On 5/4/23 21:31, Aditya Karthik Saravanakumar wrote:

Note: dofh_uhat is a DoFHandler<dim> object corresponding to the FE_FaceQ<dim,spacedim> finite element space.

_ERROR MESSAGE_
An error occurred in line <374> of file </home/--------/software/dealii-9.4.0/source/fe/fe.cc> in function     const dealii::FullMatrix<double>& dealii::FiniteElement<<anonymous>, <anonymous> >::get_prolongation_matrix(unsigned int, const dealii::RefinementCase<dim>&) const [with int dim = 2; int spacedim = 2]
The violated condition was:
     prolongation[refinement_case - 1][child].n() == this->n_dofs_per_cell()
Additional information:
     You are trying to access the matrices that describe how to embed a
     finite element function on one cell into the finite element space on
     one of its children (i.e., the 'embedding' or 'prolongation'
     matrices). However, the current finite element can either not define
     this sort of operation, or it has not yet been implemented.

Ah yes. But this is of course not the fault of the multigrid implementation, but rather that the element in question doesn't define these matrices.

So here's the underlying question: FE_FaceQ defines shape functions that only live on the faces of the cell. If you want to interpolate ("prolong") the solution from one mesh to a finer mesh, how are you going to define the solution on the new edges that are created in the interior of the cell?

When we wrote the element, we simply didn't know the answer to this question, and I still don't. What do the papers you reference use for this operation?

Best
 W.

--
------------------------------------------------------------------------
Wolfgang Bangerth          email:                 [email protected]
                           www: http://www.math.colostate.edu/~bangerth/


--
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 the Google Groups "deal.II User Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/dealii/b5348936-b9a6-c9f7-e933-5b3102c4e396%40colostate.edu.

Reply via email to