Dear all, I apologize for my question sent once again too early - the requested functions have of course already been implemented (see https://github.com/dealii/dealii/pull/12603 for anyone interested).
I am currently trying to use this to implement a DG solver for the steady-state Porous medium diffusion equation (https://en.wikipedia.org/wiki/Porous_medium_equation) and will be happy to share the code if I succeed. Thank you very much for your understanding and best regards, Nik On Tuesday, August 5, 2025 at 4:18:13 AM UTC-7 Nik Markus Leuenberger wrote: > Dear all, > > I have a question regarding an *extension* of step.74. > > Eventually, I would like to solve a non-linear Poisson equation using DG > elements. I would also like to use automatic differentiation to obtain the > Jacobian from the Residual as in step-72. > > To start, I wanted to solve the linear problem as if it was non-linear > using a single Newton iteration. > > For the cell_worker and boundary_worker, I was able to use calls like the > following > > std::vector<ADNumberType> old_solution_values(fe_fv.n_quadrature_points); > fe_fv[u_fe].get_function_values_from_local_dof_values(dof_values_ad, > old_solution_values); > > std::vector<Tensor<1, dim, ADNumberType>> old_solution_gradients(fe_fv. > n_quadrature_points); > fe_fv[u_fe].get_function_gradients_from_local_dof_values(dof_values_ad, > old_solution_gradients); > > to then assemble the residual and automatically compute the Jacobian. The > full code is appended. > > For the face_worker, however, I think I would need functions like > std::vector<Tensor<1, dim, ADNumberType>> old_average_of_gradients(fe_fv. > n_quadrature_points); > fe_iv[u_fe].get_average_of_gradients_from_local_dof_values(dof_values_ad, > old_average_of_gradients); > > Since I didn't find these functions, I think they might not exist yet. > > Therefore my question. > - If these functions don't exist, what alternative approach do you see to > get the desired average of gradients of the old solution vector? > > Thank you very much in advance for any help and best regards, > Nik > -- 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 visit https://groups.google.com/d/msgid/dealii/815f0a1b-4f85-4f53-b9cf-3ed5f519785cn%40googlegroups.com.
