Dear David,
for debugging your code you should start with simplifying your code as much
as possible to see what is going wrong.
In particular, you should try with not refining your mesh and confirm that
the assembled matrices look as they should.
Can you modify the parameters in such a way that
> Do you know how to display this data file so that they can show the values
> in correct order just like in MATLAB.
>
> Best regards,
> David.
>
>
>
> On Tuesday, May 17, 2016 at 3:30:22 AM UTC+8, Daniel Arndt wrote:
>>
>> >Dear David,
>> >
>&g
David,
assuming that you want to access the gradients at the support points of the
element you are using, you can simply create a
appropriate Quadrature object to initialize FEValues[1]. In this case the
quadrature points correspond to the local DoF indices.
Apart from that, you can use DataPos
David,
the code snippet you posted is not going to work. You can't use a FEValues
without initializing it with a cell_iterator.
If all the information you need for your postprocessing is contained in a
data vector based on a single DoFHandler, then DataPostprocessor is what
you want to use.
If
Hey David,
Yes, this won't work since FEValues need a cell to reinit to get mapping
> data. I looded at the class DataPostprocessor as you mentioned, but don't
> quite understand the parameters passed to compute_derived_quantities_vector
> method. I don't understand what exactly is the paramete
Dear Rajat,
After the mesh partitioning is done, in a
parallel::distributed::Triangulation the local meshes are independent of
each other. In particular, moving a vertex on one process does not change
anything for all the other processes.
Therefore, it is necessary to move the vertices of a giv
that the dof is a locally owned dof by using the is_element
> of the indexset of locally owned dofs help ?
> Will this be slow ?
>
>
> On Wednesday, May 25, 2016 at 7:19:27 AM UTC-4, Daniel Arndt wrote:
>>
>> Dear Rajat,
>>
>> After the mesh partitioning is done,
Bastian,
can you provide us with a minimal example that shows your problem? What you
are trying to do should be possible even if it might not be the right
thing to do in your situation.
Best,
Daniel
Am Mittwoch, 1. Juni 2016 02:10:36 UTC+2 schrieb Bastien Lauras:
>
> Hi,
>
> I am working on a 2
e from?
>
> Many thanks!
> Best,
>
> Bastien
>
>
> On Wednesday, June 1, 2016 at 5:43:48 AM UTC-5, Daniel Arndt wrote:
>>
>> Bastian,
>>
>> can you provide us with a minimal example that shows your problem? What
>> you are trying to do should
Hamed,
There shouldn't be a problem to do what you following step-22. In
particular, you can modify step-18::get_strain as you proposed.
Did you face any difficulties so far?
Best,
Daniel
--
The deal.II project is located at http://www.dealii.org/
For mailing list/forum options, see
https://g
ere isn't any
shape_grad_component function, I am not sure that I can update the
get_strain as I mentioned.
I was wondering if you have already written a code for a thermoelastic
problem in dealii, solving heat equation and elastic equation separately.
Best,
Hamed
On Thursday, June 2, 2016
is working.
> constraints.close();
> }
>
> When I try to compile my file, I get the following error :
> error: undefined reference to 'void
> dealii::DoFTools::make_periodicity_constraints
> >(dealii::DoFHandler<1, 1> const&, unsigned char, unsigned c
d stored on a quadrature
> point related to elastic problem DofHandler, in order to solve
> stress-induced heat equation problem. In other words, I don't know how to
> exchange information between two equation.
>
> Thanks,
> Hamed
>
> On Friday, June 3, 2016 at 3:4
Bastien,
as there already is the Polynomial class HermiteInterpolation you should
create a new FiniteElement based on a
TensorProductPolynomial. A good way might be to copy
FE_Q_Base and modify it accordingly.
Of course you need to think about how to interpolate values using this
element, how
{
>>>const Point face_center = cell->face(f)->center();
>>>if (face_center[1] == 0) //Bottom
>>> cell->face(f)->set_boundary_id (0);
>>>else if (face_center[1] == 1) //Top
>>> cell->face(f)->set_boundary_id (2);
>>
Mohammad,
first of all
const double old_s = old_solution_values[q_point](1);
const double old_s = old_solution_values0[q_point](1);
const double old_s = old_solution_values1[q_point](1);
looks weird. Are you redefining old_s all the time?
