Species transport solver (passive for now) #1388

TobiKattmann · 2021-10-02T09:26:46Z

Proposed Changes

Hi all,

This PR implements a passive species transport solver. The passive means, that the mass fraction at each location does not influence any fluid-properties and thereby the mean flow (which reminds me that this is a good sanity check to compare mean flow residuals).

A new CSpeciesSolver and CSpeciesVariable were added based on the CScalar* things done in #1330 . Additionally species_convection/diffusion/sources were added just as it is the case for the turbulence solver (both based on the general scalar* framework)

Please note that prob ~1000 lines are cfg's, README's, helping scripts.

Related Work

Preliminary work was done in #1330 with creating a base class for scalar transport equations.

#1223 and #1332 already mostly implement the here shown species solver and big chunks were copied over from there, but they are not based on the new CScalarSolver. In order to chunk the work for these branches this PR will not contain any species mixing properties or TableLookUps.

After this PR #1332 will be tackled next. Whether we try to merge develop then into those branches or copy stuff over in a new branch will be seen when we get there.

PR Checklist

I am submitting my contribution to the develop branch.
My contribution generates no new compiler warnings (try with the '-Wall -Wextra -Wno-unused-parameter -Wno-empty-body' compiler flags, or simply --warnlevel=2 when using meson).
My contribution is commented and consistent with SU2 style.
I have added a test case that demonstrates my contribution, if necessary.
I have updated appropriate documentation (Tutorials, Docs Page, config_template.cpp) , if necessary.

Note that sources are empty for now. Axissymmetric source terms will be added later.

Right now, checking of these options is missing. Options also have to be added to the config_template once everything else is finalized.

Please note that Multizonedriver is not tested yet. Single and Multizonedriver get SetInitialCondition function which might be superfluous - needs checking.

The output should be able to handle arbitrary many species and should prob be generalized for comp and inc solvers to avoid code duplication.

pcarruscag

That was quick 🥇

Common/include/CConfig.hpp

SU2_CFD/src/solvers/CSpeciesSolver.cpp

SU2_CFD/include/variables/CSpeciesVariable.hpp

Common/include/CConfig.hpp

Common/src/CConfig.cpp

Plus a tiny cfg change.

SU2_CFD/include/solvers/CSpeciesSolver.hpp

SU2_CFD/include/variables/CSpeciesVariable.hpp

SU2_CFD/src/drivers/CDriver.cpp

Deleting empty destructors Const-ing some code parts OMP ready container positions in CDriver Co-authored-by: Pedro Gomes <38071223+pcarruscag@users.noreply.github.com>

Removing unnecessary (void) Const-ing some return pointers Renamming enum class entry NO_SCALAR_MODEL to NONE

TobiKattmann · 2021-11-29T22:56:18Z

Alright, I think I am finished with this PR. Big thanks so far pedro, for all the comments, help, code suggestions and contributions 🤝 In case you have some final comments let me know

pcarruscag

Just minor comments below on the recent changes, very nice addition 🥇

SU2_CFD/include/output/CFlowOutput.hpp

SU2_CFD/src/output/CFlowOutput.cpp

pcarruscag · 2021-11-30T14:59:01Z

TestCases/species_regression.py

-    species3_primitiveVenturi.test_vals = [-6.028145, -5.258104, -5.107927, -5.922051, -1.582604, -14.539029, -14.809699, 5, -0.808615, 5, -2.351160, 5, -0.288300, 1.645644, 0.499064, 0.601230, 0.545351]
+    species3_primitiveVenturi.test_vals = [-6.028145, -5.258104, -5.107927, -5.922051, -1.582604, -6.314220, -6.431771, 5, -0.808615, 5, -2.351160, 5, -0.288300, 1.645644, 0.499064, 0.601230, 0.545351]


Is any of the tests using the compressible solver?

no, I don't have one. Tbh i never even ran a compressible case 🙊 I'll give it a try and report

ok i adatpted the rans/flatplate_SA to include some species inlet (with manually adapted inlet file to include something funny to happen). 0d37b2a I did not adapt the cfg too crazy to stay close to the original one. That is a task for another day

