Re: [Libmesh-users] DG and boundary conditions

[Roy S. <roy...@ic...>]

On Wed, 2 May 2018, Renato Poli wrote:

> Well, I am working with fracture creation in between mesh elements. In
> this case I believe it sort of makes sense to have 'strong' BCs and
> internal discontinuous DOFs being tied together by interior
> penalties... (not sure though)

I think strong BCs will hurt your convergence rates, but you should
still converge, so it couldn't hurt to try and see if I'm wrong.

> I was considering to duplicate the DOFs myself during topology
> creation: I would add triangles sharing vertices _coordinates_ but not
> vertex _references_. Then I map the neighbors and tie them together
> during assembly.
>
> Would that be sound?

Gah.  At that point you'd have to add manual constraint equations
between your solution triangles and the duplicate triangles the
constrained boundary DoFs, so you might as well add manual constraint
equations between your solution triangles and the desired boundary
values directly.
---
Roy