[WIP] Val 2i - Deuterium retention in neutron-irradiated single-crystal tungsten by cticenhour · Pull Request #325 · idaholab/tmap8

cticenhour · 2025-11-07T21:09:42Z

Masa's Paper: https://www.sciencedirect.com/science/article/pii/S0920379618303880

Closes #393

simopier

Thank you @cticenhour! I'm excited for this case to be added!

I left comments and suggestions even though this is still tagged as [WIP] because I was too curious to ignore it. It looks like the main thing missing is the documentation, correct?

simopier · 2025-12-19T05:08:27Z

+# tmap_flux_for_rmspe = numerical_solution_on_experiment_input(experiment_time, simulation_time_TMAP7, simulation_flux_left_TMAP7/2 + flux_environment)
+# RMSE = np.sqrt(np.mean((tmap_flux_for_rmspe-experiment_flux)**2) )
+# RMSPE = RMSE*100/np.mean(experiment_flux)
+# ax.text(6000.0,0.85e18, 'RMSPE = %.2f '%RMSPE+'%',fontweight='bold')


Why is this commented out? Maybe it's just because this is still in WIP stage, but calculating the RMSPE between TMAP8's predictions and the experimental data is a good thing to add.

This was commented out because it added complexity while I was working through adjusting the rest of the plot. It is planned to be re-added before this is merged.

RMSPE values (for TMAP4 and TMAP8 simulation results against experimental data.....thought it would be an interesting point of comparison) were added in 8506bf3. Though, there are some quirks that needs to be ironed out. TMAP4 RMSPE looks reasonable, but TMAP8 RMSPE is off the charts (+15000%). Looks like there is an error or the data needs to be "cleaned" - like, we need to compensate for the environment or something. Need to explore further.

Looks like the full data set has quite a bit of noise that was truncated out in the data plot:

Do you have a good way off the top of your head to remove this "environment flux"? I see that an environment flux was calculated and used in the val-2d comparison script, but I am not sure if we can do the same here. Probably need to review the context in the documentation, first.

Are you talking about the values at 10^5? this sounds like basically 0, and it probably corresponds to the minimum detection limit. I don't think it's meaningful in any way. Masa might be able to confirm or deny that.

simopier · 2025-12-19T05:15:31Z

+[]
+
+[Postprocessors]
+  [flux_surface_left]


Specify units as a comment.

Addressed in b2d204c.

cticenhour · 2026-03-17T03:08:33Z

Rebased on devel to fix conflicts with the current val-2g case. This case is now val-2i. Note that "outdated" or resolved discussion from former val-2g files in this PR are still labeled as val-2g. Currently unresolved comments and discussion auto-magically were updated by GitHub to reference the renamed files.

Confirmed that temperature and diffusivity profiles over time look reasonable, and there is a reasonably high trapped population, but zero mobile concentration. Convergence fails over only a couple of timesteps.

…timestepper, and lower trap_per_free parameter

…ture ploting script

…t with paper

…concentration, and other parameter adjustments for clarity

…ts with the source publication

…settings in input file, that did not impact results.

Use random sampling to reduce the density of the scatter plot of the experimental data - similar to what Masa's doing in the paper. Cannot tell what approach he is using the reduce the plot density. As of this change, 10% of the data will be sampled and shown on the plot using the random sampler.

…) parameters

Co-authored-by: Pierre-Clement Simon <pierreclement.simon@gmail.com>

…ectly from the simulation. Add temperature function to postprocessor output for use in scripts. Start output of simulation data from timestep 0 to get the full temperature series.

Refs idaholab#393

- Fix typo in temperature history script. Now always uses gold files from TMAP8 simulation. - Add units comments to postprocessor outputs

…al data

moosebuild · 2026-03-18T18:00:33Z

Job Precheck, step Python: black format on 73ff72b wanted to post the following:

Python black formatting

Your code requires style changes.

A patch was generated and copied here.

You can directly apply the patch by running the following at the top level of your repository:

curl -s https://mooseframework.inl.gov/tmap8/docs/PRs/325/black/black.patch | git apply -v

Alternatively, you can run the following at the top level of your repository:

black --config pyproject.toml --workers 1 .

moosebuild · 2026-03-18T19:49:02Z

Job Build test summary, step Build test summary on b41cfdf wanted to post the following:

Test summary

Compared against c837769 in job civet.inl.gov/job/3665797.

