Problems of my own benchmark example - LIE Single Fracture

Usability
1

Hello,

I got the following message which caused OGS6 not to run.

image
image1735×621 27.1 KB

Most likely, the problem is caused by the SF.vtu file (my mesh file). The following shows my mesh, there is a lower-dimensional interface element or LIE fracture in the middle domain. I got the correct .msh file (gmsh v2 ascii) from GMSH (attached in the .zip file). However, when I call GMSH2OGS.exe to convert .msh file to .vtu file, I might lost the LIE fracture elements. This could be why I had the message " [LIE/HM] looking for fracture elements in the given mesh". Any help is greatly appreciated.

image
image757×755 33.8 KB

FYI, I attached the geometry (.gml), mesh (.vtu), and project file (.prj) for easier troubleshooting.

Thank you,
WenfengM_SingleFracture.zip (7.5 MB)

2

If you use ‘-e’ (exclude) argument when you run GMSH2OGS, it will exclude line elements.
If this is the case, the mesh would’ve lost the LIE…

3

Thank you for your reply, Keita.

I did not use ‘-e’ when I run GMSH2OGS, so the line elements are still in the .vtu file.
I sent emails to Nori and he kindly helped out. To properly define LIE, I need to use quadratic elements because deformation requires higher-order shape functions. This key info was not documented anywhere for use of LIE in OGS.