[Secchialessio]

Hello all


I'm learning how to use SU2, and I decided to experiment a bit.
I'm trying to simulate a question received during the lecture of internal flow (i'll attach a screenshot of the question).
My strategy was to start with the 2D model and later go for the 3D one.


The mesh I create seems to work, as I tried a simple case with uniform velocity.
I based my velocity profile on the example Inc_Laminar_Step so I'm quite sure it's ok.
Nonetheless when I try to input the velocity as external file it gives me the following error:


Reading inlet profile from file: velocity_ref.dat
No Inlet Interpolation being used
WARNING: Did not find a match between the points in the inlet file
and point 123 at location: [-4.000000e+00, 3.000000e+00]
Distance to closest point: 4.000312e+00
Current tolerance: 1.000000e-06

You can increase the tolerance for point matching by changing the value
of the option INLET_MATCHING_TOLERANCE in your *.cfg file.


Error in "void CSolver::LoadInletProfile(CGeometry**, CSolver***, CConfig*, int, short unsigned int, short unsigned int) const":
-------------------------------------------------------------------------
Prescribed inlet data does not match markers within tolerance.
------------------------------ Error Exit -------------------------------


--------------------------------------------------------------------------
MPI_ABORT was invoked on rank 2 in communicator MPI_COMM_WORLD
with errorcode 1.

NOTE: invoking MPI_ABORT causes Open MPI to kill all MPI processes.
You may or may not see output from other processes, depending on
exactly when Open MPI kills them.
--------------------------------------------------------------------------
[Alessio:14927] 3 more processes have sent help message help-mpi-api.txt / mpi-abort
[Alessio:14927] Set MCA parameter "orte_base_help_aggregate" to 0 to see all help / error message


I'm attaching the questions (I'm still stuck with the first anyway) and the configuration file.
If needed I can upload the mesh as well.

Thanks for any help

[bigfootedrockmidget]

How did you generate the inlet file? If you do not have one, SU2 will generate a default file for you filled with the values from the uniform boundary conditions, and this file should work out of the box. Does this work for you?

[Secchialessio]

Originally Posted by bigfootedrockmidget

How did you generate the inlet file? If you do not have one, SU2 will generate a default file for you filled with the values from the uniform boundary conditions, and this file should work out of the box. Does this work for you?

Hi, thanks for answering.
I generated the inlet file as velocity.dat in an ASCII format following the example from Inc_Laminar_Step.
Even though some points are actually slightly different (e.g. point 573 x=-4, y=2.950000000000586 in Mesh File, while in Input file x=-4, y=2.9500) the point specified by the error (123) is exactly x=-4 y=3 so i don't know what the problem is.
I don't think I should refine the mesh as this one is already a refined version of my original and the problem is the same.
The file that it generates has only 31 rows (as before I refined the mesh) but i will try and let you know

Thank you for your apprehension

EDIT:
For some reason, I don't get the example file anymore, but at the end I realized there was an error in the input file indeed, nonetheless, even after correcting it, I was getting the same error but "WARNING: Did not find a match between the points in the inlet file
and point 123 at location: [-4.000000e+00, 3.000000e+00]
Distance to closest point: 5.00000e-02"
So I changed the Inlet tolerance to 5e-2 which is definitely not ideal, as the simulation took a lot of time, but after 10021 iterations it converged
NEXT STEP, I want to make the 3D example of question, let's see how it goes :/

[bigfootedrockmidget]

Hi, Can you share the mesh and inlet profile file? I can have a look then. Maybe it's a problem with the comparison and the sign of the coordinates.

Preferably a small mesh, since some people share 300Mb meshes and I cannot even load it without going to an HPC :-)

[Secchialessio]

Hi thanks a lot for your effort.
I posted the mesh with the 0.05 error, and the semi-structured mesh, which has even an higher tolerance....
The result from the unstructured mesh is also incorrect.

https://drive.google.com/drive/folde...B6?usp=sharing

[bigfootedrockmidget]

Hi,



This works fine for me. Are you using a very old version of SU2?
The only thing I do is run

SU2_CFD config_template_v2.cfg

and then I get the file example_velocity.dat. I rename the example_velocity.dat file to velocity.dat and rerun SU2_CFD, and it works as expected.
Maybe you made an error when you edited the example_velocity.dat file? You mentioned 31 rows but my generated example_velocity.dat has 61 rows