Re: [deal.II] interpolate FE_Nelelec

Hi John, Maybe I can give some hints of how I would approach this problem (these are just some quick thoughts): Write this problem as a vector Laplace problem curl(nu*curl(A)) - grad(div(A)) = J which you then have to write as a system of PDEs. Note, this can only be the same as

Re: [deal.II] interpolate FE_Nelelec

Dear John, The class documentation states > Several aspects of the implementation are experimental. For the moment, it is > safe to use the element on globally refined meshes with consistent > orientation of faces. My

Re: [deal.II] interpolate FE_Nelelec

Dear Jean-Paul, Thanks! It is indeed a trap. John On Tuesday, April 20, 2021 at 9:23:23 PM UTC+2 Jean-Paul Pelteret wrote: > Hi John, > > Unfortunately, you’ve fallen into the trap of confusing what the entries > in the solution vector mean for the different element types. For Nedelec

Re: [deal.II] interpolate FE_Nelelec

Dear John, As Jean-Paul pointed out the entries of the solution vector for a Nedelec element have a different meaning. The entries are weights that result from edge moments (and in case of higher order also face moments and volume moments), i.e, integrals on edges. Nedelec elements have a deep

Re: [deal.II] interpolate FE_Nelelec

Hi John, Unfortunately, you’ve fallen into the trap of confusing what the entries in the solution vector mean for the different element types. For Nedelec elements, they really are coefficients of a polynomial function, so you can’t simply set each coefficient to 1 to visualise the shape

Re: [deal.II] interpolate FE_Nelelec

To add to Wolfgang’s comment, you can have a look at the compute_edge_projection_l2() function that forms a part of the function that computes

Re: [deal.II] interpolate FE_Nelelec

Dear Jean-Paul, Thank you for your reply. I am solving a very standard curl(1/mu(curl(A)) = J problem as described in the paper of Oszkar Biro. https://ieeexplore.ieee.org/document/497322 No gauge. I need interpolation to project the vector current potential on the shape functions. I just

Re: [deal.II] interpolate FE_Nelelec

Dear John, I’m not sure that there is an interpolation function that would work in that way for Nedelec elements (there is VectorTools::project_boundary_values_curl_conforming_l2() for

[deal.II] interpolate FE_Nelelec

Hello, It seems I am unable to find a function similar to VectorTools::interpolate for FE_Nedelec finite elements. The existing implementation of this functions gives the following run-time error: *An error occurred in line <556> of file in function* * void