SU2_CFD/src/output/CFlowOutput.cpp

pcarruscag · 2021-11-30T15:09:23Z

SU2_CFD/src/output/CFlowOutput.cpp

@@ -641,96 +756,13 @@ void CFlowOutput::SetAnalyzeSurface(const CSolver* const*solver, const CGeometry
  }

  /*--- Set values on markers ---*/
-  for (iMarker_Analyze = 0; iMarker_Analyze < nMarker_Analyze; iMarker_Analyze++) {
+  for (unsigned short iMarker_Analyze = 0; iMarker_Analyze < nMarker_Analyze; iMarker_Analyze++) {
    su2double SpeciesVariance = Surface_SpeciesVariance_Total[iMarker_Analyze];
    SetHistoryOutputPerSurfaceValue("SURFACE_SPECIES_VARIANCE", SpeciesVariance, iMarker_Analyze);
    Tot_Surface_SpeciesVariance += SpeciesVariance;
    // Set value into config. Necessary to access as an OF.


It's actually not needed, you can get it from the output, check if you can make the config const please.

I am currently following the template which is used to set the OF for the DA. I actually used the Config to set the variance OF for this #1388 (comment) below. So I keep that in mind for the adjoint PR that will come next

So I would postpone this thought to the species adjoint PR

.github/workflows/regression.yml

Co-authored-by: Pedro Gomes <38071223+pcarruscag@users.noreply.github.com>

TobiKattmann · 2021-11-30T15:56:01Z

I also just made a gradient validation with the species variance OF ... will have the small additions in a follow up PR.
Small and easy case but with SST and 2 species transport equations

SURFACE_SPECIES_VARIANCE gradient 
+-----------+-------------------+-------------------+-------------------+-------------------+-------------+
| DV number |       DA gradient |       FD gradient |     absolute diff | relative diff [%] | sign change |
+-----------+-------------------+-------------------+-------------------+-------------------+-------------+
|         0 |     -0.0015565547 |     -0.0015568653 |      0.0000003106 |      0.0199523955 |           0 |
|         1 |     -0.0002704356 |     -0.0002704940 |      0.0000000584 |      0.0215893397 |           0 |
|         2 |      0.0005546628 |      0.0005546735 |     -0.0000000107 |     -0.0019320096 |           0 |
|         3 |      0.0008743823 |      0.0008743497 |      0.0000000325 |      0.0037214309 |           0 |
|         4 |      0.0008629093 |      0.0008627899 |      0.0000001194 |      0.0138354969 |           0 |
|         5 |     -0.0015565547 |     -0.0015568653 |      0.0000003106 |      0.0199522614 |           0 |
|         6 |     -0.0002704356 |     -0.0002704940 |      0.0000000584 |      0.0215901515 |           0 |
|         7 |      0.0005546628 |      0.0005546735 |     -0.0000000107 |     -0.0019318288 |           0 |
|         8 |      0.0008743823 |      0.0008743497 |      0.0000000325 |      0.0037199429 |           0 |
|         9 |      0.0008629093 |      0.0008627899 |      0.0000001194 |      0.0138353462 |           0 |
+-----------+-------------------+-------------------+-------------------+-------------------+-------------+

I only consider the y-direction of the points and and fixed the first two "columns" to have a smooth transition. Additionally during deformation i have the outlet as MARKER_SYM for smooth Volume mesh morphing , i.e. only the wall is in DV_MARKER. This Gradient validation was brought to you by: FADO :)

…ecies

TestCases/rans/flatplate/turb_SA_flatplate_species.cfg

lgtm-com · 2021-12-01T11:53:02Z

This pull request introduces 1 alert and fixes 2 when merging d7757db into 42dcf5a - view on LGTM.com

new alerts:

1 for Resource not released in destructor

fixed alerts:

2 for Resource not released in destructor

TobiKattmann added 11 commits October 2, 2021 11:02

Add CSpeciesSolver files. Child of CScalarSolver.

dcae5d5

Add CSpeciesVariable. Child of CScalarVariable.

