Implement file and database IO functions in `OpenMCDepcode` #189

yardasol · 2023-02-01T18:22:03Z

This PR adds full support for OpenMC by implementing file and database IO functions in OpenMCDepcode.

To my reviewers: Please do not panic at the large change size! Most of these are .h5 files! You do not need to look at them!

Note that for depletion simulations, the total mass density of the materials does not change in Serpent whereas in OpenMC it does.

Summary of changes

Specifically, this PR:

Implements a run_depletion_step() method to Depcode to reduce code repetition.
Adds the following abstract methods to Depcode:
- convert_nuclide_code_to_name()
- _convert_name_to_nuccode(), which is implemented in both SerpentDepcode and OpenMCDepcode
- preserve_simulation_files(), which preserves all input and output files for each depletion step.
Adds time constants to the top of app.py
Adds a code block in app.py to process the depletion_settings argument for OpenMC-coupled runs.
Adds the following functions in SerpentDepcode:
- resolve_include_paths(), that converts relative paths to absolute paths in include cards.
- _convert_name_to_nuccode, which is a helper function to assist in converting nuclide names to nuclide codes.
Adds the following functions in OpenMCDepcode:
- _check_for_material_names() which fails fast if a material doesn't have a name.
- Implementation of read_step_metadata
- _find_xs_path(), a helper function for read_step_metadata
- Implementation of read_neutronics_parameters()
- _calculate_breeding_ratio(), a helper function for read_neutronics_parameters()
- _calculate_delayed_quantity() a helper function for read_neutronics_parameters()
- _get_values_with_uncertainties(), a helper function for _calculate_delayed_quantity()
- _calculate_fission_masses(), a helper function for read_neutronics_parameters()
- Implementation of read_depleted_materials()
- _create_mass_percents_dictionary(), a helper function for read_depleted_materials()
- _convert_nucname_to_pyne(), a helper function for _create_mass_percents_dictionary()
- _get_power_from_tallies(), a helper function for read_depleted_materials()
- Implementation of convert_nuclide_code_to_name()
- _convert_name_to_nuccode()
- Implementation of update_depletable_materials()
- Overhaul of write_saltproc_openmc_tallies()
- _get_fissile_fertile_nuclides(), a helper function for write_saltproc_openmc_tallies()
Adds Simulation._add_missing_nuclides(), a function that enables expansion of previously stored results. We need this functionality due to the presence of decay-only nuclides, which are not always stored in the depletion matrix and thus the results.
Overhaul to existing openmc tests, mainly due to using the MSBR geometry instead of the TAP geometry
Updates to test suite to account for new functionality
Consistency changes

Types of changes

Bug fix (non-breaking change which fixes an issue)
New feature (non-breaking change which adds functionality)
Breaking change (fix or feature that would cause existing functionality to change)

Required for Merging

Associated Issues and PRs

Issue: #

Associated Developers

Dev: @

Checklist for Reviewers

Reviewers should use this link to get to the
Review Checklist before they begin their review.

- Add helper functions - Add tests for read_depleted_materials() and helper functions - add `saltproc_runtime_ref` directory in `openmc_data` - remove unneeded import statements - make importable time variables in app.py - add assertion for `depcode_input[ouput_path]` in test_app.py

- add relevant tests - move SerpentDepcode.read_plaintext_file to Depcode.read_plaintext_file - fix call to `_check_for_material_names` in OpenMCDepcode - fix `write_runtime_input` in OpenMCDepcode - simplify `file_interface_openmc` integration tests - fix `test_check_for_material_names` unit test - add helper function to app.py

OpenMCDepcode - add helper funtions to calculate neutronics parameters - ...