Apart from that it looks so far good to me. Can you
gt;> I agree with Daniel. So in this case, you probably just need to change
>> your order of operations and set the boundary_id's before doing any
>> refinement.
>>
>> On Wednesday, June 8, 2016 at 2:07:09 AM UTC+2, Daniel Arndt wrote:
>>>
>&
Pankaj,
Hello,
> I am trying to write a code to calculate buoyancy force which is the
> product of temperature and gravity. In my case temperature is available.
>
>
>
> TrilinosWrappers::MPI::Vector temperature ;//given
> const dealii::Point gravity = -( (dim == 2) ? (dealii::Point (
> 0,1))
Praveen,
I solve a PDE in serial and save the solution to file.
>
Do you mean that your triangulation is not a
parallel::distributed::Triangulation or that you are simply running with
one process?
> Is there a way to now read this solution into a
> TrilinosWrappers::MPI::Vector object ?
>
If
Praveen,
Am Montag, 20. Juni 2016 18:05:12 UTC+2 schrieb Praveen C:
>
> Hello Daniel
>
> I have a normal Triangulation and I solve on this using a Vector.
> I could save this to file calling Vector::print.
>
> Is it now possible to read this into a TrilinosWrappers::MPI::Vector ?
>
No, there is n
Dear ?,
dealii::IndexSet::ElementIterator::operator*() returns the stored
global_dof_index [1].
Therefore, all you have to do is to dereference index:
dealii::types::global_dof_index gdi = *index
Best,
Daniel
[1]
https://www.dealii.org/8.4.1/doxygen/deal.II/classIndexSet_1_1ElementIterator.h
Pasha,
First of all: The sum of values in the right-hand side vector only
corresponds to a sum of values of the represented right-hand side function,
if you use interpolating Finite Elements (such as FE_Q).
Furthermore, you have to be sure that you are really interested in the
values and not in
Hamed,
Have a look at how temporary non-ghosted vectors are used in step-40.
Similar to VectorTools::interpolate
the solution vector in SolverCG::solve must not be ghosted.
Best,
Daniel
Am 07/01/2016 um 01:15 AM schrieb Hamed Babaei:
> Hi Daniel,
>
> I hope you are doing well.
> I got an error i
Ehsan,
All I can say: After switching the order of arguments in SparseMatrix::add,
your code runs for me with a recent developer version and Trilinos at least.
Best,
Daniel
Am Montag, 4. Juli 2016 18:59:05 UTC+2 schrieb Ehsan Esfahani:
>
> Dear Professor Bangerth,
>
> Thanks for your response.
Pablo,
Did you see https://groups.google.com/forum/#!topic/dealii/rTx7Fea65dM ?
Can you try without Opencascade and a clean build directory? Do you get
this error also with the latest release tarball?
Best,
Daniel
Am Montag, 4. Juli 2016 10:15:34 UTC+2 schrieb Pablo Perez del Castillo:
>
> Hell
Benhour,
Have a look at the implementation of GridGenerator::half_hyper_ball [1].
You probably just want to use the first 4 vertices and then
p,
p+Point<2>(0,-1) *(radius/std::sqrt(2.0)*a),
p+Point<2>(0,-1) *radius,
additionally.
Can you specify in formulas what symmetry boundary conditions mea
Benhour,
Did you try what I suggested in
https://groups.google.com/d/msg/dealii/QPLvZIQx7yw/LQ7t0-xWBgAJ ?
Best,
Daniel
--
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
Benhour,
It would be helpful if you tell us what the problem with the newly created
problem is.
Looking at the implementation of GridGenerator::half_hyper_ball [1],
you can observe that there are cells which have vertices both with positive
and negative
y-components. This will likely lead to und
Anup,
if you know which node you want to constrain, you just have to find the
corresponding degree of freedom and add a constraint to the
ConstraintMatrix.
For finding this DoF, DoFTools::map_dofs_to_support_points[1] might be
useful. The alternative is to loop through all the cells and ask eac
Benhour,
you can find a modification of step-3 in which the domain is a quarter
circle attached.
