This is a read only copy of the old FEniCS QA forum. Please visit the new QA forum to ask questions

Demo for time-dependent adaptivity?

+5 votes

Is there an easy way to do time-dependent adaptive simulations in pydolfin? As far as I can see, there is adapt, but there seems to be no obvious mechanism for mesh coarsening implemented. Any hints?

asked Jun 5, 2014 by Christian Waluga FEniCS Expert (12,310 points)

2 Answers

+2 votes
 
Best answer

Hi Christian, I don't believe there is a mechanism for mesh coarsening in fenics right now. I discussed this with one of the developers recently and he is looking into it. It's fairly complex (3D, partitioned meshes, etc.) so it might take a while.

Regards,
Dave

answered Jun 8, 2014 by david.bernstein FEniCS User (2,000 points)
selected Jun 12, 2014 by Christian Waluga

(deleted, changed to answer)

commented Jun 10, 2014 by KristianE FEniCS Expert (12,900 points)
+2 votes

If you have a non-curved 2D geometry, pragmatic (https://github.com/ggorman/pragmatic) might be an option. It allows anisotropic adaptation. Let me know, if you decide to go this route and need help.

I am using it with FEniCS, and I have made a number of simulations with this combination (https://sites.google.com/site/ejlebjerg/home/viscous-fingering).

answered Jun 10, 2014 by KristianE FEniCS Expert (12,900 points)

Thanks for your answer! I want to use this for teaching, so I would prefer to use plain FEniCS without additional dependencies.

commented Jun 11, 2014 by Christian Waluga FEniCS Expert (12,310 points)
...