Hello,
Wolfgang wrote:
> GridRefinement::refine_and_coarsen_fixed_{number,fraction} have a trailing
> defaulted argument that specifies a maximum number of cells. It doesn't
> always hit the target exactly but at least it allows you to specify that it
> shouldn't refine cells above just because they are above a certain threshold.
> It stops when the number of projected cells is above the given maximum.
Yes, I saw that, but that does not help a lot due to
> "Be aware though that through proliferation of refinement due to
> Triangulation::MeshSmoothing, this number is only an indicator"
(i.e. the limit is before considering smoothing)
Guido wrote:
> Controlling the exact number of cells is difficult, because of mesh
> smoothing. You should rather shoot for an approximation.
Yes, sure. An exact match is often even not possible (consider one
coarse cell and I ask for 3 :-). An approximation is certainly enough.
The problem I see is that I can keep the number of cells roughly in
the same order in a single adaption without a problem. But when I run
a time dependent problem over a long time interval with thousands of
adaption steps, the number of cells can easily change a lot.
--
Timo Heister
http://num.math.uni-goettingen.de/~heister
On Tue, Sep 28, 2010 at 2:53 PM, Wolfgang Bangerth
<[email protected]> wrote:
>
> Timo,
>
>> So, what are you using and do you have any advice for me? Surely,
>> someone had a similar problem already? :-)
>
> GridRefinement::refine_and_coarsen_fixed_{number,fraction} have a trailing
> defaulted argument that specifies a maximum number of cells. It doesn't
> always hit the target exactly but at least it allows you to specify that it
> shouldn't refine cells above just because they are above a certain threshold.
> It stops when the number of projected cells is above the given maximum.
>
> Best
> W.
>
> -------------------------------------------------------------------------
> Wolfgang Bangerth email: [email protected]
> www: http://www.math.tamu.edu/~bangerth/
>
>
_______________________________________________
dealii mailing list http://poisson.dealii.org/mailman/listinfo/dealii
