Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Urom and Umag at a final time step looks different #18

Open
chandukec opened this issue Mar 13, 2022 · 4 comments
Open

Urom and Umag at a final time step looks different #18

chandukec opened this issue Mar 13, 2022 · 4 comments

Comments

@chandukec
Copy link

Hello,
Many thanks for this utility. I tried to use this utility for the sample case that you have provided, when I tried to run podROM 0 it gave me some display output but at the final line it says segmentation fault . I ignored that went with the mpirun -np 4 podFlowReconstruct -time 0.5:15 -parallel. But when I compared the urom and umag velocity fields at the final time step, they looks different-- the screenshot is attached herewith. Is that fine, or I am doing something terribly wrong? Any comment/suggestion will be a great help, thanks in advance.
urom_umag_compare

@skyabc355
Copy link

@tpgrunloh @tjwilso @akashaero, Hello,Developers:
Thanks for this POD utility,it is very powerful.
I have seen the same error message as chandukec,that is to say,when use "podROM 5",I saw the "segmentation fault",too.
I have tried the following things to avoid this:
1,At first,I was using OpenFoam 4 under CentOS 7.8.When I saw this error,I thought there maybe something with the Operating System as CentOS 7.8 is not in the supported OS list according to the README.md.
Because of this "segmentation fault" message,I thought the results have been polluted.
chandukec's finding validated this.
2,I then erased my computer completly and installed Ubuntu 16.04 and OpenFoam 7.They are both in the supported list of README.md,however,I saw the same "segmentation fault" again.
I noticed that a c++11 compiler was required according to your instructions.
The default GCC/G++ version on Ubuntu 16.04 is 5.4,which support C++11.
I also tried downgrade GCC/G++ to 4.8 and 4.9,however,this made things worser.
Even "podBasisCalc 5 -time 0.0:15.0" did not work.
I also tried to use Intel C++(icc+) to compile the code instead of using G++/GCC,there was no success,neither.
By the way,when I firstly compiled the code under Ubuntu 16.04,I saw many warnings,for example:
warning: comparison between signed and unsigned integer expressions [-Wsign-compare]
warning: unused variable ‘sizeOfString’ [-Wunused-variable]
warning: use of old-style cast [-Wold-style-cast]
......
Please check the attached AccelerateCFD_Compile.log for more details.
I hope the descriptions could help diagnose this issue.
AccelerateCFD_Compile.log
Thanks again for your excellent POD program.

@akashaero
Copy link
Contributor

Let me take a look at this. Will get back to you soon!

@chandukec
Copy link
Author

Hello @skyabc355 Many thanks for acknowledging my issues. But I figured this out correctly using Ubuntu 20.04.4 LTS and OpenFOAM v1906. The issue could be that the inconsistency in using the <# of basis> for different commands that you have used. And moreover I increased the value of nDim from 5 to 15 as: nDim nDim [0 0 0 0 0 0 0] 15; // 5 //original in the constant/podDict file. And after that, if I match the urom and umag they are exact.

@skyabc355
Copy link

Hello@chandukec@akashaero,Thanks for your replies.I have successfully used this POD utility with Ubuntu 20.04 and OpenFOAM 7.Everything is working as expected.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

3 participants