Re: [deal.II] Solving time dependent heat equation with MPI (PETsc)

The code was submitted to the github (https://github.com/dealii/code-gallery/pull/92 Oh, doh, my apologies for forgetting about it! I hope to get around to reviewing it soon! Best W. -- Wolfgang Bangerth

Re: [deal.II] Solving time dependent heat equation with MPI (PETsc)

that's good. feel free to discuss any questions related on this topic, I will get back to you as soon as possible. syed ansari 于2022年1月25日周二 14:41写道： > Thank you very much, Mark, for sharing. This is indeed very helpful for me. > > Best Regards, > Syed Ansari S. > > On Tue, Jan 25, 2022 at 5:32

Re: [deal.II] Solving time dependent heat equation with MPI (PETsc)

Thank you very much, Mark, for sharing. This is indeed very helpful for me. Best Regards, Syed Ansari S. On Tue, Jan 25, 2022 at 5:32 PM Mark Ma wrote: > Hi, > > The code was submitted to the github ( > https://github.com/dealii/code-gallery/pull/92) but is waiting for > checking.

Re: [deal.II] Solving time dependent heat equation with MPI (PETsc)

Hi, The code was submitted to the github ( https://github.com/dealii/code-gallery/pull/92) but is waiting for checking. Alternatively, you can find the code at my github at: https://github.com/MarkMa1990/code-gallery/tree/master/Distributed_Moving_Laser_Heating Have a nice day, Mark syed ansari

Re: [deal.II] Solving time dependent heat equation with MPI (PETsc)

Thank you very much Prof. Bangerth for your kind reply. On Sun, Jan 23, 2022 at 10:33 PM Wolfgang Bangerth wrote: > On 1/22/22 2:56 AM, syed ansari wrote: > > I tried to find this code in the deal.ii code gallery but I > am > > unable to locate it. I request the group members to

Re: [deal.II] Solving time dependent heat equation with MPI (PETsc)

On 1/22/22 2:56 AM, syed ansari wrote:     I tried to find this code in the deal.ii code gallery but I am unable to locate it. I request the group members to help me to find this code. I don't think it was ever submitted. Best W. --

Re: [deal.II] Solving time dependent heat equation with MPI (PETsc)

Hello all, I tried to find this code in the deal.ii code gallery but I am unable to locate it. I request the group members to help me to find this code. Thanks in advance. Best Regards, Syed Ansari S. On Tuesday, October 17, 2017 at 7:46:16 PM UTC+5:30 Wolfgang Bangerth wrote: >

Re: [deal.II] Solving time dependent heat equation with MPI (PETsc)

On 10/17/2017 07:53 AM, Mark Ma wrote: Sure, no problem. Excellent -- let us know if you need help! We're eager to grow the code gallery! Best Wolfgang -- Wolfgang Bangerth email:

Re: [deal.II] Solving time dependent heat equation with MPI (PETsc)

Sure, no problem. 2017-10-17 15:23 GMT+02:00 Wolfgang Bangerth : > On 10/17/2017 07:21 AM, Mark Ma wrote: > >> >> PS: I think this could be another tutorial show how to implement this, I >> would like to contribute if you think it is interesting for Deal.II project. >> >

Re: [deal.II] Solving time dependent heat equation with MPI (PETsc)

On 10/17/2017 07:21 AM, Mark Ma wrote: PS: I think this could be another tutorial show how to implement this, I would like to contribute if you think it is interesting for Deal.II project. Yes, I think that would be great. Or, maybe simpler and with less documentation requirements: make it

Re: [deal.II] Solving time dependent heat equation with MPI (PETsc)

Hi Wlofgang, Good news, finally I resolved this problem by assembling the left_Mass_matrix and right_rhs by the code following, local_rho_c_T_matrix(i,j) += (rho_C_fun_T.value(fe_values. quadrature_point(q)) * phi_i_u *

Re: [deal.II] Solving time dependent heat equation with MPI (PETsc)

On 10/16/2017 01:53 PM, Mark Ma wrote: I think this problem lies in the time updating of solution using old_solution, since the mass_matrix and laplace_matrix have already eliminated the constraint node, /*mass_matrix_T.vmult (system_rhs, old_solution_T_cal*//*);*/ is no longer valid for

Re: [deal.II] Solving time dependent heat equation with MPI (PETsc)

source code are attached. 在 2017年10月16日星期一 UTC+2下午9:14:13，Wolfgang Bangerth写道： > > On 10/16/2017 09:23 AM, Mark Ma wrote: > > > > It almost looks to me like you're applying both Dirichlet values (via > > the ConstraintMatrix) and Neumann boundary values (via a boundary > > integral). But then

Re: [deal.II] Solving time dependent heat equation with MPI (PETsc)

> On 10/16/2017 09:23 AM, Mark Ma wrote: > > > > It almost looks to me like you're applying both Dirichlet values (via > > the ConstraintMatrix) and Neumann boundary values (via a boundary > > integral). But then I haven't taken a look at the code, so I can't > > really say for sure. > >

Re: [deal.II] Solving time dependent heat equation with MPI (PETsc)

On 10/16/2017 08:55 AM, Mark Ma wrote: > So when you visualize the solution, the error is at the boundary but it looks > correct in the interior? > > Yes, it is. Thereafter, the error at boundaries does propagate and then > interfere with the interior.

