MultiApp transfer variable to initial condition of subapp #29676
Unanswered
garcs2
asked this question in
Q&A General
Replies: 1 comment 10 replies
-
|
hello Instead of using a ConstantIC, if you want a value from another application, you should transfer it. but here you should be good already?
since Griffin executes first and sends its power field to the HT solve, the HT solve has the updated power density already for its first solve? Guillaume |
Beta Was this translation helpful? Give feedback.
10 replies
-
|
I do not have contact with the Griffin team, though I can post to their discussion board. I thought this was more of a transfers issue not specific to Griffin which is why I made the post here. I am affiliated with UW-Madison. |
Beta Was this translation helpful? Give feedback.
-
|
Alright so try:
|
Beta Was this translation helpful? Give feedback.
-
|
Okay so I've done what you suggested, and I've made the Griffin execution verbose to track the solver. This is part of the output where you see that the power_source in HT is not updated but that the multiapp is being executed within the power iteration loop: I cut the output before and after it goes back to the power iteration in Griffin. In summary:
|
Beta Was this translation helpful? Give feedback.
-
|
so transfers is not often enough? You can try MULTIAPP_FIXED_POINT_END instead then |
Beta Was this translation helpful? Give feedback.
-
|
I tried MULTIAPP_FIXED_POINT_END and I'm still not getting the transfer to pass on properly, I also tried different configurations between TIMESTEP_END and MULTIAPP_FIXED_POINT_END between the main app and sub app and haven't gotten the transfers to function the way that I would like to. Given this issue I'm considering having the scheme shift in the other direction, where HT-TH solve first (with the power density set in HT) and then pass that information onto Griffin. I figure this would work as when I explicitly provide an IC HT-TH solve just fine, but I'm still curious why the transfer isn't solving the way that I was originally intending. |
Beta Was this translation helpful? Give feedback.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
Check these boxes if you have followed the posting rules.
Question
Hello everyone,
I am working on making a neutronics-heat transfer-thermohydraulics coupled simulation, using Griffin as the neutronics solver and heat transfer-thermohydraulics (HT-TH) being solved with the base modules in MOOSE. I have Griffin and HT-TH running successfuly separately. The HT-TH is solved using
TransientMultiAppwhere HT acts as the main app and TH is the sub-app. Ideally, the workflow would look like the following:For clarity, the
FullSolveMultiAppexecutes ontimestep_endso that Griffin solves first, then transfers the power density to HT-TH where HT solves first. My question is then: how do I designate the initial condition of HT to be the transferred power density? Since the power density is a scalar, I would like to use ConstantIC but a value is required as input. Below is the relevant portion of my input file, with the IC block not filled in since I don't know what to do:Beta Was this translation helpful? Give feedback.
All reactions