15ccc6e

Add species_convection/diffusion/sources. Children of scalar_*.

c11181e

Note that sources are empty for now. Axissymmetric source terms will be added later.

Add new Species files to both (meson, automake) buildsystems.

d77d33c

Add SpeciesSolver to SolverFactory.

07a8b64

Add necessary new options to Config and option_*.

459350b

Right now, checking of these options is missing. Options also have to be added to the config_template once everything else is finalized.

Smaller additional changes to ScalarSolver.

5982908

Add species contributions to Driver level. Filling numerics containers.

937b9a0

Please note that Multizonedriver is not tested yet. Single and Multizonedriver get SetInitialCondition function which might be superfluous - needs checking.

Add Species Iteration to FluidIteration.

88779c5

Add species output for inc solver and for 1 species only.

f141acd

The output should be able to handle arbitrary many species and should prob be generalized for comp and inc solvers to avoid code duplication.

Merge branch 'develop' of https://github.com/su2code/SU2 into develop

53ea78f

pr-triage bot added the PR: unreviewed label Oct 2, 2021

TobiKattmann changed the title ~~Species transport solver,~~ [WIP] Species transport solver (passive for now) Oct 2, 2021

pr-triage bot removed the PR: unreviewed label Oct 2, 2021

TobiKattmann marked this pull request as draft October 2, 2021 09:27

pr-triage bot added the PR: draft label Oct 2, 2021

TobiKattmann mentioned this pull request Oct 2, 2021

Feature species transport su2code/TestCases#78

Merged

TobiKattmann added 2 commits October 2, 2021 12:14

Add passive species transport testcase.

42a76ef

Quick fix to make regression tests work.

e5af296

pcarruscag reviewed Oct 2, 2021

View reviewed changes

TobiKattmann commented Oct 2, 2021

View reviewed changes

Common/include/CConfig.hpp Show resolved Hide resolved

TobiKattmann commented Oct 2, 2021

View reviewed changes

Common/include/CConfig.hpp Show resolved Hide resolved

TobiKattmann commented Oct 2, 2021

View reviewed changes

Common/src/CConfig.cpp Outdated Show resolved Hide resolved

Remove memory leak in species_sources.

8709766

Plus a tiny cfg change.

TobiKattmann commented Oct 3, 2021

View reviewed changes

SU2_CFD/include/solvers/CSpeciesSolver.hpp Show resolved Hide resolved

TobiKattmann commented Oct 3, 2021

View reviewed changes

SU2_CFD/include/variables/CSpeciesVariable.hpp Outdated Show resolved Hide resolved

TobiKattmann commented Oct 3, 2021

View reviewed changes

SU2_CFD/src/drivers/CDriver.cpp Outdated Show resolved Hide resolved

TobiKattmann commented Oct 3, 2021

View reviewed changes

SU2_CFD/src/drivers/CDriver.cpp Outdated Show resolved Hide resolved

TobiKattmann and others added 2 commits October 3, 2021 11:26

Apply suggestions from code review

6b003c6

Deleting empty destructors Const-ing some code parts OMP ready container positions in CDriver Co-authored-by: Pedro Gomes <38071223+pcarruscag@users.noreply.github.com>

Cleaning some code bits after review.

34023cc

Removing unnecessary (void) Const-ing some return pointers Renamming enum class entry NO_SCALAR_MODEL to NONE

su2code deleted a comment from lgtm-com bot Nov 29, 2021

TobiKattmann added 2 commits November 29, 2021 19:17

Cleanup

5daa05e

Rename Species_inf to CScalars Solution_inf

9653fc3

TobiKattmann added the changelog:feature label Nov 29, 2021

Cleanup 2

c7c9dfb

su2code deleted a comment from lgtm-com bot Nov 29, 2021

pcarruscag approved these changes Nov 30, 2021

View reviewed changes

pcarruscag reviewed Nov 30, 2021

View reviewed changes

.github/workflows/regression.yml Outdated Show resolved Hide resolved

Func input by ref, remove unnecessary nDim definition

07f22c0

Co-authored-by: Pedro Gomes <38071223+pcarruscag@users.noreply.github.com>

TobiKattmann added 2 commits November 30, 2021 19:15

Move species specific reg tests to parallel_regression.

