[Gmsh] Problem with importing a mesh in Abaqus

Christophe Geuzaine cgeuzaine at ulg.ac.be
Thu Apr 11 12:25:32 CEST 2013


On 11 Apr 2013, at 01:21, Omid Mahabadi <omid.mahabadi at geomechanica.com> wrote:

> Dear Christophe,
> I have been using the updated Gmsh with the Abaqus NSET export and it has been working very well so far. I've only come across a tiny issue, that is, the names of Physical Entities are not used from the geometry file. Instead, they are numbered automatically in the exported inp file (e.g., PhysicalLine2,..., PhysicalSurface1,...). Is there a way to preserve the names specified in the geo file?
> 

This should be fixed in the latest svn.

Thanks for the feedback,

Christophe

> Thank you,
> Omid
> 
> 
> On 05/04/13 12:02 PM, Christophe Geuzaine wrote:
>> On 05 Apr 2013, at 17:49, Omid Mahabadi <omid.mahabadi at geomechanica.com> wrote:
>> 
>>> On 05/04/13 10:25 AM, Christophe Geuzaine wrote:
>>>> On 05 Apr 2013, at 16:13, Omid Mahabadi <omid.mahabadi at geomechanica.com> wrote:
>>>> 
>>>>> Hi,
>>>>> I thought quite a bit about element types in Abaqus. The reality is that there are so many different element types in Abaqus that makes defining a "generic" type rather impossible. I would suggest to keep the default Abaqus export as currently is, but give the user the option to define custom types in the gmsh geometry file. For instance, in the case of a 2D mesh, the user may define a maximum of two strings for the 1D and 2D element types, which can look like:
>>>>> Mesh.Abaqus.ElementType.1D = "C1D2"
>>>>> Mesh.Abaqus.ElementType.2D = "C2D3"
>>>>> (I'm just mimicking gmsh's syntax here)
>>>>> 
>>>>> I discussed the above suggestion with another colleague of mine who has used Abaqus in the past and it made sense to them. What are your thoughts?
>>>>> 
>>>> With SVN r15202 I at least changed the default tags to tags that should be understood by Abaqus/Standard (had to change lines, triangles and quads). We might want more flexibility in the future, but it might also be overkill (we need types for 1st order and 2nd order elements, etc.).
>>> I totally agree as user-defined types can get complicated.
>>> On a side note, for your information I've realized that now Element numbers don't start from 1 anymore.
>> Indeed: that's a side-effect of the new scheme. Since we must remember the numbers in the ELSETS, we cannot just renumber on-the-fly as before.
>> 
>>>> 
>>>>> Thank you,
>>>>> Omid
>>>>> 
>>>>> 
>>>>> On 05/04/13 06:35 AM, Christophe Geuzaine wrote:
>>>>>> On 05 Apr 2013, at 12:07, Christophe Geuzaine <cgeuzaine at ulg.ac.be> wrote:
>>>>>> 
>>>>>>> On 18 Mar 2013, at 18:03, Pierre-Frederic Villard <pierre-frederic.villard at univ-lorraine.fr> wrote:
>>>>>>> 
>>>>>>>> Hi,
>>>>>>>> 
>>>>>>>> I have issues with importing gmsh-created meshes in Abaqus 6.12. I tried two methods, which equally failed:
>>>>>>>> 1 - I tried to save the mesh as OpenCASCADE step format but I have an error:   "Error   : No OpenCASCADE model found"
>>>>>>> Hi Pierre-Frederic - This would actually not export the mesh, but the CAD model - so it's not what you want.
>>>>>>> 
>>>>>>>> 2 - I tried to save the model as an Abaqus INP file but I have a problem importing it in Abaqus : "Element type "C2D3" is not supported." with triangles and "Element type "C1D2" is not supported" with segments.
>>>>>>>> 
>>>>>>> Indeed. I just had a look at the Abaqus documentation, and C2D3 and C1D2 do not seem to be in the general "Element Indexes" available in Abaqus/Standard.
>>>>>>> 
>>>>>>> What should we use when we export "generic" elements (i.e., not linked to any physics)?
>>>>>>> 
>>>>>>> @Omid - any ideas?
>>>>>> I've committed r15202 with the ideas from
>>>>>> 
>>>>>> https://geuz.org/trac/gmsh/ticket/129
>>>>>> 
>>>>>> @Abaqus users: let me know if this is better...
>>>>>> 
>>>>>> 
>>>>>> 
>>>>>>>> Does anyone have any idea how to solve this problem?
>>>>>>>> 
>>>>>>>> Cheers,
>>>>>>>> 
>>>>>>>> Pierre
>>>>>>>> 
>>>>>>>> 
>>>>>>>> _______________________________________________
>>>>>>>> gmsh mailing list
>>>>>>>> gmsh at geuz.org
>>>>>>>> http://www.geuz.org/mailman/listinfo/gmsh
>>>>>>> -- 
>>>>>>> Prof. Christophe Geuzaine
>>>>>>> University of Liege, Electrical Engineering and Computer Science
>>>>>>> http://www.montefiore.ulg.ac.be/~geuzaine
>>>>>>> 
>>>>>>> 
>>>>>>> 
>>>>> -- 
>>>>> Omid Mahabadi
>>>>> Geomechanica, Inc.
>>>>> Tel: (647) 478-9767 x824
>>>>> 
>>> -- 
>>> Omid Mahabadi
>>> Geomechanica, Inc.
>>> Tel: (647) 478-9767 x824
>>> 
> 
> -- 
> Omid Mahabadi
> Geomechanica, Inc.
> Tel: (647) 478-9767 x824
> 
> 
> _______________________________________________
> gmsh mailing list
> gmsh at geuz.org
> http://www.geuz.org/mailman/listinfo/gmsh

-- 
Prof. Christophe Geuzaine
University of Liege, Electrical Engineering and Computer Science 
http://www.montefiore.ulg.ac.be/~geuzaine