[DoctorRobert]

I am trying to run an analysis that involves multizone fluid-fluid flow with species. To create a simple test case I took the .su2 mesh file for the incompressible flow/species transport tutorial and duplicated and shifted one mesh to create a simple multizone case. The problem works beautifully without species transport but when I add that in the code bombs in the setup phase and I get the following error message. Can anyone help me out here?

Error in "virtual void CMultizoneOutput::SetMultizoneHistoryOutputFields( const COutput* const*, const CConfig* const*)":
-------------------------------------------------------------------------
Per Surface output group unknown: SPECIES_COEFF_SURF
------------------------------ Error Exit -------------------------------

[bigfootedrockmidget]

Hi,


We have not implemented/tested this keyword for multizone yet.

You could first try by removing the SPECIES_COEFF_SURF keyword from the config file. Do you need to optimize a species related objective?

[DoctorRobert]

I never included the term in the cfg files because I didn't know it was needed. Should I be defining it? I am not trying to optimize anything species related. I'm trying multizone because I want to use a body force in certain zones as a crude but simple way to mimic a porous media.

[bigfootedrockmidget]

I did a quick test. If I run the CHT case:
https://su2code.github.io/tutorials/Static_CHT/
and add species transport, I do not get this message. I just add this to the flow_cylinder.cfg fluid setup:

% -------------------- SCALAR TRANSPORT ---------------------------------------%
%
KIND_SCALAR_MODEL= SPECIES_TRANSPORT
DIFFUSIVITY_MODEL= CONSTANT_DIFFUSIVITY
DIFFUSIVITY_CONSTANT= 0.001
%
CONV_NUM_METHOD_SPECIES= SCALAR_UPWIND
MUSCL_SPECIES= NO
SLOPE_LIMITER_SPECIES = NONE
%
TIME_DISCRE_SPECIES= EULER_IMPLICIT
%
SPECIES_INIT= 1.0, 0.0
SPECIES_CLIPPING= YES
SPECIES_CLIPPING_MIN= 0.0, 0.0
SPECIES_CLIPPING_MAX= 1.0, 1.0

Maybe you can check what the difference in setup is.


Can you share your setup somewhere? If you can create a bug report on github with a case that reproduces this error, we can look into it.

https://github.com/su2code/SU2/issues

[DoctorRobert]

The problem comes when you specify MARKER_ANALYZE to compute mass flux . However when you remove that there is no communication of species information across the interface. as can be seen in the following images of velocity, species_0, and species_1

[bigfootedrockmidget]

Hi,



OK, there were actually 2 problems here. One is that the species mass fractions were not transported across the zone interface. The other is that for MARKER_ANALYZE, we did not add species information, but later in the code we assumed that it was present.



Both are now fixed in the feature branch, but it will be a while before it ends up in the master version. In the mean time, you could clone the SU2 git repository and then do a git checkout feature_species_multizone and compile the fixed branch yourself.
Thanks for spotting the problem!

[bigfootedrockmidget]

By the way, in the develop branch, species transport has been extended with species dependent properties. the keyword PASSIVE_SCALAR has therefore also been replaced by SPECIES_TRANSPORT
If you clone the code from github, you can find some examples in the Testcases/species_transport directory