[Dune] DUNE for mix FEM+FVM methods?

Oliver Sander sander at mi.fu-berlin.de
Sat Feb 26 22:27:30 CET 2011


Hi Ali!
 From the description of your problem it seems that Dune is a
good solution for you.  The way Dune handles grids is extremely
flexible, and it is no problem at all to have two (or more) different ones
at the same time.  In fact I do that on a regular basis for various
kinds of domain decomposition and multiphysics problems.
We have various grids with excellent support for local grid
refinement.  We do not have an actual dedicated AMR grid yet,
but I think that can be added easily.
I suggest you have a look at our tutorials and online documentation.
Don't hesitate to ask on the list if you have further questions.

Best,
Oliver

Am 26.02.2011 01:36, schrieb Ali Roustaei:
> Hello,
>
> I am looking for a library for solving some special fluid problems and DUNE
> looks a very complete and nice one!
>
> Our current FEM codes are based on a library called
> PELICANS and it used to work well.
>
> But it has been for sometime that we are interested in2 phase fluid
> problems and we've faced several problems. Because we want to
> use a combination of FEM+FVM
>
> FEM for solving the flow and
> FVM is solving the Volume Of Fluid (VOF) equation.
>
> One of the biggest problems is mesh. We want different meshes
> for FEM and FVM and PELICANS has no idea of it. Also it would be nice
> to have AMR that PELICANS has not by default.
>
> Currently using old FEM codes but it needs many tricks and we can
> use it only for very simple geometries.
>
> We plan to change our base library or at least use a new one for later works and I
> am looking for it.
>
> Can anyone help us. Is DUNE able to attack such problems in a clean and nice way?
>
> Thank you all
>
>
> _______________________________________________
> Dune mailing list
> Dune at dune-project.org
> http://lists.dune-project.org/mailman/listinfo/dune
>    





More information about the Dune mailing list