Porous media modelling

[CTrinca]

Hello everyone,

I have just started working on my PhD project and I have never used Code Saturne before. I could hardly be more inexperienced, but I have been browsing this forum for a while and I have seen how lively and helpful this community is, so my task seems a bit less daunting...

My first question is a somewhat theoretical one: I need to include a porous region in my model to simulate the effects of a rod bundle on a gas flow. I understand that this can be done by including the routine usporo in the case run, but I haven't been been able to find any information about how porous media are treated by CS for what concerns the equations for momentum, energy and turbulent quantities. Are they modelled by additional source terms as in other software? Is it possible to model an anisotropic porous medium?

I would be grateful if someone could point me to some resource on this topic.
Thank everybody in advance and regards,

Cosimo

[Brian Angel]

Hello

As far as I'm aware porous media is modelled as a source terms in Code_Saturne and yes you can model an anisotropic porous medium. In at least v3.2.1 and v4, which I've used for porous media modelling, the porous media can be set-up via the GUI or the user routines usporo.f90 and uskpdc.f90.

Please take a look at the GUI and these routines and if you have any questions pls let me know. In the meantime I'll prepare a small example case that will help you to get started.

Best regards,

Brian Angel.

[Brian Angel]

Hello,

Attached is an example of porous media modelling in Code_Saturne v3.2.1. The geometry and mesh were generated using SALOME and the calculation has run for 1000 iterations in Code_Saturne.

The file Porous_Media_GeomMesh.hdf.tgz is the SALOME file which contains the geometry, mesh and various boundary and volume groups that are used by Code_Saturne.

The file POR_MEDIA.tgz contains all the files for the Code_Saturne run.

Best regards,

Brian Angel.

[CTrinca]

Thank you very much for your reply.

As you suggested I gave a look to the subroutines you mentioned and yes, I do have a couple of questions.

As far as I can understand, all usporo.f90 does is to set porosity for each cell and to store it in an array called porosi, which then can be used to define a momentum sink in uskpdc.f90[\i], is that correct? If so, is it possible to use this value for other purposes, for instance to include the conductivity of the solid medium in the energy equation?

Also, I see that in the example usporo.f90 routine, porosi contains scalar values. How can it be modified in the anisotropic case?

I am using version 3.0.6 and I wasn't able to find where in the GUI the porous treatment is set up, either it wasn't included in that version or I have looked poorly...

Thank you again for your help.
Regards,

Cosimo

Edit: while I was writing, you published your example, so perhaps I will find there the answers to my question. Many thanks for your time.