Thanks, that's a nice feature. I tried this for my problem which has an
equation_systems with two systems attached to it. For my Navier-Stokes
system the coarse solution has been projected correctly on the fine mesh.
However for my turbulence model system1 the coarse solution seems to be
scrambled on the fine mesh. The solution of a coarse mesh node in the fine
mesh seems to be connected to a different node now. I'm not sure why it is
correct for one and not for the other. It could be with how I write the
solution to a text file, though I never had problems with this before. Here
is the code for that:
std::vector<Number> soln ;
std::vector<Number> soln1;
system.solution->localize_to_one(soln);
system1.solution->localize_to_one(soln1);
if (mesh.processor_id() == 0)
{
std::vector<Number> u_permute(mesh.n_nodes());
std::vector<Number> v_permute(mesh.n_nodes());
std::vector<Number> p_permute(mesh.n_nodes());
std::vector<Number> x_coord(mesh.n_nodes());
std::vector<Number> y_coord(mesh.n_nodes());
std::vector<Number> nuE1(mesh.n_nodes());
MeshBase::const_node_iterator nd = mesh.nodes_begin();
MeshBase::const_node_iterator nd_end = mesh.nodes_end();
for (; nd != nd_end; ++nd)
{
const Node* node = *nd;
u_permute[node->id()] = soln[node->dof_number(0,0,0)];
v_permute[node->id()] = soln[node->dof_number(0,1,0)];
p_permute[node->id()] = soln[node->dof_number(0,2,0)];
x_coord[node->id()] = (*node)(0);
y_coord[node->id()] = (*node)(1);
nuE1[node->id()] = soln1[node->id()];
}
std::ostringstream filen;
filen << "NavierStokesSolutionMatlabCoarse_"
<< std::setw(3)
<< std::setfill('0')
<< std::right
<< t_step + 1
<< ".txt";
std::ofstream datfile;
datfile.precision(12);
datfile.open(filen.str());
for (uint ii=0; ii<u_permute.size(); ii++)
{
datfile << x_coord[ii] << " " << y_coord[ii] << " " << u_permute[ii]
<< " " << v_permute[ii] << " " << p_permute[ii] << " " << nuE1[ii] <<
std::endl;
}
datfile.close();
}
Otherwise it may have to do with node renumbering. I also tried to just use
the turbulence model system and comment out the N-S system, but this
yielded the same result. Do you have an idea what could have happened?
Kind regards,
Pepijn
On Sat, May 28, 2016 at 11:00 PM, Derek Gaston <fried...@gmail.com> wrote:
> Don't treat them as separate meshes. Get the values you want on the
> coarse mesh and then uniform_refine() it in libMesh as many times as you
> want... libMesh will automatically project the solution on the the finer
> meshes each time you call EquationSystems::reinit(). Then start solving on
> that fine mesh...
>
> Derek
>
> On Sat, May 28, 2016 at 4:32 PM Pepijn Kessels <pepijnkess...@gmail.com>
> wrote:
>
>> Hi there,
>>
>> I have a coarse structured mesh, and a couple more that are integer
>> refinements of it. I have the solution for the coarse mesh and could not
>> find if libMesh has the possibility to interpolate this solution onto the
>> finer meshes. That would be nice since I can then use that as an initial
>> solution for solving the finer meshes, saving me a lot of time. Any help
>> would be appreciated.
>>
>> Kind regards,
>>
>> Pepijn
>>
>> ------------------------------------------------------------------------------
>> What NetFlow Analyzer can do for you? Monitors network bandwidth and
>> traffic
>> patterns at an interface-level. Reveals which users, apps, and protocols
>> are
>> consuming the most bandwidth. Provides multi-vendor support for NetFlow,
>> J-Flow, sFlow and other flows. Make informed decisions using capacity
>> planning reports.
>> https://ad.doubleclick.net/ddm/clk/305295220;132659582;e
>> _______________________________________________
>> Libmesh-users mailing list
>> Libmesh-users@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/libmesh-users
>>
>
------------------------------------------------------------------------------
What NetFlow Analyzer can do for you? Monitors network bandwidth and traffic
patterns at an interface-level. Reveals which users, apps, and protocols are
consuming the most bandwidth. Provides multi-vendor support for NetFlow,
J-Flow, sFlow and other flows. Make informed decisions using capacity
planning reports. https://ad.doubleclick.net/ddm/clk/305295220;132659582;e
_______________________________________________
Libmesh-users mailing list
Libmesh-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/libmesh-users