The four cells from GridGenerator::half_hyper_ball have been replaced by
just three.
The vertex (0,-radius) has been removed and all the remaining vertices
with a negative y-component have been move
Junchao,
You want to use parallel::distributed::SolutionTransfer instead if you are
on a parallel::distributed::Triangulation
Executing
$ grep -r "parallel::distributed::SolutionTransfer" .
in the examples folder, tells me that this object is used in step-32,
step-42 and step-48.
Have for exampl
Junchao,
It seems that the documentation is outdated for this piece of information.
In fact, neither PETScWrapper::MPI::Vector nor TrilinosWrappers::MPI::Vector
does have update_ghost_values.
What you should do is exactly what is done in the few lines of step-42 you
referenced.
"solution = distri
Junchao,
For what matters here, you are just interesting in how to deal with
boundary constraints for a time-dependent problem.
I am not quite sure what your problem is. You have time-dependent boundary
conditions so you want to use boundary
conditions corresponding to the correct time step. Wh
Junchao,
>> The interpolated solution should already have the correct (old) boundary
>> conditions.
>>
> Is it true? As you see, I call
> constraints.distribute(interpolated_solution) to impose boundary
> constraints, which are computed in setup_system() as follows
>
What I meant to say is tha
Pasha,
there are tutorial programs, such as step-22[1], that use BlockMatrices and
solve for individual blocks.
What exactly is the problem you are facing? Does it have anything to do
with Trilinos objects or is it a conceptual problem?
Can you tell us how exactly you want to split solving you
Jiaqi,
what you really want is that
(u_n_plus_1, v_h) = (u_stat-dt*dp)
holds for all v_h in your velocity ansatz space. This means inverting a
mass matrix for the velocity
and may come down to just assigning function values depending on what your
ansatz spaces are.
In that case, you can loop ov
Pasha,
> *if* (direct_solver)
>
> {
>
> SolverControl cn;
>
> TrilinosWrappers::SolverDirect solver(cn);
>
> solver.*solve*(system_pde_matrix.block(0,0),
>
> newton_update.block(0), system_pde_residual.block(0));
>
>
> constraints_update.distribute(newton_update);
>
>
> *return* 1;
>
> }
>
>
>
>
Ehsan,
can you show us what your "detailed.log" in the build directory looks like?
This file stores the configuration that is used for building deal.II.
` -DCMAKE_INSTALL_PREFIX= /home/General_for_All_Users/deal_II_install_dir`
should tell deal.II to copy the build library into that folder
when y
Deepak,
could you provide us with a minimal example that shows your problem?
Are you running in debug mode? What exactly does the program tell you when
aborting?
Best,
Daniel
--
The deal.II project is located at http://www.dealii.org/
For mailing list/forum options, see
https://groups.google.
Deepak,
without a working example I can run, I don't see an obvious error.
Note that there are two tests, namely "numerics/create_point_source_hp" and
"hp/vectors_point_source_hp_01"
that test `VectorTools::create_point_source` for hp. Maybe this helpful for
debugging.
Best,
Daniel
--
The dea
Jiaqi,
the algorithm in step-35 is the one that is proposed in
"L. Timmermans, P. Minev, and F. Van De Vosse, “An approximate projec-
tion scheme for incompressible flow using spectral elements”, International
journal for numerical methods in fluids, vol. 22, no. 7, pp. 673–688, 1996."
The veloc
krei,
If you want to solve different PDEs on different domains that can be
discretized by a common mesh, the preferred approach is to use a hp-vector
finite element.
This means that on each of your subdomains all blocks of your finite
element but one are of type FENothing.
You might want to hav
Daniel,
1.) You can use complex valued algebra if you split your problem into two
equations. Apart from that most of the algebra objects can be used with
std::complex.
At the moment, this is not true for constraints given by a ConstraintMatrix
object, but this is WIP (https://github.com/dealii/
Wolfgang,
Interesting that this isn't even mentioned in the introduction to that
> program, or anywhere else, for that matter. How did you find out?
>
A major part of my PhD thesis is based on these projection methods.
