Implement a new MeshMaterialFilter

[paulromano]

Description

This PR implements a new MeshMaterialFilter that allows tallying specific combinations of mesh elements and materials. In principle, one can already do this by combining a MeshFilter and a MaterialFilter. However, in many cases a single mesh element only contains a few materials, which means tallying every single material for each mesh element would be very wasteful. This new filter works in conjunction with the Mesh.material_volumes() method, which returns a MeshMaterialVolumes object. This filter has a from_volumes() classmethod that will create a filter with bins for all mesh element-material combinations that were found from the call to material_volumes(). So, the workflow is basically:

# Build your model
model = openmc.Model()
...

# Set up a mesh
mesh = openmc.RegularMesh()
...


# Determine materials present in each mesh element
mat_vols = mesh.material_volumes(model, ...)

# Create corresponding filter
mm_filter = openmc.MeshMaterialFilter.from_volumes(mat_vols)

This new filter is one of the last infrastructural pieces needed to carry out mesh-based subvoxel R2S calculations.

Checklist

• I have performed a self-review of my own code
• I have run clang-format (version 15) on any C++ source files (if applicable)
• I have followed the style guidelines for Python source files (if applicable)
• I have made corresponding changes to the documentation (if applicable)
• I have added tests that prove my fix is effective or that my feature works (if applicable)

[pshriwise]

Nice @paulromano! I'm looking forward to trying this out. Some potential small refactors suggested here.

[shimwell]

Awesome, thanks Patrick, looks like lots of good changes.

I've been putting this branch to use to see how the workflow goes.

I can probably improve this draft notebook. It does appear possible to do and R2S workflow with this new MeshMaterialFilter. I can't say it was easy and I'm not 100% sure I've done all the processing correctly but I've got something here.

https://github.com/fusion-energy/neutronics-workshop/blob/testing-MeshMaterialFilter/tasks/task_11_CSG_shut_down_dose_tallies/3_R2S_regularmesh_shutdown_dose_rate.ipynb

I think we could consider adding some utility functions to make some of the stages more compact.

[paulromano]

@shimwell @pshriwise Thanks for the reviews on this PR and glad to hear that you are already incorporating it into your workflows Jon. There are a few more pieces that are needed IMO to make the mesh-based R2S workflow correct/seamless including, as you suggested, some utility functions. I have another branch or two I'm working on with exactly that.

In any case, all the comments have been addressed other than a few which I've responded directly to about why I haven't incorporated them. Let me know if you have any further requests for changes.

[shimwell]

I don't have any further changes. Looks good from my point of view

I clicked resolve on my only remaining comment as you addressed it with the /n

I would be keen to evolve the R2S example I made to make use of your utility functions as they arrive. It will be really satisfying to shrink that example down to a more manageable script.

[pshriwise]

Looks good @paulromano! I'll give this a short cooldown and merge tomorrow.