You can try out MGTransferGlobalCoarsening (see https://dealii.org/developer/doxygen/deal.II/step_75.html). The infrastructure is quite flexible.
Best, Peter On Wednesday, 3 September 2025 at 20:18:18 UTC+2 [email protected] wrote: > I am trying to develop a Stokes-Darcy solver. In this regard, I would like > to apply geometric multigrid as a preconditioner to the Darcy part of the > domain. Overall, I'm treating the Stokes-Darcy system discretization using > an hp::FECollection object, and attaching it to a dof handler. > > Following step-56, I would like to attach a darcy_dof_handler to the darcy > degrees of freedom. However, this is not possible through the existing > distribute_mg_dofs() functionality, as hp::FECollection is not supported > through the DoFHandler. > > I can understand why applying geometric multigrid to a part of the > existing grid would be challenging, as we do not discriminate between > Cartesian and generalized domains within the deal.II framework. > > Are there any work-arounds which do not require me to setup dofs for the > entire FESystem everywhere on the mesh, and apply constraints on individual > parts of the domain? > > > -- 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 visit https://groups.google.com/d/msgid/dealii/3e54feeb-f65f-4b62-9d9f-34eb290afc72n%40googlegroups.com.