Reading the literature this seems to be the first paper in which the
rotatio
Marek,
You would just set up your ConstraintMatrix anew. This is also the way it
is done in step-44.
Best,
Daniel
--
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 becau
t;>>
>>>> How should I solve this problem.
>>>> Best regards.
>>>> Ehsan
>>>>
>>>> On Monday, August 1, 2016 at 2:21:31 PM UTC+2, Ehsan wrote:
>>>>>
>>>>> Dear Daniel,
>>>>&g
It seems that there is something wrong with your TBB installation. Which
version of TBB are you using?
Could you send your "detailed.log" file located in the deal.II build
directory?
TBB is also bundled with deal.II and you can force to use that via
$ cmake -DDEAL_II_FORCE_BUNDLED_THREADS=ON .
i
Pete,
I remember someone having something like this in mind. This is the link to
that discussion:
https://groups.google.com/forum/#!topic/dealii/QKuDndKojug
A different way would be to derive from DataOut and overwrite
DataOut::first_cell and DataOut::next_cell by skipping all the cells that
d
Oded,
If you include
ADD_CUSTOM_TARGET(debug
COMMAND ${CMAKE_COMMAND} -DCMAKE_BUILD_TYPE=Debug ${CMAKE_SOURCE_DIR}
COMMAND ${CMAKE_COMMAND} --build ${CMAKE_BINARY_DIR} --target all
COMMENT "Switch CMAKE_BUILD_TYPE to Debug"
)
ADD_CUSTOM_TARGET(release
COMMAND ${CMAKE_COMMAND} -DCMAKE_
make target `debug'. Stop.
>
> [ yaakobioy L02029080
> ~/dealii/dealii-8-5-0pre-v2/dealii-8-5-0pre-v2-Install/examples/step-1-mod
> ]$ make release
>
> make: *** No rule to make target `release'. Stop.
>
> Thanks,
> Oded
>
>
> On
nderstood and I can somehow evaluate the boundary
> conditions? Or what would a better approach be?
>
> On Saturday, July 30, 2016 at 1:38:06 PM UTC+3, Daniel Arndt wrote:
>>
>> krei,
>>
>> If you want to solve different PDEs on different domains that can be
Giorgis,
This looks like a problem with your PETSc installation. Can you confirm
that it is installed correctly?
Do any problems appear if you compile deal.II without PETSc?
Best,
Daniel
Am Donnerstag, 11. August 2016 22:16:49 UTC+2 schrieb Giorgos Kourakos:
>
> Hi,
>
> I'm trying to compile a
Joel,
Does the equation you want to solve for have multiple components? Otherwise
it would not make sense to multiply in the right hand side something with a
gradient.
If you want to project the gradient into a vector valued finite element
ansatz space, then the matrix you are assembling looks
Joel,
Yes the matrix you are assembling is a vector-valued mass matrix now.
For me your code is failing with
void dealii::FEValuesBase::get_function_gradients(const
InputVector&, std::vector >&) const [with InputVector =
dealii::Vector; int dim = 3; int spacedim = 3; typename
InputVector::valu
Bruce,
The p_local_solution vector is filled with values as the result of solving
> a linear equation, p_local_rho is filled with values by looping over
> individual elements and assigning values calculated from other quantities.
> Part of the reason the p_local_rho vector doesn't have ghost no
Benhour,
What exactly is your problem? Lots of the examples programs, i.e. step-14,
step-15, step-18, step-31, step-32, step-35, step-43 and step-44, use
FEValues::get_function_gradients.
Best,
Daniel
Am Montag, 15. August 2016 03:13:14 UTC+2 schrieb benhour@gmail.com:
>
> Dear All,
> I wa
Bruce,
[...] 3. Like in 2., but for adding values to individual elements. Use
> VectorOperation::add.
>
> I guess 3) is obscure. Does this mean that you are adding a new element
> where none existed before or that you are incrementing an existing element?
> If the former, aren't the total numb
Bruce,
You were just missing
p_rho.compress(VectorOperation::insert);
after assemble_system().