Re: [deal.II] Solving time dependent heat equation with MPI (PETsc)

在 2017年10月16日星期一 UTC+2下午4:31:25，Wolfgang Bangerth写道： > > > > So when you visualize the solution, the error is at the boundary but > it looks > > correct in the interior? > > > > Yes, it is. Thereafter, the error at boundaries does propagate and then > > interfere with the interior.

Re: [deal.II] Solving time dependent heat equation with MPI (PETsc)

So when you visualize the solution, the error is at the boundary but it looks correct in the interior? Yes, it is. Thereafter, the error at boundaries does propagate and then interfere with the interior. From the pictures you posted in a follow-up, it looks like the boundary

Re: [deal.II] Solving time dependent heat equation with MPI (PETsc)

> On 10/15/2017 12:41 PM, Mark Ma wrote: > > > > Now the projection of initial values (rewrite the code by manually > assemble > > the matrix and system_rhs and calculate) run OK, but the time updating > of T is > > not correct, same phenomenon appears. I believe this may arise from the >

Re: [deal.II] Solving time dependent heat equation with MPI (PETsc)

On 10/15/2017 12:41 PM, Mark Ma wrote: Now the projection of initial values (rewrite the code by manually assemble the matrix and system_rhs and calculate) run OK, but the time updating of T is not correct, same phenomenon appears. I believe this may arise from the fact that direct using

Re: [deal.II] Solving time dependent heat equation with MPI (PETsc)

Prof. W. Bangerth, Please find attached source code, where the projection part is ok now, but time updating of solution part results is incorrect I believe (see boundary results). This confused me a few weeks and still did not find a solution. Best, Mark 在 2017年10月13日星期五

Re: [deal.II] Solving time dependent heat equation with MPI (PETsc)

Prof. Wolfgang Bangerth, Now the projection of initial values (rewrite the code by manually assemble the matrix and system_rhs and calculate) run OK, but the time updating of T is not correct, same phenomenon appears. I believe this may arise from the fact that direct using matrix vmult (i.e.

Re: [deal.II] Solving time dependent heat equation with MPI (PETsc)

Dear Wolfgang, Thank you for your explanation. Currently, I am using a code that was not written by me, and uses the MatrixTools::apply_boundary_values() approach. I try to change it to use the ConstraintMatrix one. For that, step-40 seems to be the best starting point, like Mark did. Thanks

Re: [deal.II] Solving time dependent heat equation with MPI (PETsc)

Dear Bangerth, When you mention In general, using MatrixTools::apply_boundary_values() is not the way to go > with MPI programs. Rather, use a ConstraintMatrix and incorporate the > boundary > values into the same object as you do with hanging node constraints. This is the way to go due to

Re: [deal.II] Solving time dependent heat equation with MPI (PETsc)

On 10/13/2017 02:06 AM, Mark Ma wrote: later, I changed the control into | | SolverControlsolver_control (5*system_rhs.size(),1e-12*system_rhs.l2_norm()) | | this works well for structure size of um or nm. I think previous setting may lead to a loss of precision so that the results are always

Re: [deal.II] Solving time dependent heat equation with MPI (PETsc)

It is actually my stupid mistake in solvercontrol, I used SolverControl solver_control (dof_handler.n_dofs(),1e-12) This works when the geometry size in the order of 1, but fails at 1e-6, or even 1e-9. Here I actually want to make a micrometer or nanometer size structure. later, I changed the

Re: [deal.II] Solving time dependent heat equation with MPI (PETsc)

On 10/12/2017 03:02 AM, Mark Ma wrote: Thanks for your useful advice. Finally I solved this problem. So what was the problem? Maybe we can learn from your example? Best W. -- Wolfgang Bangerth email:

Re: [deal.II] Solving time dependent heat equation with MPI (PETsc)

Hi Prof. Wolfgang Bangerth, Thanks for your useful advice. Finally I solved this problem. Best, Mark 2017-10-10 19:50 GMT+02:00 Wolfgang Bangerth : > On 10/10/2017 11:40 AM, Mark wrote: > >> >> I really appreciate for your quick reply. Now I do as what you said, >>

Re: [deal.II] Solving time dependent heat equation with MPI (PETsc)

On 10/10/2017 11:40 AM, Mark wrote: I really appreciate for your quick reply. Now I do as what you said, using VectorTools in setup_system(). It seems working, but there is still one question that I always find very large values at points near boundaries, see attached figures. Is there a way

Re: [deal.II] Solving time dependent heat equation with MPI (PETsc)

Hi Prof. Wolfgang Bangerth, I really appreciate for your quick reply. Now I do as what you said, using VectorTools in setup_system(). It seems working, but there is still one question that I always find very large values at points near boundaries, see attached figures. Is there a way to solve

Re: [deal.II] Solving time dependent heat equation with MPI (PETsc)

On 10/10/2017 08:40 AM, Mark Ma wrote: I want to solve a heat equation in the time domain with distributed memory using MPI, but the results are incorrect. In order to do so, I reference tutorial step-23 for time updating method and step-40 for implementing MPI. May I ask whether my boundary

[deal.II] Solving time dependent heat equation with MPI (PETsc)

Dear experienced deal.II users and developers, I want to solve a heat equation in the time domain with distributed memory using MPI, but the results are incorrect. In order to do so, I reference tutorial step-23 for time updating method and step-40 for implementing MPI. May I ask whether my