionObject entry in the controlDict dictionary, contains at least
the following information:
function_object_name Name of functionObject
type function_object_to_use; functionObject to use
functionObjectLibs ("function_object_library.so"); Library to use
enabled true; Turn on/off functionObject
log true; Show on screen the output

of the functionObject
writeControl outputTime;
timeStart 0; Output frequency
timeEnd 20;
//... Keywords and sub-

//functionObject // dictionaries specific to the
//keywords and sub-dictionaries // functionObject
//...
51
• There are many functionObjects implemented in OpenFOAM®, and sometimes is not very
straightforward how to use a specific functionObject.
• Also, functionObjects can have many options and some limitations.
• Our best advice is to read the doxygen documentation or the source code to learn how to use
functionObjects.
• The source code of the functionObjects is located in the directory:
$WM_PROJECT_DIR/src/postProcessing/functionObjects
• Here after we are going to study a few commonly used functionObjects.
52
The controlDict dictionary
51 functions • Let us take a look at the bottom of the controlDict

52 { dictionary file.
name_of_the_functionObject_dictionary • Here we define the functionObjects, which are functions
{
Dictionary with the functionObject entries
that will do a computation while the simulation is running.
}
• In this case, we define the functionObjects in the sub-
dictionary functions (lines 51-115).
113 #include "functionObject0"
114
115 }
• Each functionObject we define, has its own name and its
compulsory keywords and entries.
• Notice that in line 113 we use the directive include to call an
external dictionary with the functionObjects definition.
• If you use the include directive, you will need to update the
controlDict dictionary in order to read any modification
done in the included dictionary files.
• By the way, you can give any name to the input files defined
in line 113.
53
56 minmaxdomain • fieldMinMax functionObject

57 {
58 type fieldMinMax;
59 • This functionObject is used to compute the
60 functionObjectLibs ("libfieldFunctionObjects.so"); minimum and maximum values of the field
61
62 enabled true; variables.
63
64
65
mode component; • The output of this functionObject is saved in
66 writeControl timeStep; ascii format in the file fieldMinMax.dat located
67 writeInterval 1;
68 in the directory
69 log true;
70 postProcessing/minmaxdomain/0
71 fields (p U);
72 }
• Remember, the name of the directory where the output

data is saved is the same as the name of the
functionObject (line 56).
Note:
Use the banana method to know all the options available
for each entry.
54
78 field_averages • fieldAverage functionObject

79 {
80 type fieldAverage;
81 functionObjectLibs ("libfieldFunctionObjects.so"); • This functionObject is used to compute the
82 enabled true; average values of the field variables.
83
84 writeControl outputTime;
85 //writeControl timeStep; • The output of this functionObject is saved in the
86 //writeInterval 100; time solution directories.
87
88 //cleanRestart true;
89
90 timeStart 0.05;
91 timeEnd 0.1;
92
93 fields
94 (
95 U
96 {
97 mean on;
98 prime2Mean on;
99 base time;
100 }
101
102 p
103 {
104 mean on;
105 prime2Mean on;
106 base time;
107 }
108 );
109 }
Note:
for each entry.
55
The functionObject0 dictionary
19 inlet_massflow • faceSource functionObject

20 {
21
22 type surfaceRegion; • This functionObject is used to compute the
23 functionObjectLibs ("libfieldFunctionObjects.so"); mass flow in a boundary patch.
24 enabled true;
25
26 //writeControl outputTime; • In this case, we are sampling the patch inlet.
28 writesInterval 1; • We are using the operation sum with the field phi.
29
30 log true; This is equivalent to compute the mass flow.
31
32 writeFields false; • The output of this functionObject is saved in
33
34 regionType patch; ascii format in the file faceSource.dat located
35
36
name inlet;
in the directory
37 operation sum;
38 fields postProcessing/inlet_massflow/0
39 (
40 phi
41 );
42 }
Note:
for each entry.
56
48 outlet_massflow • faceSource functionObject

49 {
52 enabled true; mass flow in a boundary patch.
53
54 //writeControl outputTime;
55 writeControl timeStep; • In this case, we are sampling the patch outlet.
56 writeInterval 1;
57 • We are using the operation sum with the field phi.
58 log true;
59 This is equivalent to compute the mass flow.
61 • The output of this functionObject is saved in
62 regionTyp patch;
63 Name outlet; ascii format in the file faceSource.dat located
64
65 operation sum;
in the directory
66 fields
67 ( postProcessing/outlet_massflow/0
68 phi
69 );
70 }