- add machinery in abc.py to preserve all depcode output - add machinery in app.py that allows fission_q to be set correctly - add machinery in openmc_msbr_model.py to make stochastic volume calculation optional - Move duplicated code in the `run_depletion_step` methods to Depcode.run_depletion_step - Make `convert_nuclide_code_to_name` a required function for all Depcode child classes - Streamline SaltProc-generated tallies for OpenMC coupled simulatiosn - add helper functions to ensure data consistency when storing results in the SaltProc results database

move serpent integration tests

Co-authored-by: Sam Dotson <[email protected]>

…enmc-file-io

yardasol · 2023-02-23T21:05:30Z

@samgdotson please allow other people some time to get their reviews in before merging.

abachma2

Hi @yardasol! thanks for getting this in. Overall it looks pretty good to me, I just have a couple of questions.

abachma2 · 2023-02-23T21:09:37Z

.github/workflows/test-saltproc.yml

          path: |
            /usr/share/miniconda3/envs/saltproc-env
            ~/openmc_src
+            ~/mcpl_src


What is mcpl? I'm not familiar with that, assuming that the directory name is for a code source.

I'm not entirely sure, but it appears to be some sort of file format. I added MCPL to SaltProc's CI in #182, but I forgot to add full caching support, which I am doing here.

Ok I just hadn't heard of it before and I was curious.

abachma2 · 2023-02-23T21:10:54Z

.github/workflows/test-saltproc.yml

+
+      - name: Test SaltProc
+        run: |
+          pytest --ignore tests/integration_tests/run_no_reprocessing_serpent --ignore tests/integration_tests/run_no_reprocessing_openmc --ignore tests/integration_tests/run_constant_reprocessing_serpent --ignore tests/integration_tests/run_constant_reprocessing_openmc tests/


Why are you ignoring all of these integration tests?

The serpent ones require export controlled software, which I don't even know how I would get onto a GH actions runner, and the OpenMC ones take a long time to run, too long for CI in my opinion.

Ok. I get not including the serpent tests. Is there a way to speed up the OpenMC tests? Also, what is even being tested in the CI if you include tests/ in the list of directories to ignore?

I second Amanda's question.

Is there a way to speed up the OpenMC tests?

I haven't been able to find a way to do this.

Also, what is even being tested in the CI if you include tests/ in the list of directories to ignore?

tests/ is not being ignored, only the directories that have the --ignore flag are being ignored. You can look at the Checks tab to verify this.

abachma2 · 2023-02-23T21:16:16Z

examples/msbr/openmc_msbr_model.py

    -------
    deplete : bool
        Flag indicated whether or not to run a depletion simulation.
+    volume : bool


Suggested change

volume : bool

volume_calculation: bool

This just feels like a more descriptive variable name

Seconded, though I also agree with later comments that volume_calculation is very close to a function name, and something like stoch_volume or stoch_vol might work.

abachma2 · 2023-02-23T21:16:55Z

examples/msbr/openmc_msbr_model.py

                        type=bool,
                        default=False,
                        help='flag for running depletion')
+    parser.add_argument('--volume',


Suggested change

parser.add_argument('--volume',

parser.add_argument('--volume_calculation',

If you change it earlier the code

abachma2 · 2023-02-23T21:17:10Z

examples/msbr/openmc_msbr_model.py


    args = parser.parse_args()
-    return bool(args.deplete)
+    return bool(args.deplete), bool(args.volume)


abachma2 · 2023-02-23T21:17:27Z

examples/msbr/openmc_msbr_model.py



-deplete = parse_arguments()
+deplete, volume = parse_arguments()


abachma2 · 2023-02-23T21:17:37Z

examples/msbr/openmc_msbr_model.py

-#msbr_volume_calc.set_trigger(1e-03, 'rel_err')
-settings.volume_calculations = [msbr_volume_calc]
-settings.run_mode = 'volume'
+if volume:


abachma2 · 2023-02-23T21:18:23Z

examples/msbr/openmc_msbr_model.py

+if volume:
+    msbr_volume_calc = openmc.VolumeCalculation([fuel, moder], int(1e10), ll, ur)
+    #msbr_volume_calc.set_trigger(1e-03, 'rel_err')
+    settings.volume_calculations = [msbr_volume_calc]


Ok, maybe my suggestion wasn't that great because it's pretty similar to this one.

Maybe something like stoch_volume?

Or stoch_vol, if you're worried about length?

abachma2 · 2023-02-23T21:41:05Z

saltproc/openmc_depcode.py

+        tallies = openmc.Tallies()
+        tallies += list(delayed_lambda.tallies.values())
+        tallies += list(beta.tallies.values())


What is the structure of this object? Is is a giant list of the tallies? Are there labels for the tallies anywhere?