Best,
Daniel
Am Mittwoch, 17. August 2016 21:01:22 UTC+2 schrieb bjpal...@gmail.com:
>
>
>
> On Tuesday, August 16, 2016 at 2:47:18 AM UTC-7, Daniel Arndt wrote:
>>
>> Bruce,
&
dealii.org is online again.
Am Mittwoch, 17. August 2016 19:33:48 UTC+2 schrieb Wolfgang Bangerth:
>
> On 08/17/2016 11:30 AM, thomas stephens wrote:
> > Currently unable to load https://www.dealii.org/ along with tutorials
> > (problem began at approximately 12:30pm ETD)
>
> Yes, there is a po
Deepak,
I am aware of the fact that ConstraintMatrix.n_constraints() gives the
> number of hanging nodes for h-refinement, but if I only use p-refinement,
> can it give the number of hanging support points occurring due to the
> difference in the order of bases between two adjacent elements? I
at?
>
> Best,
> Joel
>
> On Friday, August 12, 2016 at 5:35:07 PM UTC+2, Daniel Arndt wrote:
>>
>> Joel,
>>
>> Yes the matrix you are assembling is a vector-valued mass matrix now.
>> For me your code is failing with
>>
>> void dealii::FEValuesBase:
An old version of step-12[1] has an implementation that doesn't use
MeshWorker.
In my opinion, it is covenient to use MeshWorker if you can when you want
to consider discontinuous ansatz spaces due to the
different cases that can occur for face contributions. If you want or need
to use more than
Dear Franck,
> I am looking at step-43 using different method. I do a semi discretization
> in space and end up with an ode for the saturation and use different
> technique for time integration for the saturation.
>
> For this I will need to assemble the matrix H after solving for the
> veloci
Yongchao,
If you are still interested in using MeshWorker, there are some more
examples (and in particular Stokes) at
https://bitbucket.org/guidokanschat/amandus
Best,
Daniel
Am Montag, 29. August 2016 11:15:22 UTC+2 schrieb yocz...@gmail.com:
>
> Dear Wolfgang ,
> Thanks for your advice, I ha
>
>
> There actually *is* a function that takes two DoFHandler objects, but it
> doesn't take a constraint matrix or the last boolean argument.
>
[...]
> You can of course see what variants of the make_sparsity_pattern function
> there are by looking here:
>https://www.dealii.org/8.4.1/do
Eldar,
As you already found out, interpolate_boundary_values can only be used with
a ComponentMask if the element you are using is primitive.
Therefore, the workaround you are using seems to be suitable for the moment.
Since BDM1 is div-conforming, project_boundary_values and
project_boundary_v
Joel,
In GridTools::find_cells_adjacent_to_vertex(dof_handler,center_id) you need
to give the number of the vertex_number as center_id, not the number of the
degree of freedom.
Then, you want to use
Quadrature
quadrature_formula(dof_handler.get_fe().get_unit_support_points());
to create a Q
Kyusik,
Can you summarize what you have tried and which errors/problems you got
with these approaches?
What exactly do you mean with "csc(theta) or sec(theta) can't be used in
cylindrical coord system"?
Best,
Daniel
Am Freitag, 9. September 2016 06:17:44 UTC+2 schrieb hank...@gmail.com:
>
> Th
Rajat,
Since you are using p4est and PETSc, I suspect that you want to run with
more than one MPI process.
If you are using a parallel::distributed::Triangulation you will probably
run into trouble when the pair of degrees of freedom
you want to identify are not known by the same process.
You mi
Rajat,
The first equation is the usual equilibrium equation. It is solved to give
> displacements. The stiffness matrix is then multiplied by this global
> displacement vector to give the global traction vector which forms the rhs
> for the second equation.
>
The easiest solution to deal with c
Jiaqi,
Currently you can access first, second and third partial derivatives in
FEValues. The FiniteElement class also implements fourth derivatives.
If this is sufficient for your purposes, then using
FEValues*::third_derivative should be fine for you. I don't see a easier
way at the moment.
B
Jaekwank,
You would indeed expect significantly better results with higher polynomial
degree on the same mesh, if the solution is sufficiently regular.