Note:
for each entry.
57
76 probes_online • probes functionObject

77 {
78 type probes;
79 functionObjectLibs ("libfieldFunctionObjects.so"); • This functionObject is used to probe field data at
80 enabled true; the given locations.
82
83 probeLocations • The output of this functionObject is saved in
84 ( ascii format in the files p and U located in the
85 (0 0 0)
86 (0 0 0.1) directory
87 (0 0 0.2)
88 ); postProcessing/probes_online/0
89
90 fields
91 (
92 U
93 p • Remember, the name of the directory where the output
94 );
95 data is saved is the same as the name of the
96 }
Note:
for each entry.
58
131 pressureDrop
132 { • fieldValueDelta functionObject
133 type fieldValueDelta;
135 enabled true;
136 difference/average/min/max of two field values.
137 region1
138 { • We are using the operation sum with the field phi.
139 writeFields off;
140 type surfaceRegion; This is equivalent to compute the mass flow.
141 regionType patch;
142 name inlet; • We are using the operation subtract between the
143
144 operation sum; two field values.
145
146 fields • The output of this functionObject is saved in
147 (
148 phi ascii format in the file fieldValueDelta.dat
149
150 }
);
located in the directory
151
152 region2 postProcessing/pressureDrop/0
153 {
154 writeFields off;
156 regionType patch;
157 name outlet; • Remember, the name of the directory where the output
158
159 operation sum;
160 functionObject (line 131).
161 fields
162 (
163 phi
164 );
165 } Note:
166 Use the banana method to know all the options available
167 operation subtract;
168 } for each entry.
59
174 innerpatch_massflow • faceSource functionObject

175 {
178 enabled true; mass flow in an inner patch.
179
181 writeInterval 1; • In this case, we are sampling the faceZone
182
183 log true;
internalfacepatch.
185 • We are using the operation sum with the field phi.
186 regionType faceZone;
187 Name internalfacepatch; • The output of this functionObject is saved in
188
189 operation sum; ascii format in the file faceSource.dat located
190
191
fields
(
in the directory
192 phi
193 ); postProcessing/innerpatch_massflow/0
194
195 }

Note:
for each entry.
60
201 sets_online • sets functionObject

202 {
203 type sets;
204 functionObjectLibs ("libfieldFunctionObjects.so"); • This functionObject is used to sample field data
205 enabled true; in a line.
207
208 interpolationScheme cellPointFace; • The output of this functionObject is saved in
209 setFormat raw; ascii format in the files set1_p.xy and
210
211 sets set1_U.xy located in the time directories inside
212 (
213 the folder
214 set1
215 { postProcessing/sets_online
216 type lineCellFace;
217
218 axis x;
219 start (-0.002 -0.002 0.2);
220 end ( 0.002 0.002 0.2); • Remember, the name of the directory where the output
221
222
}
223 ); functionObject (line 201).
224
225 fields
226 (
227 U
228 p
229 );
230
231 }
Note:
for each entry.
61
237 forces_object • forces functionObject

