"ODE" quantities and user defined external sources #427
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
…tiple ode vars. Still working on source terms.
…loops through levels.
… including external sources in SDC loop.
…th SDC iterations.
baperry2
requested changes
Oct 16, 2024
There was a problem hiding this comment.
Looks good. I'm able to run the test case and get the expected results. Just a couple minor changes requested.
One slightly larger thing is to have this run in the CI. You'll need to add it in Tests/CMakeLists.txt and you'll probably also need to add something to define NUM_ODE with the CMake build system similar to NUM_AUX in PeleC: https://github.com/AMReX-Combustion/PeleC/blob/b9ebe80bd3b00db0f50710048e50bac01212ef2b/CMake/BuildPeleExe.cmake#L25
And a clarification is that these quantities aren't spatially invariant, but don't advect or diffuse, so spatial terms don't come up in their governing equations.
dmontgomeryNREL
changed the title
baperry2
reviewed
Oct 18, 2024
![github-advanced-security[bot]](https://avatars.githubusercontent.com/in/57789?s=60&v=4){kind=small}
Comment on lines
+36
to
+56
| /* | ||
| Notes: | ||
| 1) ext_src contains sources from velocity forcing coming in. | ||
| This function should add to rather than overwrite ext_src. | ||
| 2) Requires peleLM.user_defined_ext_sources = true in input file | ||
|
|
||
| // Example: Exponential decay ode quantity | ||
| auto ext_src_arr = ext_src->arrays(); | ||
| auto const& state_old_arr = state_old.const_arrays(); | ||
|
|
||
| ParallelFor( | ||
| *ext_src, | ||
| [=] AMREX_GPU_DEVICE(int box_no, int i, int j, int k) noexcept { | ||
| for (int n = 0; n < NUM_ODE; n++) { | ||
| Real B_n = state_old_arr[box_no](i, j, k, FIRSTODE + n); | ||
| Real src = -1.0 * pow(10.0, n + 1) * B_n; | ||
| ext_src_arr[box_no](i, j, k, FIRSTODE + n) += src; | ||
| } | ||
| }); | ||
| Gpu::streamSynchronize(); | ||
| */ |
Check notice
Code scanning / CodeQL
Commented-out code Note
baperry2
approved these changes
Oct 22, 2024
terencelehmann
pushed a commit
to ITV-RWTH/PeleLMeX
that referenced
this pull request
Nov 14, 2024
* DivU constraint with arbitrary sources (AMReX-Combustion#428) * Included additional terms in divu constraint to account for external forces on density. * Updated Docs to include external source terms in model equations and divergence constraint. * Formatting * fix a bug in divu calc * Update deriviation of divu in Docs * Restored code to development * Reverted text in line 168 * Minor fix in Docs * Comment in divu about extRho --------- Co-authored-by: Bruce Perry <[email protected]> * "ODE" quantities and user defined external sources (AMReX-Combustion#427) * Initial commit. Needs debugging for source terms. * Added ode_var to typical values. Still debugging initialization and source. * Minor edit in set_ode_names function. * Fixed misalignment in plotting. IC's appear to be correct now for multiple ode vars. Still working on source terms. * Fix indexing. * Updated problem_modify_ext_sources. Works with local case definition. * Comment arguments in problem_modify_ext_sources in Source code. * Added compilation variable USE_MODIFIED_SOURCES to avoid unnecessary loops through levels. * Tested plotting for Efield and Spray. Tested all args of ProblemSpecificFunctions. * Added ode_srcstrength to test case, fixed minor bugs. * Changed flag for user defined external sources. Added future flag for including external sources in SDC loop. * Set peleLM.user_defined_ext_sources to true in input file. * Added Euler step for advancing the ODE quantities in time. * Changed name of advanceODEQty() to predictODEQty() for consistency with SDC iterations. * Updated input.2d for validation * Added .cpp file for ProblemSpecificFunctions to avoid need for static functions. * Added convergence test for ODE Qty. * Updated the test case to include 3 odes with increasing stiffness. * Renamed case * Combined all external forces in single function call to clean up PeleLMeX_Advance.cpp * Update README for case. * Update README for case. * Update README.md * Removed NUM_AUX for now. * Cleaning up * Documentation for ode quantities. * Clang-format * Remove validation tests from case directory. * Minor fix for soot and spray * Spelling and unused variable handling. * Unused variable handling * Formatting * Removed unwanted print statements, renamed some files/vars for consistency. * Added case to CI * Add USE_EB to EB_ODEQty/CmakeLists.txt * Add EB_ODEQty to RegTests/CMakeLists.txt * CI fix * More CI * Add ODEQty.cpp to cmake * More CI... * Only define predictODEQty if NUM_ODE > 0 * Updated CMake to handle user defined EB's and ProblemSpecificFunctions * Now passes in MultiFabs for state_old and state_new. This provides access to state_old.Factory() etc. * Update PeleLMeX_ProblemSpecificFunctions.cpp * Removed lev argument in problem_modify_ext_sources * Update comment --------- Co-authored-by: Bruce Perry <[email protected]> * pin intel-oneapi version in CI (AMReX-Combustion#434) * Bump Submodules/AMReX-Hydro from `3ab9864` to `3788361` (AMReX-Combustion#433) Bumps [Submodules/AMReX-Hydro](https://github.com/AMReX-Fluids/AMReX-Hydro) from `3ab9864` to `3788361`. - [Commits](AMReX-Fluids/AMReX-Hydro@3ab9864...3788361) --- updated-dependencies: - dependency-name: Submodules/AMReX-Hydro dependency-type: direct:production ... Signed-off-by: dependabot[bot] <[email protected]> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> Co-authored-by: Bruce Perry <[email protected]> * Add full manifold model capability (AMReX-Combustion#415) * basics of having eos_parm in PeleLMeX * add eosparm to PeleLMeX_K functions * more eosparm: now everywhere except Efield & BCs * eosparm in BC stuff * re-order functions for template * remove manifold-specfic changes, will go in later PR * gnu make changes for manifold * create manifold version of adjust fluxes function * Add abort for Closed Chamber + Manifold * Better setup chackes for Manifold + wbar, closed chamber, mixfrac/prog * advection of rho corrections for manifold models * manifold-specific transport, derives, divu * remove divide by zero * go to templated EOS functions to use cellData * fix oops in pelephysics * fix compile errors for Manifold derives * fix compile condition for transparm initialization * fix missing RY2RRinvY conversion * use PelePhysics that free manfunc shared pointer * clang-formatting * fux derives for manifold * add flamesheet test case for manifold * remove extraneous changes in file * Update Source/PeleLMeX_Setup.cpp Co-authored-by: Marc T. Henry de Frahan <[email protected]> * address Marc HdF's comments with minor formatting * remove copied line * update PP for merged RY2R stuff --------- Co-authored-by: Marc T. Henry de Frahan <[email protected]> * Update PelePhysics and add path to new utility (AMReX-Combustion#435) * Update PelePhysics and add new utility to Make.PeleLMeX * Add PelePhysics utilities to CMake * Test PelePhysics:tidy against clang-tidy. * Update PelePhysics submodule * Bump Submodules/AMReX-Hydro from `3ab9864` to `3788361` (AMReX-Combustion#438) Bumps [Submodules/AMReX-Hydro](https://github.com/AMReX-Fluids/AMReX-Hydro) from `3ab9864` to `3788361`. - [Commits](AMReX-Fluids/AMReX-Hydro@3ab9864...3788361) --- updated-dependencies: - dependency-name: Submodules/AMReX-Hydro dependency-type: direct:production ... Signed-off-by: dependabot[bot] <[email protected]> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * Bump Submodules/PelePhysics from `327aadb` to `f54b084` (AMReX-Combustion#439) Bumps [Submodules/PelePhysics](https://github.com/AMReX-Combustion/PelePhysics) from `327aadb` to `f54b084`. - [Release notes](https://github.com/AMReX-Combustion/PelePhysics/releases) - [Commits](AMReX-Combustion/PelePhysics@327aadb...f54b084) --- updated-dependencies: - dependency-name: Submodules/PelePhysics dependency-type: direct:production ... Signed-off-by: dependabot[bot] <[email protected]> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> --------- Signed-off-by: dependabot[bot] <[email protected]> Co-authored-by: Dave Montgomery <[email protected]> Co-authored-by: Bruce Perry <[email protected]> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> Co-authored-by: Marc T. Henry de Frahan <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This PR introduces a quantity that varies in time, but is neither advected or diffused. I'll refer to it as an
ode_qtyfor now. In PeleC these would beAUXquantities, but there already exists a placeholder forAUXin PeleLMeX that is not a state variable. I intend to rename the existing aux quantities toADVand these ode quantities toAUXfor consistency with PeleC, but that will be a large PR in itself.The ode quantities are included in the state variable and satisfy$\frac{\partial B_k}{\partial t} = S_{ext,B_k}$ , for $k = 0, 1,...,$NUM_ODE. Currently this is only first order accurate in time, but is consistent with the treatment of external sources from soot, spray and radiation, which are calculated prior to the SDC iterations. In the future I would like to add the ability to update the source terms within the SDC iterations, but this may require solving additional nonlinear systems depending on the source terms. For now we assume that the source terms are constant from $t^{n} \rightarrow t^{n+1}$ .
In addition to the ode quantities, this PR includes$S_{ext,\rho}$ , affect the velocity constraint but this has not been addressed in this PR. For reference, the velocity constraint should be:
$\nabla \cdot \vec{u} = \frac{1}{T}\frac{DT}{Dt} + W \displaystyle\sum_m \frac{1}{W_m} \frac{DY_m}{Dt} + \frac{1}{\rho} S_{\text{ext},\rho} = S.$
PeleLMeX_ProblemSpecificFunctions.cppwhich allows users to define the name of the ode quantities as well as external source terms for any state variable. Note that external source terms for density,Note that the existing external force terms for soot, spray and radiation are now included in the function called
getExternalSourcesinPeleLMeX_Forces.cpp.I tested 3 ode quantities of increasing stiffness with$S_{ext} = -10^{k+2} \cdot B_k$ , for $k = 0,1,2$ . The results below demonstrate first order accuracy and the need for SDC iterations in the future.