Add Physics base class and a heat conduction example

[GiudGiud]

Open for comments.
Once the dependent PRs are in, this could actually be merged
refs #25642
see original PR #25704

I dont think the syntax has been decided yet, so I m open to a CCB meeting on this

Currently the Physics base class only implements:

• some basic parameter checking
• an insertion point for relationship managers at the equation level (could be removed for now, will come in for NS)
• an insertion point for restarting from Exodus (could be removed for now, will come in for NS)

[moosebuild]

Job Documentation on bd7ff03 wanted to post the following:

View the site here

This comment will be updated on new commits.

[moosebuild]

Job Coverage on bd7ff03 wanted to post the following:

Framework coverage

	796cb8	#25977 bd7ff0
	Total	Total	+/-	New
Rate	85.24%	85.20%	-0.04%	74.95%
Hits	99138	99509	+371	377
Misses	17166	17283	+117	126

Diff coverage report

Full coverage report

Modules coverage

Heat transfer

	796cb8	#25977 bd7ff0
	Total	Total	+/-	New
Rate	88.28%	88.16%	-0.11%	83.49%
Hits	3938	4029	+91	91
Misses	523	541	+18	18

Diff coverage report

Full coverage report

Full coverage reports

Reports

• framework
• chemical_reactions
• combined
• contact
• electromagnetics
• external_petsc_solver
• fluid_properties
• fsi
• functional_expansion_tools
• geochemistry
• heat_transfer
• level_set
• misc
• navier_stokes
• optimization
• peridynamics
• phase_field
• porous_flow
• ray_tracing
• rdg
• reactor
• richards
• scalar_transport
• solid_properties
• stochastic_tools
• tensor_mechanics
• thermal_hydraulics
• xfem

Warnings

• framework new line coverage rate 74.95% is less than the suggested 90.0%
• heat_transfer new line coverage rate 83.49% is less than the suggested 90.0%

This comment will be updated on new commits.

[GiudGiud]

Should probably move this forward. It doesnt have to be merged right right now, we could also wait until I dig deep into Component as Actions and assess if this is going to change the deal for Physics. I dont expect it to. But work on syntax for Components as Actions could enable interesting syntax for Physics.

[lindsayad]

want to add a framework test?

[GiudGiud]

need a framework class then? Thoughts on which physics? A test object?

[lindsayad]

The thing that came to my mind is DiffusionPhysics or PoissonPhysics. @nmnobre just recently created a coupled_elestrostatics.i framework level test in his Raviart-Thomas PR.... It could be a good demonstration of Physics to make both CFEM and Raviart-Thomas implementations of the Poisson equation.... Although I can't remember if you wanted Physics to be useful for the purpose of allowing different algebraic discretizations of the same physics. Also you probably don't want to balloon this PR too much, so maybe just a CFEM implementation of Poisson would be good to start, and then we could add RT later.

I thought we had an issue where we had a pretty large community (I thought it involved @dschwen at least) discussion about the Physics class but maybe i've lost my marbles. #25642 doesn't seem like it.

[GiudGiud]

I'm happy to implement that.

[GiudGiud]

A lot of the discussion is on the original Physics PR #25704

wanted Physics to be useful for the purpose of allowing different algebraic discretizations of the same physics

I do want that. I dont think it s feasible to do a [Discretization] block with our current ways of parsing the input file without being very convoluted. So I think when we instantiate a Physics object in the input file, it should have the discretization as well

[lindsayad]

A lot of the discussion is on the original Physics PR #25704

Ah there it is, thanks

[GiudGiud]

Let me know when this is ready for a thorough review. I'll have a lot of requests for documentation, but I'd probably just have us fix anything wrong and then I can do a follow-up PR to improve the documentation to my satisfaction, and you can review that.

Alex has been thoroughly reviewing this. I dont think there is anything left wrong! He did catch a lot of wrong things

[joshuahansel]

I dont think there is anything left wrong! He did catch a lot of wrong things

Oh don't worry, there's wrong stuff still. But just doc/typos so far 😄

[joshuahansel]

You should probably make a newsletter entry

[GiudGiud]

I ll make one for all of my January work soon

[GiudGiud]

noice thanks a lot for the reviews. Merging so it makes the newsletters

now for the tough part. NS, TM and THM