Hello Mihai,
you can also get the patchwise quadratic interpolation through the
FETools::extrapolate -
function. In this case, there is no need to coarsen your mesh.
Kind regards,
Christian
Quoting Markus Bürg <[email protected]>:
Hello Mihai,
I think SolutionTransfer is what you are looking for.
Best Regards,
Markus
Am 02.03.11 15:31, schrieb mihai alexe:
Hello,
I have a question concerning local higher-order interpolation (more
precisely, patch-wise bi-quadratic interpolation).
I start from a solution on a "fine" mesh Omega_h, with a bilinear
discrete solution u_h defined on it. I coarsen all cells in that
mesh once to get Omega_2h. I then have to build from the data on
Omega_h a bi-quadratic discrete solution u_2h. Is there API support
for getting u_2h in deal.ii?
I guess I can take every active cell on Omega_2h and get the
support point values from Omega_h that lie in that (bigger) cell,
and then build the bi-quadratic solution by hand. Due to Omega_2h
being obtained directly from Omega_h by pure coarsening, I should
have enough point values for a bi-quadratic interpolation. Is there
a cleaner way to accomplish this?
-- Thanks!
_______________________________________________
dealii mailing list http://poisson.dealii.org/mailman/listinfo/dealii
--
Christian Goll
Institut für Angewandte Mathematik
Universität Heidelberg
Im Neuenheimer Feld 293 (Raum 210)
69120 Heidelberg
Telefon: +49 6221 54-6154
_______________________________________________
dealii mailing list http://poisson.dealii.org/mailman/listinfo/dealii
