v23 picard changes #801
v23 picard changes #801
Conversation
| nekrs::finishStep(); | ||
| const PostprocessorValue * iter = &getPostprocessorValueByName("fp_iteration"); | ||
| std::cout << "Current fixed point iteration number read in NekRSProblemBase is " << *iter | ||
| << std::endl; // JUST FOR TESTING |
There was a problem hiding this comment.
Is this being kept in the final code? Comment indicates otherwise
There was a problem hiding this comment.
I think this should be kept in the final code. Depending on how the postprocessor for the iteration is passed into nek.i, it can be possible to mess it up (if the correct execute_on flags are not used), so I think this is necessary for the user to check that the correct iteration count is being received.
There was a problem hiding this comment.
Maybe the comment should be removed, otherwise everything looks good to me, thanks @regerdavid @aprilnovak
| { | ||
| converged = nekrs::runStep(corrector++); | ||
| } while (!converged); | ||
| // TODO: time is somehow corrected here |
There was a problem hiding this comment.
Has the TODO item been accomplished i.e. the time correction? @regerdavid
There was a problem hiding this comment.
I don't believe that that is my TODO comment (looks like it exists in the current master branch of cardinal) so I'm not sure if that is resolved or not.
| nekrs::finishStep(); | ||
| const PostprocessorValue * iter = &getPostprocessorValueByName("fp_iteration"); | ||
| std::cout << "Current fixed point iteration number read in NekRSProblemBase is " << *iter | ||
| << std::endl; // JUST FOR TESTING |
There was a problem hiding this comment.
| << std::endl; // JUST FOR TESTING | |
| << std::endl; |
|
As an update, further testing is needed to make sure that the fixed point iterations are working correctly for non-FSI applications. I was trying to use them with a CHT case today and was not getting expected behavior. |
|
Following up on my previous comment, the issue with CHT is related to a bug in NekRS that currently exists. This is supposedly fixed in next, but I have not gotten to try it yet |
aprilnovak
force-pushed
the
david_devel
branch
from
a313c7d
to
6ebd385
Compare
|
This is finally ready to consider for merging, now that the CHT bug has been fixed in NekRS's |
|
Job Documentation on 6ebd385 wanted to post the following: View the site here This comment will be updated on new commits. |
|
I'm not actively working on making a test for it, but I can try to come up with a simple test case for it. If there is one thing I'd like to see, its to get a change/addition in the execute_on flags in MOOSE so that we have one that will allow the Picard iterations to work properly without having to manually add an EXEC_CUSTOM flag. |
|
I think I have found a workaround for the |
Same PR as #790, but I did not want to overwrite @regerdavid's branch to split out into 2 PRs.