.msh file triangulation failed

[yuesu jin]

Dear all,

  I have one problem when I try to input the .msh file with grid_in function. The error message is :

*****************************************************************

Exception on processing:

--------------------------------------------------------
An error occurred in line <3139> of file </project/cacds/build/dealii-9.2.0/source/grid/tria.cc> in function
static void dealii::internal::TriangulationImplementation::Implementation::create_triangulation(const std::vector<dealii::Point<dim> >&, const std::vector<dealii::CellData<3>, std::allocator<dealii::CellData<3> > >&, const dealii::SubCellData&, dealii::Triangulation<3, spacedim>&) [with int spacedim = 3]
The violated condition was:
    false
Additional information:
    While trying to assign a boundary indicator to a line: the line with end vertices -1 and 0 does not exist.
--------------------------------------------------------

I attached the gmsh .geo file in this email. There is no vertex -1 actually. Could you tell me what's wrong with my mesh generation?

Thank you very much!

Best regards,

--

Yuesu Jin,

Ph.D student,

University of Houston,

College of Natural Sciences and Mathematics,

Department of Earth and Atmospheric Sciences,
Houston, Texas 77204-5008


346-404-2062

[Bruno Turcksin]

Yuesu,

I had the same problem recently. Our gmsh reader assumes that the user has set boundary id and material id to the mesh. It looks like you don't have those and so the file is read incorrectly.

Best,

Bruno

[yuesu jin]

Dear Bruno,

  Thank you for your help! I added physical groups for volume, surfaces, curves, and points. But the problem remains the same.  The modified geo file is attached below.

Best regards,

Yuesu

--
The deal.II project is located at http://www.dealii.org/
For mailing list/forum options, see https://groups.google.com/d/forum/dealii?hl=en
---
You received this message because you are subscribed to the Google Groups "deal.II User Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to dealii+un...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/dealii/5bf5008c-865d-47df-a49f-e4408028ef1dn%40googlegroups.com.

[Bruno Turcksin]

You don't want to set the physical groups for the curves and the
points. This means nothing in deal.II. A vertex does not have a
boundary ID. You need all the external surfaces and the volume to have
a physical group. That's it.

Bruno

> You received this message because you are subscribed to a topic in the Google Groups "deal.II User Group" group.
> To unsubscribe from this topic, visit https://groups.google.com/d/topic/dealii/HmoTVW7YfAw/unsubscribe.
> To unsubscribe from this group and all its topics, send an email to dealii+un...@googlegroups.com.
> To view this discussion on the web visit https://groups.google.com/d/msgid/dealii/CA%2B25a%3DLt6Dknw5DzkL0SCjozogiVoEVZQp%3DeurYpBWiUrCrLiQ%40mail.gmail.com.

[yuesu jin]

Dear Bruno,

   I did some modification according to your suggestion :
********************************************************
Physical Volume("4") = {1};
//+
Physical Surface("0") = {4, 3, 2, 5, 1, 6};

********************************************************

I add all external surfaces in one physical group and the volume in another group. But the error message remains the same.  I tried a few types of grouping combinations, the error message did not change.

Best regards,

Yuesu

To view this discussion on the web visit https://groups.google.com/d/msgid/dealii/CAGVt9eOsJoe4r_2t_eKVKMArv-qS5b93cScOxcivC0SP%3DVXz%3DQ%40mail.gmail.com.

[Bruno Turcksin]

Hmmm. Can you try without the SetFactory("OpenCASCADE"); and can you
send me your .msh

Bruno

> To view this discussion on the web visit https://groups.google.com/d/msgid/dealii/CA%2B25a%3DJAJ8cXUc7%3Dsw8UPCyFPys%3Dx98Ry%2B3H7fXE8szmzRB4Qg%40mail.gmail.com.

[yuesu jin]

Dear Bruno,

  The problem was solved after I removed the 'SetFactory("OpenCASCADE")'.  Here are two mesh files with and without 'SetFactory("OpenCASCADE")'. Could you tell me why this command made the system unable to work? Thank you very much for your help!

Best regards,

Yuesu

  

To view this discussion on the web visit https://groups.google.com/d/msgid/dealii/CAGVt9ePzJykr8OJOgrRqib_DJ4DhAVW2WMf2-UjOc8iZdCmnfQ%40mail.gmail.com.

[Bruno Turcksin]

Honestly I don't know. I wonder if the terms -1e-07 are the problem. I
told you to remove SetFactory("OpenCASCADE") because it was the only
difference with my .geo files.

Bruno

> To view this discussion on the web visit https://groups.google.com/d/msgid/dealii/CA%2B25a%3DJJ2WOKezdVURChx7FTLs1TJ685LpVMg8H8Ge%2BJFijZiA%40mail.gmail.com.

[yuesu jin]

Dear Bruno,

  Sorry for the late reply. Here is flooding, I was driving home hours before. You are right, in the linux version of gmsh, it always gives the error message that cannot find the corresponding vertex, but actually I do nothing on the cell. It is the 1e-7 rounding error that makes the system very unstable. Thank you again for your patience and kindness. Have a good day!

Best regards,

Yuesu

To view this discussion on the web visit https://groups.google.com/d/msgid/dealii/CAGVt9eO-iaz1tZJbynyz2pQ3fxoR7ATcXhep9DRk_XY5T-xdRg%40mail.gmail.com.