Dear Wolfgang, I am sorry for not having answered to your reply. I started using the constraints at some point and they worked well for my problem. But I came back to analyzing the issue again now.
I eliminated all the coefficients, unnecessary loops and so on, as you
suggested. The code has still somewhat 450 lines, but I tried to leave only
the core parts. As I was eliminating things, I found out, where the problem
occurs, but I have not an explanation for it.
Basically, now I only have one additional factor in my local_matrix (apart
from it's "skeleton") within the Assembly module - density22, which is a
complex value (0,1). When I remove it, the Dirichlet BC get applied as
expected through the MatrixTools::apply_boundary_values function. But when
i leave it in the system - boundary conditions are no longer correct after
solving the system.
Would you please give a hint, what is going on here?
Thank you for your support!
Here is the simplified code.
Kind regards,
Mariia
понедельник, 25 октября 2021 г. в 22:21:31 UTC+2, Wolfgang Bangerth:
> On 10/21/21 7:33 AM, Мария Бронзова wrote:
> >
> > I believe I answered you privately, but I cannot find my email anywhere.
> > In case you didn"t receive it, I am sending such a test case here again.
> > The testcase is a box porous material model with 8 cells to be able to
> > interpolate BC's and to see the problematics clearlier. The confusing
> > thing might be the Time function, as I am working with the frequency
> > domain, but I simply used it artificially as a counter for frequencies
> > to generate a series of output files.
> >
> > The problem appears in the pressure boundary condition, as I am trying
> > to force (250,0) complex-valued boundary values on certain faces. After
> > solving the system, the Dirichlet boundary values are not held, so
> > (250,0) becomes somewhat else.
> >
> > Please, let me know, if some other information is required!
>
> Maria -- the program is still too large. For debugging these kinds of
> problems, it is often useful to strip everything that is not necessary
> out. Set all of your coefficients to one (i.e., omit them from the code
> altogether), remove the loop over frequencies, remove the postprocessing
> steps other than graphical output, etc. I bet that you can reduce your
> current 724 lines to 250 this way, and it will be much easier to see
> what could possibly go wrong in your code then.
>
> Best
> W.
>
>
> --
> ------------------------------------------------------------------------
> Wolfgang Bangerth email: [email protected]
> www: http://www.math.colostate.edu/~bangerth/
>
--
The deal.II project is located at http://www.dealii.org/
For mailing list/forum options, see
https://groups.google.com/d/forum/dealii?hl=en
---
You received this message because you are subscribed to the Google Groups
"deal.II User Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To view this discussion on the web visit
https://groups.google.com/d/msgid/dealii/4654b463-2a1b-4486-95da-0059cf8e1d09n%40googlegroups.com.
Dirichlet_Issue
Description: Binary data