What kind of mesh are you using? Are you resolving all the boundary layers
suitably? Are you using a Mapping of the same order as your velocity
Franck,
One possibility is to use A.mmult(C,B,V) where B is an IdentityMatrix.
Another way is to create an IdentityMatrix B, modify its entries suitably
and use A.mmult(C,B).
A third approach would be to just write the the scaling yourself.
Best,
Daniel
Am Sonntag, 18. September 2016 18:01:17 U
Joel,
> [...]
> for (unsigned int q_index=0; q_indexfor (unsigned int i=0; i {
> velocity[local_dof_indices[i]] = local_velocity_values[q_index];
> }
>
> Here you assign the same Tensor<1,dim> at all DoFs multiple times and you
end with velocity[*]=local_vel
Jaekwang,
This seems to be unrelated. Did you setup the Triangulation before calling
fe.distribute_dofs()?
Best,
Daniel
Am Dienstag, 20. September 2016 17:04:02 UTC+2 schrieb JAEKWANG KIM:
>
> yes...I tried that before, but I get an error message as follow..
>
> As you mentioned, I first learn
...of course I meant dof_handler.distribute_dofs(). Does the DoFhandler
know at this point about the Trinagulation?
Best,
Daniel
Am Dienstag, 20. September 2016 17:58:02 UTC+2 schrieb Daniel Arndt:
>
> Jaekwang,
>
> This seems to be unrelated. Did you setup the Triangulation be
Jaekwang,
Have a look at step-10. There you see how the order of the mapping
influences the error.
Basically, you have to set the mapping for your FEValues objects, when you
compute error via integrate_difference
and when you project or interpolate some Function to an FE Vector.
Best,
Daniel
A
Chenchen,
We only support to read in meshes that are not refined. Translated to your
situation this means that you are only able to use Q1 meshes as initial
grid.
Of course, you can then use triangulation.set_manifold to define the
correct behavior when you refine your mesh within dealii along
Juan,
[...]
> 1) My first question is about how the "direction" works, is it 1
> correspond to "x" and 2 corresponds to "y"?
>
No, direction 0/1/2 means that the support_points of the dofs to be
identified just differ in the x-/y-/z-component.
What you are doing should be perfectly fine.
> 2)
Marek,
[...] I would like to process it in dealii by rotating it around y-axis, to
> get 3D grid.
> Is it possible in dealii?
>
No, there is no such function yet.
> [...]
> My objective is the mesh in the file the attached file perf_chamber.png,
> which was obtained by merging
> the inner cy
Marek,
the attached file should be close to what you want to do.
Basically, you need to create two geometries: One for the cylinder shell
and one for the cylinder.
Here, we take advantage of the fact that GridGenerator::cylinder always
creates four subdivisions w.r.t height.
Furthermore, the cyl
Joel,
Its been a week since my latest post, I was just wondering if you could
> help me.
>
Did you try what I suggested? What were the problems?
Looking at the latest files you sent, local_velocity_values has still not
the type std::vector >
and you are still using
for (unsigned int q_index=0;
Everything should be available again.
Am Mittwoch, 28. September 2016 15:19:15 UTC+2 schrieb sotelo...@gmail.com:
>
> it is already 28th, the server still down. is it going to take longer
> than announced?
> Edith
>
> On Monday, September 12, 2016 at 11:21:13 AM UTC-5, Guido Kanschat wrote:
>>
>
Joel,
I did try it, but I cant just change the local_velocity_values from a
> std::vector> to std::vector. Then the
> fe_values[velocities].get_function_values function doesnt compile.
>
That's true, but you should use std::vector > and not
std::vector, see also [1].
> (correct me if wrong)
Erik,
Am Donnerstag, 29. September 2016 07:32:04 UTC+2 schrieb Erik Svensson:
>
> Hi,
>
> I’m trying to implement homogeneous Neumann bc on part of the boundary in
> tutorial example step-20.
>
> The Neumann bc should be imposed strongly (as for Dirichlet bc in normal,
> non-mixed, formulation
Joel,
Yes, this looks about right. I would really replace "std::vector double
x,y,z" by "Vector x,y,z".
In the end, entries in these vectors only make sense with the corresponding
DoFHandler.
