[deal.II] deal.II Newsletter #48
Hello everyone! This is deal.II newsletter #48. It automatically reports recently merged features and discussions about the deal.II finite element library. ## Below you find a list of recently proposed or merged features: #7178: Add an assertion. (proposed by bangerth) https://github.com/dealii/dealii/pull/7178 #7177: document TrilinosWrappers::SparseMatrix VectorType (proposed by tjhei; merged) https://github.com/dealii/dealii/pull/7177 #7176: Use correct dof sets in case of multigrid renumbering. (proposed by jodlbauer) https://github.com/dealii/dealii/pull/7176 #7173: Extend constrained linear operators to work with any LinOp Payload. (proposed by jppelteret) https://github.com/dealii/dealii/pull/7173 #7172: Mark EpetraExt as optional (proposed by masterleinad; merged) https://github.com/dealii/dealii/pull/7172 #7171: Allow TrilinosWrappers::SparseMatrix::mmult for 64-bit indices (proposed by masterleinad; merged) https://github.com/dealii/dealii/pull/7171 #7170: Update the glossary entry on distorted cells. (proposed by drwells; merged) https://github.com/dealii/dealii/pull/7170 #7169: Replace SparsityPattern::Iterator with LinearIndexIterator. (proposed by drwells; merged) https://github.com/dealii/dealii/pull/7169 #7168: Fix hp_step-40_variable_01 (proposed by masterleinad; merged) https://github.com/dealii/dealii/pull/7168 #7167: Improve the MatrixFree doxygen formatting (proposed by drwells) https://github.com/dealii/dealii/pull/7167 #7166: Add some more entries to .gitignore. (proposed by drwells; merged) https://github.com/dealii/dealii/pull/7166 #7165: Shared tria improvements (proposed by drwells; merged) https://github.com/dealii/dealii/pull/7165 #7164: Silence another warning in compute_n_point_to_point_communications (proposed by masterleinad; merged) https://github.com/dealii/dealii/pull/7164 #7163: small typo (proposed by tcclevenger; merged) https://github.com/dealii/dealii/pull/7163 #7161: [WIP] Add jenkins (proposed by tjhei) https://github.com/dealii/dealii/pull/7161 #7160: quadrature: improve documentation (proposed by tjhei; merged) https://github.com/dealii/dealii/pull/7160 #7159: Extend DoFRenumbering::hierarchical(). (proposed by bangerth; merged) https://github.com/dealii/dealii/pull/7159 #7158: Check for the ability to call non-constepr functions in constexpr ones (proposed by masterleinad; merged) https://github.com/dealii/dealii/pull/7158 #7157: Avoid warning about unused variable in compute_n_point_to_point_communications (proposed by masterleinad; merged) https://github.com/dealii/dealii/pull/7157 #7156: Fix: 'iota' is not a member of 'std' (proposed by mwichro; merged) https://github.com/dealii/dealii/pull/7156 #7154: Avoid some warnings compiling with CUDA support (proposed by masterleinad; merged) https://github.com/dealii/dealii/pull/7154 #7152: Use workaround in 'execute_transfer()' only on older p4est versions. (proposed by marcfehling; merged) https://github.com/dealii/dealii/pull/7152 #7151: Add another test for hp parallel DoFs. (proposed by bangerth; merged) https://github.com/dealii/dealii/pull/7151 #7150: Add a test for parallel hp handling. (proposed by bangerth; merged) https://github.com/dealii/dealii/pull/7150 #7149: Add a test. (proposed by bangerth; merged) https://github.com/dealii/dealii/pull/7149 #7145: Implement a function for computing the number of messages received per MPI process (proposed by eldarkh; merged) https://github.com/dealii/dealii/pull/7145 #7134: Work around ICC issue with __builtin_expect in constexpr function (proposed by masterleinad; merged) https://github.com/dealii/dealii/pull/7134 #7091: Matrixfree vectorization mask (proposed by schoeder; merged) https://github.com/dealii/dealii/pull/7091 #6976: AD Helpers: Introduce base class for cell-level AD helper classes (proposed by jppelteret; merged) https://github.com/dealii/dealii/pull/6976 #6965: Change step-16: add alternative without MeshWorker (proposed by tjhei; merged) https://github.com/dealii/dealii/pull/6965 ## And this is a list of recently opened or closed discussions: #7175: dealii-9.0.0.dmg OSX and "illegal instruction 4" (opened) https://github.com/dealii/dealii/issues/7175 #7174: Bug in parallel component-wise multigrid renumbering (opened) https://github.com/dealii/dealii/issues/7174 #7162: Bug in hp::DoFHandler::distribute_dofs() (opened) https://github.com/dealii/dealii/issues/7162 #7155: Compilation issue (opened and closed) https://github.com/dealii/dealii/issues/7155 #7153: Zero volume (cohesive) elements (opened) https://github.com/dealii/dealii/issues/7153 #7148: Using deal.II-9.0.0 Docker image with my C++ project seems to require C++14 (closed) https://github.com/dealii/dealii/issues/7148 #7146: error when compiling (closed) https://github.com/dealii/dealii/issues/7146 #7074: boost 1.68.0 fails serialization configure check (closed) https://github.com/dealii/dealii/issues/7074
Re: [deal.II] implementing traction neumann boundary
Hi Jean-Paul
I was able to find my error. When generating my grid i neglected to set
colorize to 'true' and as a result i only had boundary IDs of zero for all
my faces. My thought at the time was that the interpolate_boundary_values()
was negating my Nuemann contributions, but i realize now that commenting it
out was silly of me. My code now generates physically sensible results.
