Meshing complex geometries

[Kaushik Das]

Hi all,

I am trying to mesh a complex geometry by first creating a triangulation of the bounding box of the geometry and then successively refining the cells that are within the geometry.  I use a hp::FECollection to assign  FE_Nothing<DIM>() to cells that are outside the boundary and FE_Q<DIM>(1) to cells inside the geometry. Please see the attached figure, which shows only FE_Q cells.

A limitation that I am facing is I cannot use an anisotropic refinement case because make_hanging_node_constraints does not support hp::FECollection.

Is there a plan for relaxing this limitation in the future? Is there any other way to do this?

I want to keep the axis-aligned nature of cells that are generated this way. It is an additive manufacturing application. Therefore, using an external tet mesh generator is not a good option for me.

Thanks,

Kaushik

[Kaushik Das]

Forgot to attach the figure. Here it is. 

[Peter Munch]

Hi Kaushik,

> Is there a plan for relaxing this limitation in the future? Is there any other way to do this?

Unfortunately, there are currently no active developments regarding anisotropic refinement. Would it be an option to refine isotropically? Latter works very well parllel::distributed::Triangulation in parallel.

PM

[Kaushik Das]

Hi Peter,

Yes, I am using isotropic refinement. For some geometries, using anisotropic refinement, I may be able to keep the problem size smaller. I am also experimenting with using p::d::triangulation, and there is a limitation that anisotropic refinement does not work with p::d::triangulation. 

Ideally, I would like to use anisotropic refinement with hp::FECollection on p::d::triangulation. My problems sizes are often > 3M DOFs and they are transient heat transfer problems that take 10s hr to a few days on 32 CPUs.  

Thanks for the help.

Kaushik 

--
The deal.II project is located at http://www.dealii.org/
For mailing list/forum options, see https://groups.google.com/d/forum/dealii?hl=en
---
You received this message because you are subscribed to a topic in the Google Groups "deal.II User Group" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/dealii/agfsxb_wKqM/unsubscribe.
To unsubscribe from this group and all its topics, send an email to dealii+un...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/dealii/652623a8-1308-4cce-93b0-79c646a0410bn%40googlegroups.com.