A Tallies object is a list of tallies with some additional class functions. Yes the tallies have their own IDs and names.

And you're summing various tallies into the same list? Why?

Because that's how OpenMC works.

abachma2 · 2023-02-23T21:58:20Z

tests/integration_tests/run_no_reprocessing_openmc/test_openmc.py

@@ -0,0 +1,95 @@
+"""Run SaltProc without reprocessing"""


Maybe this is later in the PR, but do you have a test file running SaltProc with reprocessing?

LukeSeifert

I have a few things I was curious about and some minor corrections, but overall looks good.

.github/workflows/test-saltproc.yml

examples/msbr/msbr_endfb71.serpent

LukeSeifert · 2023-02-24T20:00:40Z

examples/msbr/openmc_msbr_model.py

+if volume:
+    msbr_volume_calc = openmc.VolumeCalculation([fuel, moder], int(1e10), ll, ur)
+    #msbr_volume_calc.set_trigger(1e-03, 'rel_err')
+    settings.volume_calculations = [msbr_volume_calc]


Maybe something like stoch_volume?

LukeSeifert · 2023-02-24T20:10:17Z

saltproc/abc.py

+    def convert_nuclide_code_to_name(self, nuc_code):
+        """Converts depcode nuclide code to symbolic nuclide name.
+
+        Parameters
+        ----------
+        nuc_code : str
+            Nuclide code
+
+        Returns
+        -------
+        nuc_name : str
+            Symbolic nuclide name (`Am242m1`).
+
+        """
+


I'm not very familiar with abc, what does this do?

saltproc/app.py

saltproc/openmc_depcode.py

saltproc/serpent_depcode.py

tests/integration_tests/file_interface_openmc/test.py

LukeSeifert · 2023-02-24T20:51:44Z

tests/integration_tests/file_interface_openmc/test.py

+    ref_mats = openmc_depcode.read_depleted_materials(True)
+    openmc_depcode.output_path = old_output_path
+
+    openmc_depcode.update_depletable_materials(ref_mats, 12.0)


What is the 12.0 here?

Co-authored-by: LukeSeifert <[email protected]>

LukeSeifert · 2023-02-27T22:46:32Z

examples/msbr/openmc_msbr_model.py

+if entropy:
+    entropy_mesh = openmc.RegularMesh()
+    entropy_mesh.lower_left = ll
+    entropy_mesh.upper_right = ur
+    entropy_mesh.dimension = (20, 20, 20)
+    settings.entropy_mesh = entropy_mesh


Nice addition!

LukeSeifert · 2023-02-27T22:47:00Z

saltproc/openmc_depcode.py

            A material
        percent_type : str
-            Percent type of material
+            Percent type of material,


Accidental comma?

LukeSeifert · 2023-02-27T22:50:02Z

saltproc/openmc_depcode.py

+        diluted_model = openmc.Model(materials=materials, geometry=geometry, settings=settings)
+        reactions, diluted_materials = MicroXS._add_dilute_nuclides(self.chain_file_path,
+                                                                   diluted_model,
+                                                                   1e3)


I think it might be helpful to include a variable inital_atom_dens = 1e3 or something similar, but if that won't need to be changed later then feel free to leave it as is.

LukeSeifert

Good changes, looks good to me. I'll wait for other people to get their reviews in before merging.

yardasol · 2023-02-28T20:50:58Z

I'm fixing one of the local integration tests, so hold off on merging

ZoeRichter

Generally I think it looks good. My biggest comment is that there are some functions without any docstrings. Even if it's just a quick one line docstring explaining the purpose of the function/test, I think it would be helpful. I left a comment marking files where I noticed functions with missing docstrings.

ZoeRichter · 2023-03-01T20:15:48Z

.github/workflows/test-saltproc.yml

+
+      - name: Test SaltProc
+        run: |
+          pytest --ignore tests/integration_tests/run_no_reprocessing_serpent --ignore tests/integration_tests/run_no_reprocessing_openmc --ignore tests/integration_tests/run_constant_reprocessing_serpent --ignore tests/integration_tests/run_constant_reprocessing_openmc tests/


I second Amanda's question.

