Dr, Bangerth,
>
>u_h(x) = \sum_i U_i \varphi_i(x)
>
>
First of all, I'm really sorry to bother you continuously...
Could I have please one more question?
I also wanted to calculate l2 norm of |u_h(x)-u_h_exact(x)| (where
u_h_exact(x) is exact solution of u_h(x)).
So what I did is
>
>
> Well, both are the solution -- one is a function of x, the other is a
> vector
> of nodal values, but they ultimately represent the same object. U_i is the
> vector you get out of the linear solver.
>
I'm sorry to make you confused I should have been more specific.
But, my question
Dr. Bangerth,
I'm sorry... I found the similar question about this problem.
That's why I eliminate my question...
Anyway Thank you very much for your reply.
Kyusik.
2016년 11월 1일 화요일 오전 2시 22분 58초 UTC+9, Wolfgang Bangerth 님의 말:
>
> On 10/31/2016 04:00 AM, hank...@gmail.com wrote:
> >
> > I
This is one way to do it. You are just projecting J_tor to a finite
element space and use that for output or further processing. As others
have already noted, the only problem I can see is that computing
solution_max may be inaccurate. This is because in general, if you have
Dr. Bangerth,
Thank you very much for your kind reply.
> This is one way to do it. You are just projecting J_tor to a finite
> element space and use that for output or further processing. As others
> have already noted, the only problem I can see is that computing
> solution_max may be
Dear Bruno,
I have already checked SSOR and AMG non of them make a difference . ILU
doesn't work and gives this error: "[0]PETSC ERROR: MatSolverPackage petsc
does not support matrix type mpiaij"
I don't know how to increase number of iteration. Although I have used the
"SolverControl
Hamed,
On Monday, October 31, 2016 at 3:01:20 PM UTC-4, Hamed Babaei wrote:
>
> This error message can indicate that you have simply not allowed a
> sufficiently large number of iterations for your iterative solver to
> converge. This often happens when you increase the size of your problem. In
Hi all,
I have parallelized a code in which I use GMRES or Bicgstab solvers with
BlockJacobi preconditioner for a symmetric but not possitive-difinite
system matrix. The problem is that it converges for 15 DoFs and fewer
(on 1 node with 16 processor) but does'nt converge for larger number
Kyusik,
I'm trying to calculate J_tor= x*solution+solution_max/y
solution is already calculated vector from 3 function (setup_system,
assemble_system, solve)
To do this calculation I added the following three function(setup_Jtor,
assmeble_Jtor, solve_Jtor)
[...]
This is one way to do it.
Daniel,
> If you know that your FiniteElement is interpolating (which I assume given
> the way you compute the maximum) and that J_tor= x*solution+solution_max/y
> is in your ansatz space (probably not) or an interpolation is sufficient,
> you can use something like
>
>
> First of all, Thank you very much for your reply
>
> This can only work if you know that the maximum is attained at one of the
> support_points
> of your FiniteElement. In general, you can use
> VectorTools::integrate_difference to find a good approximation
> to the maximum value of
Kyusik,
//by Han(To get psi0)
> Vector::iterator max;
> max=std::max_element(solution.begin(),solution.end());
>
This can only work if you know that the maximum is attained at one of the
support_points
of your FiniteElement. In general, you can use
VectorTools::integrate_difference to
Hi All,
I'm trying to calculate J_tor= x*solution+solution_max/y
solution is already calculated vector from 3 function (setup_system,
assemble_system, solve)
To do this calculation I added the following three function(setup_Jtor,
assmeble_Jtor, solve_Jtor)
template
void Step6::setup_Jtor
Dear Hamed,
when we installed dealii and its dependencies on the cluster, I think we
> didn't uncomment the "#PACKAGES="${PACKAGES} once:superlu_dist" in
> candi.cfg so it seems we need to install them again.
>
indeed, but make sure that parmetis is enabled also.
> I want to solve my
14 matches
Mail list logo
