Hi,
I did a combustion simulation using the species transport model in my PC where Ansys 2020R2 is installed. I used a Chemkin mechanism and Chemkin-CFD solver for the chemistry part. The simulation went pretty well and I got a satisfactory result.
When I tried to do the same in batch mode on an HPC where Ansys 19.2 is installed, I found the continuity starts diverging right after the simulation started, and the simulation stops after a few iterations.
I prepared my journal file as follows:
The initialization part is done on my desktop PC. After that, I run the simulation (steady) for 10 iterations. Then, I write the case&data file. In the journal file, the case and data files are read first. Then, I set a new iteration number (100). Next, I write the new data file.
N.B: Before doing so, I tried a very simple fluid flow problem on both my local PC and HPC. The results looked nearly the same in that case.
Thanks!
Hi,
ANSYS Fluent developers are constantly working on the solver implementation, so your observation with respect to deteriorated convergence behavior in an earlier ANSYS Fluent version (in comparison to 2020.R2) is not so much of a surprise. You can only compare the behavior of ANSYS 2020.R2 in GUI and batch mode. This should be fairly identical. By reverting to ANSYS Fluent 19.2 your are essentially missing a lot of solver improvements of the past 2 years. In particular in the area of combustion and mixture transport development work has been done in ANSYS Fluent over the recent years.
I even wonder on how you manage to get a CAS file from 2020.R2 to be read into a 19.2 Fluent solver. This kind of backward compatibility cannot be taken for granted and usually it results in errors.
Regards,
Dr. Th. Frank.
Dr. Frank,
Thank you so much for your insights.
I get your point and work on it. Actually, I was also a bit confused if there would be an issue of compatibility to read the case and data file. However, it worked well.
Shaon
- Badges
- Science topic
- Similar topics
- Social Science
- Media
- Journalism