d3b1914

Merge remote-tracking branch 'origin/feature_species' into feature_sp…

5ae82d7

…ecies

su2code deleted a comment from lgtm-com bot Nov 30, 2021

TobiKattmann added 2 commits November 30, 2021 21:49

Add a compressible species Testcase

0d37b2a

Merge remote-tracking branch 'origin/develop' into feature_species

0a21059

su2code deleted a comment from lgtm-com bot Nov 30, 2021

su2code deleted a comment from lgtm-com bot Dec 1, 2021

TobiKattmann commented Dec 1, 2021

View reviewed changes

TestCases/rans/flatplate/turb_SA_flatplate_species.cfg Outdated Show resolved Hide resolved

Update TestCases/rans/flatplate/turb_SA_flatplate_species.cfg

d7757db

su2code deleted a comment from lgtm-com bot Dec 1, 2021

TobiKattmann merged commit 1249f9f into develop Dec 1, 2021

TobiKattmann deleted the feature_species branch December 1, 2021 13:06

pr-triage bot added PR: merged and removed PR: unreviewed labels Dec 1, 2021

This was referenced Dec 2, 2021

Feature species - more adjoint parts #1461

Merged

Species transport theory + tutorials su2code/su2code.github.io#64

Merged

TobiKattmann mentioned this pull request Jan 19, 2022

[WIP] Develop for transition model. #1496

Closed

5 tasks

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Species transport solver (passive for now) #1388

Species transport solver (passive for now) #1388

TobiKattmann commented Oct 2, 2021 •

edited

Loading

pcarruscag left a comment

TobiKattmann commented Nov 29, 2021

pcarruscag left a comment

pcarruscag Nov 30, 2021

TobiKattmann Nov 30, 2021

TobiKattmann Nov 30, 2021

pcarruscag Nov 30, 2021

TobiKattmann Nov 30, 2021

TobiKattmann Nov 30, 2021

TobiKattmann commented Nov 30, 2021 •

edited

Loading

lgtm-com bot commented Dec 1, 2021

		species3_primitiveVenturi.test_vals = [-6.028145, -5.258104, -5.107927, -5.922051, -1.582604, -14.539029, -14.809699, 5, -0.808615, 5, -2.351160, 5, -0.288300, 1.645644, 0.499064, 0.601230, 0.545351]
		species3_primitiveVenturi.test_vals = [-6.028145, -5.258104, -5.107927, -5.922051, -1.582604, -6.314220, -6.431771, 5, -0.808615, 5, -2.351160, 5, -0.288300, 1.645644, 0.499064, 0.601230, 0.545351]

Species transport solver (passive for now) #1388

Species transport solver (passive for now) #1388

Conversation

TobiKattmann commented Oct 2, 2021 • edited Loading

Proposed Changes

Related Work

PR Checklist

pcarruscag left a comment

Choose a reason for hiding this comment

TobiKattmann commented Nov 29, 2021

pcarruscag left a comment

Choose a reason for hiding this comment

pcarruscag Nov 30, 2021

Choose a reason for hiding this comment

TobiKattmann Nov 30, 2021

Choose a reason for hiding this comment

TobiKattmann Nov 30, 2021

Choose a reason for hiding this comment

pcarruscag Nov 30, 2021

Choose a reason for hiding this comment

TobiKattmann Nov 30, 2021

Choose a reason for hiding this comment

TobiKattmann Nov 30, 2021

Choose a reason for hiding this comment

TobiKattmann commented Nov 30, 2021 • edited Loading

lgtm-com bot commented Dec 1, 2021

TobiKattmann commented Oct 2, 2021 •

edited

Loading

TobiKattmann commented Nov 30, 2021 •

edited

Loading