search

precice / fenics-adapter

preCICE-adapter for the open source computing platform FEniCS
GNU Lesser General Public License v3.0
29 stars 15 forks source link

Double-boundary points warning #166

Open mmichelis opened 8 months ago

mmichelis commented 8 months ago

I've been running a few FSI examples with OpenFOAM and FEniCS, and have been working into the FEniCS documentation, and like the simplicity of the framework so far. With some of the simulations, mostly soft bodies with neohookean material models, right after the precice coupling is done, some warnings are printed for "Found a double-boundary point at (x,y,z)".

I was wondering what causes this warning, and how I can look more into where this occurs. Even with a very simple BoxMesh, something is causing this warning. I will look further into the adapter and my code to understand what could potentially cause this warning, but perhaps others have seen this warning before. Thanks a lot!

BenjaminRodenberg commented 7 months ago

Hi @mmichelis!

The warning is raised at this place:

https://github.com/precice/fenics-adapter/blob/102022d5c3cdd6452bb0b5ab3900d20ec8508bca/fenicsprecice/adapter_core.py#L141

Some history about the warning can be found here: https://github.com/precice/fenics-adapter/issues/90

I think we should definitely improve this warning, because at the moment it is obviously confusing and not clear when this warning is something to worry about and when it can be safely ignored.

If you want to dig deeper into this issue: Any PR improving on this warning or (simple) test cases where this warning actually leads to a problematic situation would be highly appreciated. I unfortunately only vaguely remember the context and cannot spend a lot of time on it right now.

mmichelis commented 7 months ago

Hey Benjamin,

Thanks a lot for this, that will help me look into it a bit more! I will dig a bit as well, the simulation coupling is currently still exploding after a few iterations, so I'll dig a bit to see what is causing this (perhaps it's my fluid simulation). If it is because of this warning, I'll try to follow-up on this!

Cheers, Mike