[Dune] Local Refinement in a DUNE Grid

[Oliver Sander]

That depends on the grid manager.  UGGrid and AlbertaGrid will ensure that
automatically, ALUGrid will not (and that's not a bug).

--
Oliver

Am 26.12.2010 12:41, schrieb S. Swayamjyoti:
> Now, will that ensure that the refined grids are *conforming* with the 
> other level elements?
>
> On Sun, Dec 26, 2010 at 12:36 PM, Jö Fahlke <jorrit at jorrit.de 
> <mailto:jorrit at jorrit.de>> wrote:
>
>     Am Sun, 26. Dec 2010, 06:18:46 +0100 schrieb S. Swayamjyoti:
>     > I am using DUNE for creating a grid over which I will apply
>     Finite Element
>     > Methods. I have created a structured 3D grid of cubes. Now, I
>     want to refine
>     > the cubes at the center of the domain. I require fine meshing
>     for cubes at
>     > the center while course meshing is OK as we go outward from the
>     center.
>     > Course meshing away from the center will also save me
>     computational cost.
>     >
>     > Could someone please suggest how to refine the grids at the
>     center - if
>     > there is a local refinement class? Please also suggest the
>     header file to
>     > include for the same.
>
>     You have to iterate over the elements (the codim-0 entities) of the
>     LeafGridView.  For each element that you want to refine, call
>     grid.mark(1,
>     elem).  Once Iteration is done, call grid.preAdapt(),
>     grid.adapt(), and
>     grid.postAdapt().  Repeat until you are satisfied with the grid.
>
>     If you know a priori that a certain element can be refined n times
>     uniformly,
>     you can also tell that to grid.mark() by calling grid.mark(n, element)
>     instead.
>
>     Bye,
>     Jö.
>
>     --
>     Sie können die Unversität jetzt ausschalten.
>
>     -----BEGIN PGP SIGNATURE-----
>     Version: GnuPG v1.4.10 (GNU/Linux)
>
>     iQIVAwUBTRco1emx9CKgc0OWAQjG5RAA2MkRdSxkVRsNftEZj5chH21/QQ3PynzF
>     m7wpygqbVfJv1HXzccy+X5PeKT2fsjU5hcBAir3Trd2B2B8OULjrcf9u3SP4fSgA
>     4zCidHKzq0Fw+x1S7v+TLkenxMLlCaBtYflLWAC0EK65Utlrrn5e2d5VhBthcEEN
>     RIHN538dVFxBYMOOCLzUAIZ84Cpvxhg14Tu+d3x/zNXytsOo3QM8FA/ekGIZZXOB
>     v6pApN3mFjRrDD9u2b5lETmNNxP3WQkb3QF+n6csXHcJQ6Y7eTQkfjhP0OZlisAf
>     5fo4TNe+Q2R2yAbOlwh+0BMG8yUL4IcEY1HZ+8FFathCpnNaJcRQkDojMDHkGIad
>     myje8E6hJFKXpOmNTm7aOIy/xJrg1cxZCPk5Lfa3TMPafdnRk3xZSw7j4CoACkTO
>     l03+wZvTO7yjmDSIkbXLwduUHPAgGDMPQ6j0sykwm/ipYPdbj0RzhxKBWKfGAw50
>     5IgFXl4GDmGh83AEvIMNhgCBA6/p7T+GShibQcwiMhidkqsWDHW/8rJjgErZK8PM
>     mSYXfEfeO3PUvWvV3aRqEohMx1wOnsD0biYK/f3b6AbnNna1VTmDwgFb9OzmBaJQ
>     5K8haH285fGty8BhtUUU0e4bpgP2lQTJCcPcuw6u/TvpHu8lVEqf29WcPfloca5P
>     2kcTkKP210M=
>     =hWEp
>     -----END PGP SIGNATURE-----
>
>
>
> _______________________________________________
> Dune mailing list
> Dune at dune-project.org
> http://lists.dune-project.org/mailman/listinfo/dune
>    

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.dune-project.org/pipermail/dune/attachments/20101226/62e8dee7/attachment.htm>