This would also allow you to use these Vectors for postprocessing in
deal.II.
Best,
Daniel
--
The de
Marek,
[...]
>
I want to ask a question, which could be interesting for other users of
> dealii-
> do we have some bulletproof procedure, how to produce (not so simple)
> geometries,
> meshes in gmsh and then merge them "safely" in the dealii? Safely meaning-
> without
> the creation of the do
Metin,
[...]
> For the vectors, the implementation (for the else case) is as following;
>
> *for* (*unsigned* *int* shape_function=0;
>
>shape_function
> {
>
> *for* (*unsigned* *int* d=0; d<*spacedim*; ++d)
>
> *if*
> (shape_function_data[shape
Marek,
Only the vertices with indices in considered_vertices are tested for
>> equality. This speeds up the algorithm, which is quadratic and thus quite
>> slow to begin with. However, if you wish to consider all vertices, simply
>> pass an empty vector.
>>
>
> i.e. I have assumed it is called
Metin,
I understand that the components are in space (x/y/z),
> and (*shape_gradient_ptr++)[d] would be derivative of phi in each
> direction; is that correct?
>
Yes, that is the derivative of the dth component in direction d.
>
> (compared to vectors) the implementation of divergence function
Kyusik,
As Wolfgang already mentioned, you want to use
FEValues::get_function_values()
to extract from a global FE Vector the local values at the quadrature
points in the cell you are on.
These quadrature points normally don't coincide with with the mapped unit
support points of the
FiniteElem
Edith,
If I understand you correctly, you basically consider step-7 and replace
the Neumann boundary conditions by inhomogeneous Dirichlet boundary
conditions.
The system matrix is symmetric and positive definite and hence CG should
work.
Can you verify that the solution is correct if you use
Lisa,
> [...]
> I do not know how to output below values:
>
> 1- u_r
> 2- sigma_xx, sigma_yy
> 3- sigma_rr , sigma_tt
>
For computing these values you should have a look at the DataPostProcessor
class [1]. Its use is explained in step-29.
What you basically want to do for u_r in there is somethi
Hamed,
>>> It still works with Petsc and Trilinos is not implemented at all. I
>>> appreciate it if you could let me know what changes should be made in
>>> step-40 to make it work with Trilinos.
>>>
>> What exactly, do you mean by that? So you have both PETSc and Trilinos
installed and step-4
In case there is an error with Trilinos, what is it?
Am Sonntag, 16. Oktober 2016 17:27:26 UTC+2 schrieb Daniel Arndt:
>
> Hamed,
>
>
>>>> It still works with Petsc and Trilinos is not implemented at all. I
>>>> appreciate it if you could let me know what cha
Jaekwang,
>From this symmetric gradient tensor, do we have one easy way to calculate
> the following "the second invariant of rate-of-strain tensor"?
>
There is SymmetricTensor::second_invariant [1] which is defined as I2 =
1/2[ (trace sigma)^2 - trace (sigma^2) ]
and SymmetricTensor::first_inv
Hamed,
running your code with a Trilinos installation alone fails for me in line
1617 with
An error occurred in line <279> of file
in function
void dealii::TrilinosWrappers::SolverBase::do_solve(const
dealii::TrilinosWrappers::PreconditionBase&)
The violated condition was:
false
Addit
Hamed,
which solvers do you think I can examin and see if they can be used instead
> of SolverCG?
>
Running with the deal.II CG solver gives me
An error occurred in line <424> of file
in function
void
PhaseField::Material_Compressi
Hamed,
At least I can't open your file in a way to clearly see what your
bilinearform is, but could just guess from the implementation. Can you
either write it here (using LaTeX or similar) or upload the file as pdf
again?
Best,
Daniel
--
The deal.II project is located at http://www.dealii.o
Jaekwang,
It seems that no fortran compiler is found. You should be able to configure
Trilinos without fortran using
-DTrilinos_ENABLE_Fortran:BOOL=OFF
After that it's the usual make install [1].
Best,
Daniel
[1] https://www.dealii.org/developer/external-libs/trilinos.html
Am Samstag, 22. Okt
1 - 100 of 591 matches
Mail list logo