Added tests

Test	Time (s)	Memory (MB)
`test:val-2i.val-2i_673K_light`	4.68	204.66
`test:val-2i.val-2i_comparison`	2.43	149.25
`test:val-2i.val-2i_673K`	SKIP	0.00
`test:val-2i.val-2i_873K`	SKIP	0.00
`test:val-2i.val-2i_973K`	SKIP	0.00

Run time changes

Test	Base (s)	Head (s)	+/-	Base (MB)	Head (MB)
`test:ver-1fb.thermal_transient_physics`	4.58	7.14	+55.91%	53.46	59.70
`test:ver-1fb.ver-1fb`	2.44	3.69	+51.01%	43.90	44.89

moosebuild · 2026-03-18T19:55:52Z

Job Documentation, step Sync to remote on 8a04362 wanted to post the following:

View the site here

This comment will be updated on new commits.

Some of this was needed during debug, but is no longer. This change also makes it easier to turn off Exodus output for testing.

The MUMPS change will hopefully help with robustness in parallel / distributed mesh

simopier

I know you are still working on this, but here is a review with comments for you to consider.
Thank you for adding this case!

The light test currently fails:

DIFF test:val-2i.val-2i_673K_light FAILED (CSVDIFF)

simopier · 2026-03-23T02:53:34Z

+trapping_site_fraction_1 = 0.002 # (-)
+trapping_rate_prefactor = '${fparse ${diffusivity_coefficient} / ${W_lattice_constant}^2}' # (1/s)
+release_rate_prefactor = '${units 1e13 1/s}'
+trap_per_free_1 = 1e6 # (-)


The trap per free coefficient is not in the equations in the documentation. You should add it and provide its value.

simopier · 2026-03-23T02:54:26Z

+[]
+
+[Variables]
+  [concentration]


I suggest changing this variable name to mobile since concentration could apply to the trapping too.

simopier · 2026-03-23T02:54:49Z

+  [trapped_1]
+    order = FIRST
+    family = LAGRANGE
+    outputs = none


why not? I suggest you output it.

Suggested change

outputs = none

simopier · 2026-08-01T17:07:33Z

@cticenhour, do you need any help with this?

edbworth · 2026-08-17T17:29:06Z

@cticenhour, do you need any help with this?

This is quite close to my current draft PR. If you would like me to help with this, I can after I finish up that work.

cticenhour · 2026-08-17T17:49:44Z

I plan on making good progress on this over this week, so I think I'll be good. Thanks!

Co-authored-by: Pierre-Clement Simon <pierreclement.simon@gmail.com>

…consistently shifted Add a note about why we do this to the comparison script

- Document use of the `trap_per_free` scaling factor in val-2i.md and report its value used in val-2i.i. - Also in val-2i.md, replace TODO citation placeholders with the actual references and consolidate listing in input files section to being paragraph-based. - Simplify variable/parameter naming in val-2i.i since only one trapping site is modeled. That is, drop the "_1" suffixes (trapped_1 -> trapped, trap_per_free_1 -> trap_per_free, trapping_site_fraction_1 -> trapping_site_fraction, etc.). Also, update the matching `cli_args` in the tests file for the renamed trapping_site_fraction parameter. - Make the mobile species variable naming more clear (concentration -> mobile) and enable output of the trapped variable. - Improve comparison_val-2i.py plot readability: outline the TMAP8 and TMAP4 curves with a black stroke path effect, and color-match each RMSPE annotation to its corresponding curve.

cticenhour changed the title ~~Deuterium retention in neutron-irradiated single-crystal tungsten (Val-2g)~~ [WIP] Deuterium retention in neutron-irradiated single-crystal tungsten (Val-2g) Nov 7, 2025

cticenhour marked this pull request as draft November 7, 2025 21:09

cticenhour force-pushed the isfnt-val-2g branch from 4b39d0e to c937c4f Compare November 10, 2025 19:21

cticenhour changed the title ~~[WIP] Deuterium retention in neutron-irradiated single-crystal tungsten (Val-2g)~~ [WIP] Deuterium retention in neutron-irradiated single-crystal tungsten Nov 11, 2025

cticenhour force-pushed the isfnt-val-2g branch from 135be67 to 4c88923 Compare November 24, 2025 20:28

simopier requested changes Dec 19, 2025

View reviewed changes

simopier added the V&V Relevant to V&V label Dec 19, 2025

cticenhour force-pushed the isfnt-val-2g branch 2 times, most recently from ffd05b1 to a9d8db5 Compare January 23, 2026 10:02

cticenhour force-pushed the isfnt-val-2g branch 2 times, most recently from 64e353e to 71ff949 Compare March 17, 2026 03:05

cticenhour and others added 18 commits March 16, 2026 21:24

Add temperature series plotting script for ISFNT case

bbd4562

Add initial val-2i input file

aa7be7d

Confirmed that temperature and diffusivity profiles over time look reasonable, and there is a reasonably high trapped population, but zero mobile concentration. Convergence fails over only a couple of timesteps.

