[cyss38]

Dear all,

The fluidfoam package (https://github.com/fluiddyn/fluidfoam) provides useful Python classes for easily reading and manipulating OpenFoam output files.
You can find some examples here : https://fluidfoam.readthedocs.io/en/...les/index.html

fluidfoam is also used to read 1D simulation outputs and put them in swak4foam format in order to use 1D data for boundary conditions and / or initial fields : https://fluidfoam.readthedocs.io/en/...cessing1d.html. (Today there is no example on this subject in the example gallery... Coming soon)

Best,

Cyrille

PS : fluidfoam is available with pip, conda or via sources from github:
https://github.com/fluiddyn/fluidfoam

[Edgar Alejandro Martínez]

I have an example from a 2D actuator turbine. You can learn how to get the coordinates as well as the desired fields: https://www.youtube.com/watch?v=wFZeFDUQflQ

[cyss38]

Originally Posted by Edgar Alejandro Martínez

I have an example from a 2D actuator turbine. You can learn how to get the coordinates as well as the desired fields: https://www.youtube.com/watch?v=wFZeFDUQflQ

Thanks you very much for this tutorial
You say that the structured option of readMesh does not work in your case.
Can you send me your blockMeshDict file to solve this issue?

The best is to open an issue on the official github repository (but the blockMeshDict may be enough )

[Edgar Alejandro Martínez]

I meant I had to set the option to False. I don't know what "structured mesh" means to fluidfoam. Remember that meshes generated by blockMesh are not exactly structured since every node has an ID. So blockMesh is not structured eventhough it is rectangular.

Here's my blockMeshDict, this is another similar mesh though:

/*--------------------------------*- C++ -*----------------------------------*\
| ========= | |
| \\ / F ield | OpenFOAM: The Open Source CFD Toolbox |
| \\ / O peration | Version: dev |
| \\ / A nd | Web: www.OpenFOAM.org |
| \\/ M anipulation | |
\*---------------------------------------------------------------------------*/
FoamFile
{
version 2.0;
format ascii;
class dictionary;
object blockMeshDict;
}
// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //

vertices
(
( -60 -60 -1)
( 100 -60 -1)
( 100 60 -1)
( -60 60 -1)
( -60 -60 1)
( 100 -60 1)
( 100 60 1)
( -60 60 1)
);

blocks
(
hex (0 1 2 3 4 5 6 7) (240 180 1) simpleGrading (1 1 1)
);

boundary
(
inlet
{
type patch;
faces
(
(0 4 7 3)
);
}

outlet
{
type patch;
faces
(
(2 6 5 1)
);
}

top
{
type patch;
faces
(
(3 7 6 2)
);
}

bottom
{
type patch;
faces
(

(1 5 4 0)
);
}

back
{
type empty;
faces
(
(0 3 2 1)
);
}

front
{
type empty;
faces
(
(6 7 4 5)
);
}
);

// ************************************************** *********************** //

[cyss38]

Ok, I tested it. Indeed in this case there is a problem of precision. I don't know exactly why but there are accuracy problems in the openfoam files (generated by blockMesh).
It's not a problem but when fluidfoam/python looks for the unique values of the center of the cells for a given direction (x for example), there are cases where it doesn't work well (for 2 different y, you can have a very slightly different x when it should be constant).
To avoid this problem, there is the "precision" option in fluidfoam (the default value is 15).

In your case, :
x, y, z = readmesh(sol, True, precision = 14)

One of the interests of using structured=True is not to have to know nx, ny, nz. fluidfoam calculates them and we can easily find them :

nx, ny, nz = np.shape(x)

[FluentStarter]

Late, but I guess it is newer too late to say thanks. Nice and helpful package!