[Getdp] RE: Adaptive mesh refinement with getDP and gmsh

Jehudi Maes Jehudi.Maes at rug.ac.be
Thu Apr 26 09:40:55 CEST 2001


Dear GetDP users,

I would like to use GetDP and Gmsh for adaptive mesh generation. Included
you can find my (the part on adaptive mesh generation) original post to Mr.
Dular and Mr. Geuzaine.

I would like to thank Mr. Geuzaine for his quick reply. Indeed the main
problem in the adaptive mesh generation is the generation of the error. I am
not a specialist in the domain of FE-calculations but from literature I have
found that for my problem a good error function could be the difference in
the tangential component of the H or B field calculated from the FE-solution
and an estimation of this component from an interpolation between an element
and it's neighbours. However for the construction of the error field I would
need:

1. To get the tangential component of H (or B) from the solution. Is this
possible in GetDP?
2. To access the neighbour values of H and B. Is this possible in GetDP ?
3. Possibly (for scaling) also the surfaces of each element.

Kind regards,

Jehudi Maes

> -----Original Message-----
> From: geuzaine at aix10.segi.ulg.ac.be
> [mailto:geuzaine at aix10.segi.ulg.ac.be]On Behalf Of Christophe Geuzaine
> Sent: Wednesday, April 25, 2001 7:21 PM
> To: gmsh at geuz.org
> Cc: Jehudi.Maes at rug.ac.be; Dular Patrick
> Subject: Re: [Gmsh] [Fwd: Adaptive mesh refinement with getDP and gmsh]
>
>
> Jehudi Maes wrote:
> >
> > Dear Mr. Dular and Mr. Geuzaine,
> >
> >  I have been working with GetDP and gmsh for the FE calculations of a
> > switched reluctance motor (magnetostatic calculations). For these
> > calculations I have had a lot of help from one of your collegues Johan
> > Gyselinck. From the websites and the documentation I concluded
> that it is
> > possible to use GetDP and gmsh for adaptive mesh generation.
> Could you be as
> > kind as to point out how I have to use the combination of GetDP
> and gmsh for
> > adaptive mesh generation.
>
> For h-type adaptation (i.e. modification of element sizes), you should
> output a post-processing view which contains the target characteristic
> lengths. Thus, you should
>
> 1) generate a first (coarse) mesh
> 2) compute target mesh sizes on this mesh, and output the results as a
>    Gmsh post-processing view
> 3) load this view in Gmsh, and apply it as the current background mesh
>    (with 'View->Apply as Background Mesh'), or load the view on the
>    command line, with the option '-bgm file'
> 4) re-mesh
>
> Step 2) is of course the difficult part, and depends on your problem.
> GetDP can help to generate the target size view if you have a mean to
> compute an error field, i.e. some 'volumic' field on the mesh
> representing (an approximation of) the error. If you can compute such a
> field, then you can use 'Adapt H1' in a GetDP PostOperation on this
> field, and you will obtain a valid background mesh for Gmsh.
>
> WARNING: if you plan to use background meshes, you need the very latest
> version (1.18) of Gmsh. This version fixes a serious bug in the
> generation
> of adapted 1D meshes.
>
> For p-type adaptation (i.e. local modification of interpolation order),
> you should output an adaptation file, and give it to GetDP on the
> command line with the '-adapt file' option. The adaptation file can be
> generated by GetDP by using 'Adapt P1' and by specifying 'Adaptation' in
> the Format of a PostOperation.
>
> >
> > Kind regards,
> >
> > Jehudi
>
> I have forwarded your mail to the gmsh mailing list. If you reply to
> this message, please reply to the mailing list.
>
>
> Christophe
>
> --
> Christophe Geuzaine
>
> Tel: 32 (0) 4 366 37 10    http://geuz.org
> Fax: 32 (0) 4 366 29 10    mailto:Christophe.Geuzaine at ulg.ac.be
>