Refine simple mesh, turn on timestep limiting guidelines in adaptive …

9d89907

…timestepper, and lower trap_per_free parameter

Add source aux for troubleshooting

e107165

Allow easier adjustment of font sizes for labels and axes for tempera…

b9aebbd

…ture ploting script

Add lower bounds for concentration and trapped_1 variables

16342f7

Add reflection coefficient impact to source term to be more consisten…

245890c

…t with paper

Parameter adjustments to be more consistent with paper usage of trap …

a0f80af

…concentration, and other parameter adjustments for clarity

Switch to CartesianMeshGenerator to refine in source region a bit better

28ebe60

Adjust trap_per_free and refine in implantation region to get to 8148 s

65bcff6

Adjust tolerances to get case running and add initial plotting script

7f15e92

Make adjustments to better match simulation and presentation of resul…

fd73aa0

…ts with the source publication

Adjustments to clean up commented out items or otherwise unnecessary …

dea88e4

…settings in input file, that did not impact results.

Add TMAP4 run data to plot comparison script

e53fd58

Improve readability of plot by adjusting grid and alpha (transparency…

89a3c5c

…) parameters

Apply suggestions from code review

6a83aed

Co-authored-by: Pierre-Clement Simon <pierreclement.simon@gmail.com>

Add units to experimental and TMAP4 data for val-2i

c9f11bc

Adjust temperature plotting script to include temperature profile dir…

669ae7d

…ectly from the simulation. Add temperature function to postprocessor output for use in scripts. Start output of simulation data from timestep 0 to get the full temperature series.

cticenhour force-pushed the isfnt-val-2g branch from 71ff949 to 669ae7d Compare March 17, 2026 03:35

cticenhour added 2 commits March 16, 2026 22:36

Add initial val-2i tests

46bbca2

Refs idaholab#393

Add light test for val-2i and mark full tests as heavy

5fc10fa

cticenhour changed the title ~~[WIP] Deuterium retention in neutron-irradiated single-crystal tungsten~~ [WIP] Val 2i - Deuterium retention in neutron-irradiated single-crystal tungsten Mar 17, 2026

cticenhour added 7 commits March 18, 2026 10:38

Add val-2i documentation

d26c108

Fixup typo in comparison script when used with documentation

54cda56

Change name of temperature history generation file and output file

2a836dc

Address review comments

b2d204c

- Fix typo in temperature history script. Now always uses gold files from TMAP8 simulation. - Add units comments to postprocessor outputs

Remove data masking and use alphas instead

d071487

Add RMSPE for both TMAP4 results and TMAP8 results against experiment…

8506bf3

…al data

Add to caption about RMSPE values

73ff72b

cticenhour changed the title ~~[WIP] Val 2i - Deuterium retention in neutron-irradiated single-crystal tungsten~~ Val 2i - Deuterium retention in neutron-irradiated single-crystal tungsten Mar 18, 2026

Perform black formatting on val-2i python files

b41cfdf

cticenhour added 3 commits March 19, 2026 09:35

Simplify outputs setup

e1a527a

Some of this was needed during debug, but is no longer. This change also makes it easier to turn off Exodus output for testing.

Switch to MUMPS and reduce bounds min to zero

8ebe7bf

The MUMPS change will hopefully help with robustness in parallel / distributed mesh

Differentiate requirements in val-2i tests

8a04362

cticenhour mentioned this pull request Mar 19, 2026

Fix the light tests failure on MacOS #382

Merged

simopier requested changes Mar 23, 2026

View reviewed changes

simopier mentioned this pull request Apr 9, 2026

Add validation case for deuterium transport in neutron-irradiated tungsten #405

Open

cticenhour and others added 2 commits August 17, 2026 12:00

Apply suggestions from code review

a3122ee

Co-authored-by: Pierre-Clement Simon <pierreclement.simon@gmail.com>

Fixup bug in RMSPE calculation; TMAP8 time series data was not being …

d845d2e

…consistently shifted Add a note about why we do this to the comparison script

cticenhour changed the title ~~Val 2i - Deuterium retention in neutron-irradiated single-crystal tungsten~~ [WIP] Val 2i - Deuterium retention in neutron-irradiated single-crystal tungsten Aug 18, 2026

Uh oh!

Conversation

cticenhour commented Nov 7, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

simopier left a comment

Choose a reason for hiding this comment

Uh oh!

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!

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!

cticenhour commented Mar 17, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

moosebuild commented Mar 18, 2026

Python black formatting

Uh oh!

moosebuild commented Mar 18, 2026

Test summary

Added tests

Run time changes

Uh oh!

moosebuild commented Mar 18, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

simopier left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

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!

simopier commented Aug 1, 2026

Uh oh!

edbworth commented Aug 17, 2026

Uh oh!

cticenhour commented Aug 17, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

cticenhour commented Nov 7, 2025 •

edited

Loading

cticenhour commented Mar 17, 2026 •

edited

Loading

moosebuild commented Mar 18, 2026 •

edited

Loading