Hi Ranjeet,
I agree with Dennis, for an incompressible test case you can't use
Neumann boundary conditions
everywhere (at least for solution-independent source/sink terms).
Instead of using a Dirichlet boundary condition you could also
implement some well model as a solution dependent source term. This
allows you
to set Neumann boundary conditions on the whole domain boundary.
Best regards,
Martin
On 09/18/2018 09:04 AM, Dennis Gläser wrote:
Hi Ranjeet,
to me the problem seems to be that you do not specify Dirichlet
boundary conditions anywhere. You must have a Dirichlet boundary
somewhere for the system to be defined properly and that is why it
works when you use the original boundaryTypes() function.
Best wishes,
Dennis
On 18.09.2018 07:58, Ranjeet Singh wrote:
Hi, I am implementing the 2P PorousmediumFlowProblem with cornerpoint
spe9 grid. Boundary conditions are no flow along all boundary and
injection source at one corner (0th element) and production source at
another opposite corner. Boundary and Initial condition are set as
follows:
//problem.hh
BoundaryTypes boundaryTypesAtPos(const GlobalPosition& globalPos)
const
{
BoundaryTypes bcTypes;
bcTypes.setAllNeumann();
return bcTypes;
}
NumEqVector source(const Element &element,
const FVElementGeometry& fvGeometry,
const ElementVolumeVariables& elemVolVars,
const SubControlVolume &scv) const
{
NumEqVector values(0.0);
int eIdx =
this->fvGridGeometry().gridView().indexSet().index(element);
if (eIdx == injectionElement_)
values[FluidSystem::phase1Idx] =
injectionRate_/element.geometry().volume();
if (eIdx == productionElement_)
values[FluidSystem::phase0Idx] =
-productionRate_/element.geometry().volume();
return values;
}
PrimaryVariables initialAtPos(const GlobalPosition &globalPos) const
{
PrimaryVariables values;
// hydrostatic pressure
Scalar densityW = 1000;
values[Indices::pressureIdx] = 5e5 +
densityW*(this->gravity()*globalPos); // inital pw
values[Indices::saturationIdx] = 0.0; // initial sn
return values;
}
I guess I am doing something wrong with boundary condition. In the
cornerpoint test problem, the boundary condition is implemented as,
BoundaryTypes boundaryTypes(const Element &element,
const SubControlVolumeFace &scvf) const
{
BoundaryTypes bcTypes;
// set no-flux on top and bottom, hydrostatic on the rest
// use the scvf normal to decide
const auto& normal = scvf.unitOuterNormal();
using std::abs;
if (abs(normal[dimWorld-1]) > 0.5)
bcTypes.setAllNeumann();
else
bcTypes.setAllDirichlet();
return bcTypes;
}
when I implement the same BC, then solution gets converged. Could you
please explain what I am doing wrong and also, in
abs(normal[dimwold-1])> 0.5
, why it is compared with 0.5, not 0?
Thank you!
Regards,
Ranjeet
_______________________________________________
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
--
M.Sc. Martin Schneider
University of Stuttgart
Institute for Modelling Hydraulic and Environmental Systems
Department of Hydromechanics and Modelling of Hydrosystems
Pfaffenwaldring 61
D-70569 Stuttgart
Tel: (+49) 0711/ 685-69159
Fax: (+49) 0711/ 685-60430
E-Mail: [email protected]
_______________________________________________
Dumux mailing list
[email protected]
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux
