Remove advection from auxiliary quantities B_k #839
Conversation
2) Updated the auxiliary quantities (B_k in documented equations) so that they are no longer advected. We believe that these equations are identical to the advected quantities. The new auxiliary quantities provide the option to include state variables that are stationary in space. 3) Added a test case that verifies the new B_k equations work with MOL, Godunov, and EB's.
…ions.rst in documentation.
|
Two things: I am see some movement of one of the stationary scalars where it intersects the EB. Maybe something to do with state redistribution? But it isn't just immediately next to the surface.
Let's have this case run in the CI, similar to what we did for the flow over a cylinder case: #827 (review) |
…ill seeing some "transport" around the EB.
… #if NUM_AUX > 0). This should ensure that the x variable is not flagged as unused when NUM_AUX or NUM_ADV = 0 during tests.
…issue with diffusion occuring at EB interface.
|
No more movement near the EB! |
…e loops in Godunov, set NUM_AUX = NUM_ADV = 2 in CMakeLists.txt for CI tests.
Exec/RegTests/CMakeLists.txt
Outdated
| @@ -38,6 +38,9 @@ endif() | |||
| if(PELE_DIM EQUAL 2) | |||
| add_subdirectory(EB-FlowPastCylinder) | |||
| endif() | |||
| if(PELE_DIM EQUAL 2) | |||
There was a problem hiding this comment.
Another nitpick: just merge this with the if dim=2 conditional immediately above.
Tests/CMakeLists.txt
Outdated
| @@ -255,7 +255,10 @@ add_test_r(eb-c10 EB-C10) | |||
| add_test_rv(eb-c11 EB-C11) | |||
| add_test_rv(eb-c12 EB-C12) | |||
| if(PELE_DIM EQUAL 2) | |||
| add_test_r(eb-flowpastcylinder-re500 EB-FlowPastCylinder) | |||
| add_test_r(eb-flowpastcylinder-re500 EB-FlowPastCylinder) # | |||
There was a problem hiding this comment.
and here. Also remove the #
|
I removed |
This PR includes the following:
RegTeststhat illustrates how to utilize the advected and auxiliary quantities. I tested the updated code with Godunov, MOL, AMR, and a simple EB.