ZoeRichter · 2023-03-01T20:37:06Z

examples/msbr/msbr_openmc.json

+       "sim_name": "msbr_openmc_test"
+   },
+   "reactor": {
+       "volume": 1.0,


Is the volume 1.0 here because the example is only examining a small portion of the salt?

ZoeRichter · 2023-03-01T20:40:04Z

examples/msbr/openmc_msbr_model.py

    -------
    deplete : bool
        Flag indicated whether or not to run a depletion simulation.
+    volume : bool


Seconded, though I also agree with later comments that volume_calculation is very close to a function name, and something like stoch_volume or stoch_vol might work.

ZoeRichter · 2023-03-01T20:44:26Z

examples/msbr/openmc_msbr_model.py


    args = parser.parse_args()
-    return bool(args.deplete)
+    return bool(args.deplete), bool(args.volume), bool(args.entropy)


Another one, if you change variable name

ZoeRichter · 2023-03-01T20:44:51Z

examples/msbr/openmc_msbr_model.py



-deplete = parse_arguments()
+deplete, volume, entropy = parse_arguments()


if you change volume to something else

ZoeRichter · 2023-03-01T23:26:01Z

tests/msbr_processes.json

+        "efficiency": {
+          "Kr": 0,
+          "Xe": 0,
+          "H": 0


The sparger has an efficiency of zero? Does that mean it's not removing anything?

I believe so.

ZoeRichter · 2023-03-01T23:28:07Z

tests/msbr_processes.json

+        "efficiency": {
+          "Kr": 0.9115218695901359,
+          "Xe": 0.9148573428740562
+        },


Or is the gas separator where you account for removal, not the sparger? Does this separator not also remove the helium gas from the liquid salt before it hits a liquid pump?

As you can see, the entrainment separator does not remove helium. I think the way the reprocessing system files are set up is the least understood/documented aspect of SaltProc right now. I'm going to be adding some guides on this.

ZoeRichter · 2023-03-01T23:36:19Z

tests/unit_tests/test_app.py

+        str((input_path / 'chain_endfb71_pwr.xml').resolve())

    modified_operator_kwargs = expected_depletion_settings['operator_kwargs'].copy()
    if filename == 'constant_fission_yield':


ZoeRichter · 2023-03-01T23:36:48Z

tests/unit_tests/test_openmc_depcode.py

+    assert openmc_depcode._convert_nucname_to_pyne('U238') == 'U238'
+    assert openmc_depcode._convert_nucname_to_pyne('Ag110_m1') == 'Ag110M'
+    assert openmc_depcode._convert_nucname_to_pyne('Am242') == 'Am242'
+    assert openmc_depcode._convert_nucname_to_pyne('Am242_m1') == 'Am242M'


ZoeRichter · 2023-03-01T23:37:57Z

saltproc/openmc_depcode.py

+        self._fissile_nucs = set([nuc for nuc in _FISSILE_NUCLIDES if nuc in burnable_nucs])
+        self._fertile_nucs = set([nuc for nuc in _FERTILE_NUCLIDES if nuc in burnable_nucs])

+        return list(self._fissile_nucs.union(self._fertile_nucs))


I personally think the function name is descriptive enough. It gets the fissile and fertile nuclides.

yardasol · 2023-03-06T17:25:56Z

@samgdotson @abachma2 @ZoeRichter @LukeSeifert the integration tests have been fixed.

samgdotson

Good job @yardasol. I can merge if others are happy.

yardasol force-pushed the openmc-file-io branch from d017f56 to 0c73c42 Compare February 1, 2023 20:01

yardasol force-pushed the openmc-file-io branch from ff248ff to 6d62da9 Compare February 2, 2023 02:46

implement read_neutronics_parameters and read_step_metadata in

d9b8501

OpenMCDepcode - add helper funtions to calculate neutronics parameters - ...