Thank you very much and kind regards
Shaqir
On Friday, August 31, 2018 at 10:19:09 AM UTC+2, Jean-Paul Pelteret wrote:
>
> Hi Shaqir,
>
> So, I have a short question to clarify something: By commenting out the
> calls to VectorTools::interpolate_boundary_values()
> and MatrixTools::apply_boundary_values(), are you implying that you do not
> impose equation 4 (the Dirichlet boundary conditions) anywhere on the
> boundary of the domain? If so then your problem is indeterminate, and I bet
> you’d get a non-trivial solution once you constrain at least one boundary
> surface.
>
> Best,
> Jean-Paul
>
> On 31 Aug 2018, at 10:09, shaxr...@gmail.com wrote:
>
> Hi all, I am quite new to deal.ii and have been tasked with making use of
> it for my Honours project.
>
> I am trying to implement a traction boundary condition for a steady state
> linear elasticity type problem. I am therefore using step-8 from the
> tutorials as the base for my project and have attempted a Neumann boundary
> implementation by making use
> of a combination of the step-44 traction boundary and lecture 21.55 by
> Prof Bangerth. The code that I added is as follows:
>
>
> for (unsigned int f=0; f::faces_per_cell; ++f)
> if (cell->face(f)->at_boundary() == true
>&& cell->face(f)->boundary_id() == 3)
> {
> fe_face_values.reinit (cell, f);
>
> for (unsigned int fq_point=0; fq_point ++fq_point)
> {
>const Tensor<1, dim> &N =
> fe_face_values.normal_vector(fq_point);
>static const double T = 7.0e6; // arbitraty magnitude
> of traction
>const Tensor<1,dim> traction = (T * N);
>
> for (int i = 0; i {
> const unsigned int component_i =
> fe.system_to_component_index(i).first;
>
> cell_rhs(i) += traction[component_i]
> *
> fe_face_values.shape_value(i,fq_point)
> * fe_face_values.JxW(fq_point);
> }
>
> }
> }
>
> The code runs without any errors, however when I set my body force to zero
> to check for the traction contribution I noticed that there seems to be no
> displacement at all. I have also commented out the interpolate boundary
> values as follows:
>
> // std::map boundary_values;
> // VectorTools::interpolate_boundary_values (dof_handler,
> 0,
>
> Functions::ZeroFunction(dim),
> boundary_values);
>
> // MatrixTools::apply_boundary_values (boundary_values,
> system_matrix,
> solution,
> system_rhs);
>
>
> I have also tried increasing the magnitude of my traction but that doesn’t
> make a difference either. I am currently at my wits end as to why this
> would be the case for there being zero displacement. I have attached the
> formulation with regards to the weak form of my problem
> as to give more insight into what I am trying to implement. Any form of
> help or insight as to what the cause of the issue might be would be greatly
> appreciated.
>
> Kind regards
> Shaqir
>
> --
> 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 dealii+un...@googlegroups.com .
> For more options, visit https://groups.google.com/d/optout.
>
>
>
>
--
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 dealii+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
