Re: [deal.II] Step-21 Neumann boundary condition

[Wolfgang Bangerth]

On 5/18/25 12:48, aibike nagyz wrote:
> but in the terminal it shows:
> Timestep 1
>    22 CG Schur complement iterations for pressure.
>    0 CG iterations for saturation.
>    Now at t=0.01, dt=0.01.
>
> Timestep 2
>    6 CG Schur complement iterations for pressure.
>    0 CG iterations for saturation.
>    Now at t=0.02, dt=0.01.
>
> Timestep 3
>    0 CG Schur complement iterations for pressure.
>    0 CG iterations for saturation.
>    Now at t=0.03, dt=0.01.
>
> Timestep 4
>    0 CG Schur complement iterations for pressure.
>    0 CG iterations for saturation.
>    Now at t=0.04, dt=0.01.
>
> Clearly the saturation front is not entering the domain. And the visualization
> of pressure and velocity are strange. I don't understand is it because I set
> the boundary ids wrong? Can anyone please guide me where seem to be the problem?

I don't know what the issue is, and you are the one best positioned to debug
the problem. Generally, you'd approach the problem by going back to the
original code and start making one change at a time -- observe that the result
is what you expected -- commit the code to a repository so that you have a
state you know is correct -- and then make the next change. If you have
specific hypotheses about what's wrong, you can of course also test that
specifically; for example, if you wonder whether you have set boundary
conditions correctly, you can use the class here

https://dealii.org/developer/doxygen/deal.II/classDataPostprocessors_1_1BoundaryIds.html

https://dealii.org/developer/doxygen/deal.II/DEALGlossary.html#GlossBoundaryIndicator
to create output boundary ids and visualize them.

Best
W.

--
------------------------------------------------------------------------
Wolfgang Bangerth email: bang...@colostate.edu
www: http://www.math.colostate.edu/~bangerth/