Dennis Gläser authored Dec 14, 2016 and Timo Koch committed Dec 16, 2016

We have to make sure that the provided diffusion coefficients are
greater than zero. Zero coefficients will end up degenerating the
local systems.

For non-zero neumann boundary conditions we did a scaling to recover
DgradU at the boundary from the prescribed flux. We don't do this anymore
and assume that when the UseTpfaBoundary property is set to false, that the
user provides -DgradU at the boundary instead of, e.g., mass fluxes.

The function to make the indices unique in the local sub-control entity seeds
was buggy and led to wrong outside scv faces in the assembly of the local matrices.

Furthermore, calling scvfTouchesBranchingPoint() for grids with dim = dimWorld would have led
to a segmentation fault.

Also, in the interaction volume, when getting the transmissibilities, we always have to compute them
again when coming from the outside face, as the normal vector of the outside face can always be different to
that of the inside face, even when not on a branching point.

Dennis Gläser authored Oct 22, 2016 and Timo Koch committed Nov 23, 2016

If dimWorld > dim, using a GlobalPosition vector is wrong, we have to
use a vector of size=dim here.

Dennis Gläser authored Oct 22, 2016 and Timo Koch committed Nov 23, 2016

Up to now the types for the inner and boundary interaction volume seeds
were always identical. For the mpfa-l and other methods, it would be more efficient
if these types could differ, as e.g. the sizes of the containers are known at compile time
for the l method. This commit introduces the capability of the types to differ.

Dennis Gläser authored Oct 19, 2016 and Timo Koch committed Nov 23, 2016

The stencil method is not necessary in the interaction volumes.
Furthermore, we store the global scvf indices and return a const reference
instead of copying every time.

Dennis Gläser authored Oct 13, 2016 and Timo Koch committed Nov 23, 2016

Treatment of interior boundaries will be specialized in the multidimension
module, where a coupling with a different on the element facets is realized.

Dennis Gläser authored Sep 16, 2016 and Timo Koch committed Nov 23, 2016

The support for interior boundaries will be handled explicitly in the
facet coupling implementation in the multidimension module. Its support
in the standard mpfa implementation causes an overhead which we don't want
here since interior boundaries do not have to be available for standard models.

Dennis Gläser authored Sep 09, 2016 and Timo Koch committed Nov 23, 2016

The mpfa-o interaction volume becomes a separate class from which the
implementation class derives. That is because the mpfa-o method with full
pressure support inherites from the mpfa-o interaction volume but has to use
a different traits class.

Dennis Gläser authored Sep 09, 2016 and Timo Koch committed Nov 23, 2016

the type for the boundary interaction volume is now specified in the
interaction volume traits of the mpfa method, where it is extracted from
in the property defaults. This way, a method can adjust the type used
at the boundary without needing to introduce new method-specific type tags.

Dennis Gläser authored Aug 17, 2016 and Timo Koch committed Nov 23, 2016

It is introduced a base interaction volume which defines the interface.
Additionally, a traits class is introduced. This defines the types of
vectors and matrices etc. to be used. The base class provides public typedefs
to be extracted from outside.

Dennis Gläser authored Aug 15, 2016 and Timo Koch committed Nov 23, 2016

This hopefully saves some memory when global caching is on

Dennis Gläser authored Aug 15, 2016 and Timo Koch committed Nov 23, 2016

Neumann boundaries can be handled by simply adding the specified
neumann flux to the residual (tpfa) or by including it into the local
systems of the interaction volumes. This property further specifies this.