[Gmsh] anisostropic meshing / bamg algorithm

Wed Feb 29 18:14:47 CET 2012

Hi,

I still do not succeed in creating an anisotropic mesh as I would like to. My problem is as follows:
I have a structure consisting of several stacked layers of different thickness (see attached file for an example). In the x-y plane, the mesh can be rather coarse (say the same with for all geometric layers), whereas in z direction, I would like to have at least 2 layers of mesh elements per geometry layer. This means that the mesh width in z direction should be small inside a thin layer, and can be wider inside a wide layer (e.g. the one at the bottom). How do I achieve this? I have tried to use the MathEvalAniso field, and I have succeeded to create an anisotropic mesh for a simple cube. But I do not fully understand from the reference manual how to use this field the right way - my math knowledge seems not to be enough to understand from the Wikipedia article what the metric tensor does. I would need the mesh width to depend on z, resp. on the geometric layer, so I would have to declare several Aniso fields and attach them somehow to the layers of my structure.

Could someone give me some hints on how to proceed here?

Thank you in advance,

Matthias

Von: Martin Vymazal [mailto:martin.vymazal at vki.ac.be]
Gesendet: Donnerstag, 16. Februar 2012 12:17
An: gmsh at geuz.org
Betreff: Re: [Gmsh] anisostropic meshing / bamg algorithm

Hello Matthias,

you can use the bamg algorithm by calling 'gmsh -2 -algo bamg my_file.geo' from command line. For 3d anisotropic mesh, you do the same with '-3 -algo mmg3d'.

You need to supply a mesh size field which will tell the anisotropic algorithm in which direction to refine the mesh.

Best regards,

Martin Vymazal

On Thursday 16 February 2012 10:56:06 Zenker, Dr. Matthias wrote:

Hi,

does anyone have an answer to the question below?

It was said that an anisotropic mesh can be achieved using the bamg algorithm. I have some questions:

1.                   My 3D geometry consists of several layers stacked in z direction. Some of these layers are rather thin. I would like to have a mesh which is finer in z direction than in x and y direction. Can this be achieved using the bamg algorithm? If so, how?
2.                   How can I access this algorithm from the GUI? I did not find it in Options – Mesh.
3.                   If not accessible from the GUI, how else can I use it?
4.                   Is there any documentation available concerning this topic? I did not find anything in the gmsh manual (which is from October 2010 btw).

Thank you,

Matthias

----------------------------------------------------------------

ERBE Elektromedizin GmbH

Dr.rer.nat. Matthias Zenker

Dipl. Phys.

Grundlagenentwicklung / Fundamental Development

Waldhoernlestrasse 17

72072 Tuebingen

Germany

Phone + 49 (0) 7071 755 - 226

Fax + 49 (0) 7071 755 - 5226

E-Mail: <mailto:Matthias.Zenker at erbe-med.com>

URL: http://www.ERBE-med.com

----------------------------------------------------------------

_____________________________________________________________________
ERBE Elektromedizin GmbH
Firmensitz: 72072 Tuebingen
Geschaeftsfuehrer: Christian O. Erbe, Reiner Thede
Registergericht: Stuttgart HRB 380137

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.geuz.org/pipermail/gmsh/attachments/20120229/3e80fdc3/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Block_3D.geo
Type: application/octet-stream
Size: 2860 bytes
Desc: Block_3D.geo
URL: <http://www.geuz.org/pipermail/gmsh/attachments/20120229/3e80fdc3/attachment.geo>