Dear Wolfgang, I had indeed a gap in understanding of some points, it is solved now! I thank you a lot for your time and effort in trying to understand my problem and to help me out!
With kind regards, Mariia среда, 6 октября 2021 г. в 03:02:52 UTC+2, Wolfgang Bangerth: > On 9/30/21 6:49 AM, Мария Бронзова wrote: > > So, there are two boundary integrals in the formulation and I am trying > to > > implement them for the case of fixed displacement boundary condition. > The > > first boundary integral falls to zero for such a BC, as no displacement > > variation is possible in this case. So, I am implementing the second > integral > > from the second equation. The integral can be represented as in the > BC.PNG > > file. There are those factors in brackets, assigned displacement values > and > > variation of the pressure multiplied together: > > > > *local_rhs*(i) += -porosity*(1.+Q[k]/R[k]) > > *d_boundary_values[q] > > *fe_face_values[pressure].value(i,q) > > *fe_face_values.JxW(q); > > > > But the way it is written now it cannot work, as the *d_boundary_values* > term > > is a vector of vectors (as we have three displacement components). So, > the > > question is, whether there is a way to treat the displacement components > > seperately in this *d_boundary_values* term? Or maybe even a smarter way > to > > make it work? > > I think that your question is actually of mathematical nature, not one of > implementation. If I read the integral I_2 correctly in your previous > email, > then what you prescribe there is > u^i_n > which I believe is not actually the displacement on the boundary (a > vector) > but only the *normal component* of the velocity (a scalar). So you have > two > options: > - You write a function that only returns the normal velocity (which is all > you can prescribe anyway) > - You write a function that returns the velocity at the boundary as a > vector > and then in the bilinear form, you take the dot product with the normal > vector (which you can get from the fe_face_values object). > > Both are reasonable, though if all you can prescribe is the normal > component, > you might as well write your function in such a way that that is what it > returns. > > 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/68727db2-9773-4cf0-9fe8-a2221bd85694n%40googlegroups.com.
