At the moment there is no function that extract a boundary mesh and generates a codimension one mesh. However, it is quite straight forward to implement. But if you generate your meshes with cubit, for example, then it is already possible to do exactly what you want to do, by exporting first the volume mesh, and then the boundary mesh into two separate files, and read them with the GridIn class (this is what I do for my own work, since I was too lazy to implement also the Boundary extractor function... :) ...).
As far as communication goes, you will have to generate a map that joins the two mesh iterators (i.e., a map that, for a face iterator on the boundary of the <3,3> mesh, associates the cell iterator of the <2,3> mesh), and that should be pretty much all you need to do. Generation of Sparsity Patterns for the coupling of the two dof handlers, of course, will have to be done by hand. You can then use FEFaceValues<3,3> on the face iterator of the volume mesh <3,3> and FEValues<2,3> on the cell iterator of the surface mesh <2,3>. Of course, all of this works also with <2,2> and <1,2>. I'd strongly suggest you try that one first... :) Let us know if this works for you! Luca. -- Luca Heltai <[email protected]> http://people.sissa.it/~heltai/ Scuola Internazionale Superiore di Studi Avanzati Phone: +39 040 3787 449, Office: 732 -- There are no answers, only cross references. On 08/set/2010, at 07.24, Markus Bürg wrote: > Hello Sebastian, > >> -Given a dim_of_world mesh be able to extract a co-dimension 1 boundary mesh >> > For this point you should have a look on tutorial step-34. However I do not > know of function to extract the boundary mesh, but you could start with two > meshes (one for the dim_of_the_world and one for the co-dimension 1). >> -These two meshes (volume and surface) are to remain conforming. If >> one gets refined the other should be also consistently refined and >> synchronized when moved. >> > The synchronization should not to be too hard, you just have to do the > refinement and coarsening on both meshes accordingly. >> -The ability to trace a bulk fe_function to the boundary and viceversa. >> > I am not sure, what you want to do here. Could you explain this a bit more? > > Best Regards, > Markus > _______________________________________________ > dealii mailing list http://poisson.dealii.org/mailman/listinfo/dealii _______________________________________________ dealii mailing list http://poisson.dealii.org/mailman/listinfo/dealii
