Dear Henry Arhin,
the output tells you that the LinearSolver did not converge, which, as a
result, causes non-convergence of the nonlinear solver.
This might be an indication that your problem is no longer well-posed.
For example, when changing from 2d to 3d you also have to make sure to
set boundary conditions related to the 3rd dimension.
If you are using permeability tensors, do you also make the according
changes in spatialParams?
So did you take into account the 3rd dimension in your problem and
spatialParams file?
Best,
Martin
On 20.11.24 04:27, Henry Arhin wrote:
Dear Community,
I was trying to apply 3D grid from yasgrid by changing:
// Set the grid type
template<class TypeTag>
struct Grid<TypeTag, TTag::TestPoroElastic> { using type =
Dune::YaspGrid<3>; }; //changed this to 3 insted of 2.
And changing the input parameters for the poroelastic problem.
[Grid]
LowerLeft 0 0 0
UpperRight = 1 1 1
Cells = 10 10 10
[Problem]
Name = poroelastic
EnableGravity = false
[Assembly.NumericDifference]
PriVarMagnitude = 1e5 1e5 1e5
[Component]
SolidDensity = 2700
However, I get this error:
Reading parameters from file params.input.
Computed bounding box tree with 1999 nodes for 1000 grid entities in
0.000275945 seconds.
-- Using the default temperature of 293.15 in the entire domain.
Overload temperatureAtPos() in your spatial params class to define a
custom temperature field.Or provide the preferred domain temperature
via the SpatialParams.Temperature parameter.
Writing output for problem "poroelastic". Took 0.01 seconds.
Colored 1000 elements with 8 colors in 0.000500432 seconds.
Newton solver configured with the following options and parameters:
-- Newton.EnableShiftCriterion = true (relative shift convergence
criterion)
-- Newton.MaxRelativeShift = 1e-08
-- Newton.MinSteps = 2
-- Newton.MaxSteps = 18
-- Newton.TargetSteps = 10
-- Newton.RetryTimeStepReductionFactor = 0.5
-- Newton.MaxTimeStepDivisions = 10
Solve: M deltax^k = rNewton: Caught exception: "NumericalProblem
[solveLinearSystem:/home/kwesi/dumux/dumux/dumux/nonlinear/newtonsolver.hh:512]:
Linear solver did not converge"
terminate called after throwing an instance of 'Dumux::NumericalProblem'
what(): NumericalProblem
[solve:/home/kwesi/dumux/dumux/dumux/nonlinear/newtonsolver.hh:363]:
Newton solver didn't converge after 0 iterations.
Aborted (core dumped)
kwesi@Kwesi:~/dumux/dumux/build-cmake/test/geomechanics/poroelastic$
Please advise me on what to do to view the poroelastic problem in 3D.
Thank you.
_______________________________________________
DuMux mailing list
[email protected]
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux
_______________________________________________
DuMux mailing list
[email protected]
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux
