Contribution Proposal

[DhyeyAero95]

Contact we made for upstream is in your email titled "CSCI-5636 Project".

[connorhorn26]

"Pull Request"

The comparison between MORIS and MoFEM packages is attached below, also attached is CUBIT mesh input file for MORIS and MoFEM, MORIS input file, and MoFEM jupyter python script.

The MoFEM google mailing list has also been contacted regarding this comparison.

CSCI_5636.pdf
pullrequest.zip

[jedbrown]

Cool. The lack of convergence is concerning. Can you do a grid refinement study to see if MoFEM is converging at its design order of accuracy? Is MORIS converging to a wrong answer or not converging at all? I wonder if you can reduce that non-convergence phenomena.

How did you choose time steps?

[DhyeyAero95]

Hi Professor, There was an issue with Boundary condition in MORIS, we fixed it and getting better results. There was also an issue with MoFEM in terms of how heat flux is defined. We are getting better results. I can do grid refinement for MORIS but not sure about MoFEM because i tried on my local machine and ran into issue. Not sure if it was just a memory or something inherent in MoFEM. Thanks, Dhyey

…

On Wed, Nov 30, 2022 at 2:28 PM Jed Brown ***@***.***> wrote: Cool. The lack of convergence is concerning. Can you do a grid refinement study to see if MoFEM is converging at its design order of accuracy? Is MORIS converging to a wrong answer or not converging at all? I wonder if you can reduce that non-convergence phenomena. How did you choose time steps? — Reply to this email directly, view it on GitHub <#2 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/A2XP75BRMOW6OQTNGSL6E6TWK7IJLANCNFSM6AAAAAARYF53EQ> . You are receiving this because you authored the thread.Message ID: ***@***.***>

[jedbrown]

Okay, sounds good. I would compare time integrator and step size (match them if possible) and then solver configuration. You should be able to profile to see where the time is spent.

[DhyeyAero95]

"Pull Request" update:

As mentioned earlier, here are the updated report file and zip file consisting of important input and mesh and result files. In terms of solution field, result are very similar but MORIS outperforms MoFEM in terms of computation time however it depends on hardware we are using for both programs.

CSCI_5636.pdf
Pull_Request_Files.zip