238 {
239 type forces;
240 functionObjectLibs ("libforces.so"); • This functionObject is used to compute the
241 Enabled true; forces on a patch.
242
243 //writeControl outputTime;
244 writeControl timeStep; • In this case, we are sampling the patch walls.
246 • The output of this functionObject is saved in
247 //// Patches to sample
248 patches ("walls"); ascii format in the file forces.dat located in
249
250 //// Name of fields
the time directories inside the folder
251 pName p;
252 Uname U; postProcessing/forces_object/0
253
254 //// Density
255 rho rhoInf;
256 rhoInf 1.;
257 • Remember, the name of the directory where the output
258 //// Centre of rotation data is saved is the same as the name of the
259 CofR (0 0 0);
260 } functionObject (line 237).
Note:
for each entry.
62
Running functionObjects a-posteriori
• Sometimes, it can happen that you forget to use a functionObject or you want to execute a
functionObject a-posteriori (when the simulation is over).
• The solution to this problem is to use the solver with the option -postProcess.
• This will only compute the functionObject, it will not rerun the simulation.
• For instance, let us say that you forgot to use a given functionObject.
• Open the dictionary controlDict, add the new functionObject, and type in the terminal,
• $> name_of_the_solver -postProcess –dict dictionary_location
• By proceeding in this way you do not need to rerun the simulation, you just compute the new
functionObject.
63
• In the directory system, you will find the following functionObjects dictionaries:
functionObject1 , functionObject2 , functionObject3 , functionObject4,
functionObject5 , functionObject6.
• Try to figure out what we are doing in every functionObject dictionary.
• At this point, let us run each the functionObject a-posteriori. In the terminal type:
1. $> pisoFoam -postProcess -dict system/functionObject1 | tee log_fo1
2. $> pisoFoam -postProcess -dict system/functionObject2
3. $> pisoFoam -postProcess -dict system/functionObject3 –time 0.05:0.1
4. $> pisoFoam -postProcess -dict system/functionObject4 –noZero
5. $> pisoFoam -postProcess -dict system/functionObject5 –latestTime
6. $> pisoFoam -postProcess -dict system/functionObject6 –latestTime
64
• In this step 1, we are reading the dictionary functionObject1 located in the directory
system (the dictionary file can be located anywhere), and we are doing the computation for all
the saved solutions. Notice that we are redirecting the output to a log file ( | tee log_fo1).
• In this step 2, we are reading the dictionary functionObject2 and we are doing the
computation for all saved solutions.
computation for the time range 0.05 to 0.1 (–time 0.05:0.1).
computation for all the saved solutions, except time zero (–noZero).
computation only for he latest save solution (–latestTime).
computation only for he latest save solution (–latestTime).
65
Some shell and awk scripting
• Let us do some shell and awk scripting on the sampled data.
• Let us go to the directory postprocesssing/minmaxdomain/0
• To erase the parentheses in the file fieldMinMax.dat and save the output in the file
out.txt, type in the terminal:
• $> cat fieldMinMax.dat | tr -d "()" > out.txt
• To extract the velocity from the file out.txt, and save the output in the file vel_minmax.txt,
type in the terminal:
• $> awk '0 == NR % 2' out.txt > vel_minmax.txt
• To extract the pressure from the file out.txt, and save the output in the file
pre_minmax.txt, type in the terminal:
• $> awk '1 == NR % 2' out.txt > pre_minmax.txt
66
• To erase the header of the file pre_minmax.txt, type in the terminal:
• $> awk '{if (NR!=1) {print}}' pre_minmax.txt > tmp
• $> mv tmp pre_minmax.txt
• To erase the header of the file vel_minmax.txt, type in the terminal:
• $> awk '{if (NR!=1) {print}}' vel_minmax.txt > tmp
• $> mv tmp vel_minmax.txt
• To compute the mean value of the seventh column of the file pre_minmax.txt (maximum
pressure in this case), type in the terminal:
• $> awk 'NR>=50 && NR { total += $7; count++} END { print
“Mean_value ” total/count}' pre_minmax.txt
We are computing the mean value of column 7 starting from row 50
• To compute the mean value of the seventh column of the file vel_minmax.txt (minimum Z
velocity component in this case), type in the terminal:
• $> awk 'NR>=50 && NR { total += $5; count++} END { print
“Mean_value ” total/count}' vel_minmax.txt
We are computing the mean value of column 7 starting from row 50
67
• In the directory scripts (located in the top level case directory) you will find the following
scripts that will do some data processing automatically,
• script_cleanfile
• script_cleanforce
• script_extractfields
• script_force_coe
• script_forces
• script_meanvalues
• script_minmax
• script_probes
• To run the scripts you need to be inside the scripts directory. In the terminal type,
• cd scripts
• sh script_name
• Feel free to reuse them and adapt these scripts according to your needs. 68
• The script script_force_coe will compute the mean value and standard deviation of the lift
and drag coefficients (in this case they are not computed).
• The script script_forces will extract the force components and saved in a clean file.
• The script script_minmax will extract the minimum and maximum values of the field
variables.
• The script script_probes will extract the information of the probes.
We just showed how to compute the average and standard deviation and do some manipulation of
the information saved in the output files, but you can do many things in an automatic way.
The power of scripting!!!
69
Plotting in gnuplot
• Let us do some plotting using gnuplot. Type in the terminal (you must be inside the scripts
directory):
1. $> sh script_minmax
2. $> gnuplot
3. gnuplot> plot [][] 'vel_minmax.txt' u 1:5 w l title "min v_z",

