[deal.II] Adaptive mesh refinement and solution transfer

Dear DEAL.II developers, I am trying to solve a Laplace problem and refine the mesh adaptively. I am using a CG solver with a symmetric Multigrid preconditioner. I plan to transfer the solution on the initial mesh to the refined mesh, and then use the interpolated solution as the initial

[deal.II] Re: SolverControl stopping criteria

()' function a different quantity instead of the l_2 norm. Best, Yang On Monday, 21 February 2022 at 16:35:37 UTC+3 bruno.t...@gmail.com wrote: > Yang, > > On Monday, February 21, 2022 at 8:00:24 AM UTC-5 Yang Liu wrote: > >> 1. Is it possible to extract the residual vector

[deal.II] SolverControl stopping criteria

Dear DEAL.II developers and users, We are using DEAL.II to address a goal-oriented PDE problem, using SolverCG<> to solve a linear system. Since we are interested in the accuracy of the quantity of interest, we want the iterative solver to stop when the inner product between a dual solution

Re: [deal.II] Building DEAL. II with an external dylib

Thank you! This works. 在 2020年5月15日星期五 UTC+8上午1:59:38，Matthias Maier写道： > > > On Thu, May 14, 2020, at 06:19 CDT, Yang Liu > wrote: > > The /usr/local/hlibpro-2.8.1/lib/libhpro.dylib dynamic library > probably sets an incorrect rpath ("/lib/libhpro.dylib&qu

[deal.II] Building DEAL. II with an external dylib

Dear DEAL.II developers and users, Currently I would like to build DEAL.II with an external library on my macOS Mojave 10.14.5. I edited the cmakelist.txt file as follows: /---/ SET(TARGET "step-6") SET(TARGET_SRC ${TARGET}.cc )

Re: [deal.II] Is there a way to traverse all DoFs?

Thank you for your answer. This helps a lot. I noticed that the active cells are traversed following the "Z-order". And the support points are labeled in a similar way up to removing duplicate points. I wonder if there is a explicit "Z-order" mapping, from {cell->active_cell_index(),

Re: [deal.II] Is there a way to traverse all DoFs?

/ grad phi_j(x_q) fe_values.JxW <https://www.dealii.org/current/doxygen/deal.II/classFEValuesBase.html#abade89efb068b71b7ced7082012a2441>(q_index)); // dx 在 2020年4月28日星期二 UTC+8下午11:52:30，Wolfgang Bangerth写道： > > On 4/28/20 9:48 AM, Yang Liu wrote: > > > > I mean I would

Re: [deal.II] Is there a way to traverse all DoFs?

Hi Wolfgang, I mean I would generate a n by n covariance matrix, based on the locations of these DoFs. 在 2020年4月28日星期二 UTC+8下午11:45:41，Wolfgang Bangerth写道： > > On 4/28/20 8:50 AM, Yang Liu wrote: > > > > I would like to generate a covariance matrix based on these DoF

Re: [deal.II] Is there a way to traverse all DoFs?

： > > On 4/28/20 5:52 AM, Yang Liu wrote: > > > > I would like to ask is there a way to traverse all DoFs in a FE? > > Well, you can always do >for (unsigned int i=0; i ...do something... > to traverse all degrees of freedom. But I suspect that's not what you

[deal.II] Is there a way to traverse all DoFs?

Dear deal developers and users, I would like to ask is there a way to traverse all DoFs in a FE? I would also like to save them in a particular order so that I can associate each DoF with a coefficient and later use them in the matrix assemblage. Best Regards, Yang -- The deal.II project is

[deal.II] Integrate the solution over a subdomain

Dear DEAL.II users, The DEAL.II software provides the function VectorTools::compute_mean_value() to integrate the solution over the whole domain. But now I would like to integrate the solution over a subdomain. fCan anyone give me some suggestions? Thanks in advance! -- The deal.II

[deal.II] On the usage of Utilities::MPI

to hearing from you soon. Thank you! Best Regards, Yang Liu -- 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 "dea