Refactor two fluid model

[szekedidavid]

No description provided.

[spossann]

This is just the first part of my review.

[spossann]

We should make clear in which units the viscosity is passed here. For this, we can add a new attribute in struphy.physics.physics.Units, like

self._nu = A_bulk * con.mH * self.n * self.x * self.v

See formula (2.17) of the master thesis. In this way the units of the viscosity will be auto-computed and can be queried by the user.

[szekedidavid]

Added to units. Do I access it via passing sim.units through Options?

[spossann]

@szekedidavid Can you put the construction of the V0-Derham complex for lifting in the Derham class already? In that way I could start from that, which would be nice.

[szekedidavid]

@szekedidavid Can you put the construction of the V0-Derham complex for lifting in the Derham class already? In that way I could start from that, which would be nice.

Done! I also removed the boundary condition flags in the propagator.
Let me know when you finished the notebook.

[spossann]

@szekedidavid Can you put the construction of the V0-Derham complex for lifting in the Derham class already? In that way I could start from that, which would be nice.

Done! I also removed the boundary condition flags in the propagator. Let me know when you finished the notebook.

Thanks! Will let you know asap.

[spossann]

Hi @szekedidavid - a new notebook is available, see this comment: #212 (comment)

[szekedidavid]

Hi @spossann , I managed to make lifting work in 1D using the FEECVariable implementation. Also expanded it to 2D and added the full solution (propagator is much, much cleaner now). The 2D lifting test case is not yet working, but it could entirely be a stabilization parameter or time stepping issue (unconstrained tangential velocity DoFs blow up in Hdiv).

[spossann]

Hi @szekedidavid - how are you? I managed to merge the branch with the current devel, which featured some re-factoring of how the propagators are stored (now in separate files). I hope that all your major changes are still there, please have a look.

Important (!!): I could not get the changes you made in struphy-parameter-files, you might have them only locally. you could push them to the repo struphy-parameter-files or just copy them to a safe place. Do not (!) pull this updated branch yet - the ensuing merge might delete your notebook files and we might lose your jupyter notebooks from the submodule struphy-parameter-files. Make sure you copy those files first to a safe location. We can then add them to the folder examples in the struphy repo.

Please contact me in case of uncertainty.