open TELEMAC-MASCARET The mathematically superior suite of solvers

Hello,

I ran the simulation with the Formatted Data File 1 and 2 successfully with 1 processor. But when I run it in parallel, boundary conditions file seemed wrong processed.

The error information was attached. What is the reason for this error?
Thanks a lot!

Here is the boundary condition files I used.
It's generated by BlueKenue. Is there anything wrong with this?

Hello,

It could be that a mismatch between geometry and the boundary conditions file is seen only when partitioning the geometry ? That is te most probable.

JMH

Yes, I think it's related to the mesh file.
Now I'm running another case with a more uniform mesh.
This problem was gone. But I don't know what is the exact reason that cases this problem.

If this problem is reproducible, please post your mesh file (slf or t3s) and the (bc2) boundary conditions file here so I can analyse any discrepancy that might exist.

Martin

Thanks Serrer.

Enclosed are the files I used for the simulation.
I have two versions of this simulation. One has a finer mesh and the other one has a coarse grid but with a very detailed harbor. The finer mesh works OK in parallel. But the coarse one has this problem.

This is the coarse one. I made it by BlueKenue.

Can you please send the bc2 file?

You have defined:
1. a level boundary (code:544) on the river.
2. a level boundary (code:544) on the offshore
3. and the rest of the boundary nodes as (code:200)

Was that your intention?

Jean-Michel, what is the meaning of LIHBOR=2,LIUBOR=0,LIVBOR=0 ??

Martin

Hello,

2 0 0 means a solid boundary with no slip condition (velocity=0) on this boundary.

JMH

Hi Martin,

Enclosed is the bc2 file and the corresponding cli file.

Yes, I have two liquid boundaries (sea boundary and river boundary). Both of them have the code 544 because I imposed water levels at these boundaries.

The other two boundaries are closed boundaries. I don't want to get any velocity at the wall so I used code 200.