Re: [Getfem-users] different mesh_fem definitions for different regions

[Renard Yves]

Umut Tabak <address@hidden> a écrit :

> Renard Yves wrote:
> > Umut Tabak <address@hidden> a écrit :
> >
> > > Renard Yves wrote:
> > > > Dear Umut,
> Hi again,
>
> I tried to apply a sample analysis where I can assemble the fluid  
> part(scalar) and the structural part(displacement). Now I have a  
> problem concerning the coupling matrix assembly. I attached the code  
> and the necessary msh files. Now the questions are(better to answer  
> these after taking a look at the source file attached. In brief, I  
> have 3 mesh structures and 3 mesh_fems attached to these mesh  
> structures and the operations should be clear for you.)
>
> + To have this kind of analysis(coupled structural-acoustic), we  
> have to set the mesh_fem dimensions to 3 and 1 for the structural  
> and fluid parts respectively, if there is only one mesh_fem  
> definition, how can I assign different dimensions to different  
> regions of my mesh?

You cannot ! Even if you have only one mesh representing both the  
structure and the fluid, the best is to have two mesh_fem, each  
defined only on the corresponding region. This is the easiest for  
computing the coupling terms but of course it means that the meshes of  
the structure and the fluid match to each other.
If you have two separate meshes you can try to compute the coupling  
term using the object "interpolated fem" which interpolate a fem from  
a mesh to another (see for instance tests/test_interpolated_fem.cc)


> + The second question follows accordingly, how to assign different  
> integration methods to different regions of the mesh? Assembly can  
> be done on regions as you suggested before.

As for the mesh_fem object, the mesh_im object allows to set a  
different integration method on
eache element.


> + Without the definition of separate mesh_fem objects, how can I  
> allocate memory for the matrices that I am going to use for the  
> assembly process? I mean to get the specific dof numbers associated  
> to regions.
>
> + With the attached file, I can assemble structural domain and apply  
> boundary conditions on that, also the fluid part, but i have a  
> problem related to a term like
>
> \begin{displaymath}
> \bold{K}_{up}=\int_{S}{\bold{N}_{s}}^{T} \bold{n}{\bold{N}_{p}}dS
> \end{displaymath}
>
> that I would like to assemble on the interface of the two above  
> defined domains.
>
> Shape functions N_s will come from the structural domain and N_p  
> will be the contributions of the fluid domain. But i could not  
> assemble this term over the surface defined in the mesh, I  
> understand the error message but I could not think a way to fix it.  
> The error is ''error: the mesh_fem/mesh_im live on different  
> meshes!''.

Yes. You cannot make a standard assembly on two different meshes. The  
only mean
to do that is to use the interpolated fem object.

Yves.

> + Lastly, are there any curved shell elements implemented in gmsh?
>
> I would appreciate further help on this.
>
> Thanks and best regards,
>
> Umut