'' u 1:11 w l title "max v_z"
70
Plotting in gnuplot
• Let us do some plotting using gnuplot. Type in the terminal (you must be inside the scripts
directory):
1. gnuplot> plot [][] '< sed "s/[()]/ /g"

../postProcessing/forces_object/0/forces.dat' u 1:7 w l
• In this step we are using sed inside gnuplot to clean the parentheses. If you do not erase the
parentheses in the input file, gnuplot will complain.
71
Exercises
• Where is located the source code of the functionObjects?
• Try to run in parallel? Do all functionObjects work properly?
• Try to run functionObjects a-posteriori in parallel? Does it work? Do all functionObjects work properly?
• Compute the Courant number using functionObjects.
• Compute the total pressure and velocity gradient using functionObjects (on-the-fly and a-posteriori).
• Sample data (points, lines and surfaces) using functionObjects (a-posteriori).
• Is it possible to do system calls using functionObjects? If so what functionObjects will you use and how do
you use it? Setup a sample case.
• Is it possible to update dictionaries using functionObjects? If so what functionObjects will you use and how
do you use it? Setup a sample case.
• What are the compulsory entries of the functionObjects.
72

Supplement Postprocessing Sampling

Uploaded by

Copyright:

Available Formats

Supplement Postprocessing Sampling

Uploaded by

Document Information

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

Supplement Postprocessing Sampling

Uploaded by

Copyright:

Available Formats

Supplement

• Let us do some sampling.

Mesh and domain

Velocity magnitude at the outlet Pressure contours at the wall

Point and lines where we want to sample

Surface type and location Coarse and fine surfaces

We can also sample in arbitrary surface

21 interpolationScheme cellPointFace; Interpolation method at the solution level (location of the

61 points Location of the sample points.

The files s1_p.xy, s2_p.xy, s1_U.xy, s2_U.xy,

Format of the output file, raw format is a generic format that

21 interpolationScheme cell; Interpolation method at the solution level (location of the

1. gnuplot> set title 'Radial velocity at z=0.2 - Time 0.1 s'

2. gnuplot> set xlabel 'Diameter'

3. gnuplot> set ylabel 'U_z'

4. gnuplot> set grid

5. gnuplot> plot [][] 'postProcessing/sampleDict1/0.1/s2_U.xy' u 1:4 w p pt 7 title "Numerical

2. gnuplot> set xlabel 'X'

3. gnuplot> set ylabel 'Y'

4. gnuplot> set zlabel 'U_z'

5. gnuplot> set grid

6. gnuplot> splot [][][] 'postProcessing/sampleDict/0.1/U_surf1.raw' u 1:2:6 pt 7 ps 0.5

3. $> pisoFoam -postProcess -dict system/functionobject4 -latestTime

Check the option Include Sets cellZone - In white

Check the option Include Zones

Select the sets you want to visualize

• Compute mass flow using sum operation with the

Remember, we can do cellSource and faceSource

1. $> postProcess -func probesDict

• The files p, and U, located in the directory postProcessing/probesDict/0 contain the

• Let us do some on-the-fly postprocessing.

Mesh and domain

• We will use this case to introduce functionObjects.

• After finding the numerical solution we will do some sampling.

function_object_name Name of functionObject

type function_object_to_use; functionObject to use

functionObjectLibs ("function_object_library.so"); Library to use

enabled true; Turn on/off functionObject

log true; Show on screen the output

//... Keywords and sub-

• Here after we are going to study a few commonly used functionObjects.

51 functions • Let us take a look at the bottom of the controlDict

56 minmaxdomain • fieldMinMax functionObject

• Remember, the name of the directory where the output

78 field_averages • fieldAverage functionObject

19 inlet_massflow • faceSource functionObject

48 outlet_massflow • faceSource functionObject

• Remember, the name of the directory where the output

76 probes_online • probes functionObject

174 innerpatch_massflow • faceSource functionObject

• Remember, the name of the directory where the output

201 sets_online • sets functionObject

237 forces_object • forces functionObject

• $> name_of_the_solver -postProcess –dict dictionary_location

1. $> pisoFoam -postProcess -dict system/functionObject1 | tee log_fo1

2. $> pisoFoam -postProcess -dict system/functionObject2

3. $> pisoFoam -postProcess -dict system/functionObject3 –time 0.05:0.1

4. $> pisoFoam -postProcess -dict system/functionObject4 –noZero

5. $> pisoFoam -postProcess -dict system/functionObject5 –latestTime