yardasol mentioned this pull request Feb 10, 2023

Feature: Consolidate nuclide name and code processing functions #191

Closed

yardasol force-pushed the openmc-file-io branch from 356005c to c483259 Compare February 10, 2023 17:48

yardasol added 2 commits February 13, 2023 09:35

runtime bugfixes

d2c6957

move serpent integration tests

yardasol force-pushed the openmc-file-io branch from c483259 to d2c6957 Compare February 13, 2023 15:36

yardasol added 2 commits February 14, 2023 17:33

fix test suite

b52b618

update release notes

7b5920b

yardasol force-pushed the openmc-file-io branch 6 times, most recently from 6c591b6 to 786d14e Compare February 15, 2023 23:13

yardasol added Status:4-In Progress Priority:1-Critical Difficulty:2-Challenging labels Feb 15, 2023

yardasol force-pushed the openmc-file-io branch 2 times, most recently from 1d0cc21 to 72d02bc Compare February 16, 2023 19:41

yardasol marked this pull request as ready for review February 16, 2023 19:41

yardasol requested review from LukeSeifert, ZoeRichter, abachma2 and samgdotson February 16, 2023 19:42

yardasol force-pushed the openmc-file-io branch 2 times, most recently from f818230 to d0f2241 Compare February 17, 2023 01:09

yardasol and others added 3 commits February 23, 2023 15:04

Apply suggestions from code review

3abb8d4

Co-authored-by: Sam Dotson <[email protected]>

fix docstring in Depcode.run_depletion_step()

34f7d75

Merge branch 'openmc-file-io' of github.com:yardasol/saltproc into op…

1765a7c

…enmc-file-io

yardasol added 2 commits February 23, 2023 15:12

fix duplicate id warnings

164ed20

typo fix

2c16079

abachma2 reviewed Feb 23, 2023

View reviewed changes

LukeSeifert reviewed Feb 24, 2023

View reviewed changes

yardasol and others added 3 commits February 27, 2023 11:00

@LukeSefiert suggestions for openmc_depcode.py

bf8a44b

update batch and particle settings for convergence

d5b7491

Apply suggestions from code review

12cfdb9

Co-authored-by: LukeSeifert <[email protected]>

yardasol requested review from LukeSeifert and abachma2 February 27, 2023 20:11

LukeSeifert reviewed Feb 27, 2023

View reviewed changes

LukeSeifert approved these changes Feb 27, 2023

View reviewed changes

yardasol requested a review from samgdotson February 28, 2023 20:51

typo fixes; fix tests; update msbr example

3749522

yardasol force-pushed the openmc-file-io branch from 03529c2 to 3749522 Compare February 28, 2023 20:52

fix run_depletion_step

b4ce1f3

ZoeRichter reviewed Mar 1, 2023

View reviewed changes

docstring fixes

652a34e

yardasol requested a review from ZoeRichter March 2, 2023 19:30

fix constant reprocessing integration test

31cbef5

samgdotson approved these changes Mar 6, 2023

View reviewed changes

samgdotson merged commit b20311a into arfc:master Mar 7, 2023

yardasol mentioned this pull request Mar 7, 2023

Feature: DepcodeOpenMC methods for starting simulation and writing/reading input files #133

Closed

yardasol removed the Status:4-In Progress label Mar 7, 2023

yardasol mentioned this pull request Mar 8, 2023

Feature: OpenMC Support #112

Closed

13 tasks

	parser.add_argument('--volume',
	parser.add_argument('--volume_calculation',



		deplete = parse_arguments()
		deplete, volume = parse_arguments()



		deplete = parse_arguments()
		deplete, volume, entropy = parse_arguments()

Uh oh!

Implement file and database IO functions in OpenMCDepcode #189

Implement file and database IO functions in OpenMCDepcode #189

Uh oh!

Conversation

yardasol commented Feb 1, 2023 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary of changes

Types of changes

Required for Merging

Associated Issues and PRs

Associated Developers

Checklist for Reviewers

Uh oh!

yardasol commented Feb 23, 2023

Uh oh!

abachma2 left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

LukeSeifert left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Implement file and database IO functions in `OpenMCDepcode` #189

Implement file and database IO functions in `OpenMCDepcode` #189

yardasol commented Feb 1, 2023 •

edited

Loading