Native Abaqus format support for mesh_converter

[Krzysztof Bzowski]

Hi,
Mesh converter is very nice tool, but I noticed that it can only work
with Abaqus inp files make in Cubit. In our department we work with
Abaqus often, so I tried to check how the tool would behave with the
native Abaqus inp files. And it didn't work good. It appeared, that
native inp files have a lot more options, which should be also
taken into account during conversion. So I developed a patch to fix some
issues and add support for native abaqus input files.

I also prepared two simple test cases with 2D and 3D models. Binary CAE
files (with models) were also attached (require at least Abaqus 6.13 to
run).

Some assumptions:
1. The default name of the material should not be changed. It means that
Material-1 is mapped to material_id = 1 etc.
2. Surface can have any name which contains a digits. That digits are
mapped to boundary_id. It can cause some problems if two surfaces have
same number in the name.

Patch passed all previous tests.

More tests are still require. Mismatch elements types, wrong names etc
should be checked during conversion. I will prepare that in the near future.

I think mesh converter tests should be moved to global tests directory
(that which are run using ctest).

Any advice and suggestions will be greatly appreciated.

Best regards,
Krzysztof

[sean....@gmail.com]

Hi Krzysztof,

I think this file is really helpful for me to convert ucd file directly from abaqus input file. Thanks.
But, I'm not sure how I can compile the mesh converter using this .path file. Could you let me know how I can use this?
Thanks in advance.

Best,

Sean

2014년 5월 17일 토요일 오전 9시 58분 13초 UTC-4, Krzysztof Bzowski 님의 말:

[Jean-Paul Pelteret]

Dear Krzysztof,

Thanks very much for this contribution to the mesh converter program. I'm sorry that I did not reply to your message! I don't remember seeing it, so I do apologise. 

Coincidentally, just this week my colleague and I started making some amendments to the program (bug fixes for unstructured grids, support for input from GiD) so I'll look at including your changes at the same time and, if you don't mind, we can include your test cases as well.

Many thanks again! All the best,

J-P

[Wolfgang Bangerth]

Krzysztof,
I still had this email in my inbox and never got around to dealing with
it. Of course, Jean-Paul Pelteret eventually did and your patches are
now integrated. Either way, a great many thanks for providing these patches!

I've run into one mesh for which I run into trouble:
https://github.com/dealii/dealii/issues/1409
I'm not an expert in the ABAQUS file format. Do you happen to have a
reference that describes this format?

Best
Wolfgang

--
------------------------------------------------------------------------
Wolfgang Bangerth email: bang...@math.tamu.edu
www: http://www.math.tamu.edu/~bangerth/

[Jean-Paul Pelteret]

Dear Krzysztof and Wolfgang,

Pending moderation, I think that this issue has now been resolved. 

J-P