code_saturne User's Forum

Hi Yvan,
Thank you, with conjugate gradient i could go a little further in the simulations (more restarts). For what i understood the problem is higly mesh related. I mean, i should have all the tetras in a very small dimension range. Usually netgen make a finer mesh at the borders and next cells grows inside the volume. After N restarts inner coarser cells are boundary cells and thus may lead to divergence. I'm trying a more regular mesh to see if there is any improvement. Just as a final question : does Saturne has any remeshing libraries included (like code_aster) ? I don't understand the smoothing mesh option since it doesn't seems to perform any change in the final mesh.

Thanks

Andrea

Hi all,
i'm just writing to thank you, Yvan, for your precious support. I'm starting to perform some dedicated CFD topology optimization simulations with CS. Now the code is quite stable (i'm still using 4.0) and i'm trying several configurations, the image attached is related to a very simple 2D flow optimization case.
I gave a fixed design space where the software try to maximize velocity removing the not worthy cells iteration after iteration.

In my opinion CS has a very strong potential in the optimization field. Now i'm trying more complex and 3d model, i hope to share with you something valuable soon.

Thanks again

Andrea