GitLab

branch


git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7445 2fb0f335-1f38-0410-981e-8018bf24f1b0

yeah, that's really exotic, but this occationally happens to me. (if i
force all mole fractions to be positive, I get worse convergence
rates.)

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7443 2fb0f335-1f38-0410-981e-8018bf24f1b0

also make the MpNc model respect absolute convergence criteria. IMHO
an unweighted absolute criterion is pretty useless in most contexts,
though. that's because the criterion is about the amount of
conservation quantities that get lost per second but a losing or
gaining a joule of energy is much less of a problem than losing or
gaining a kilogram of mass...

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7442 2fb0f335-1f38-0410-981e-8018bf24f1b0

cleanup of variables that were basically copied for debugging purposes
cleanup of old methods

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7435 2fb0f335-1f38-0410-981e-8018bf24f1b0

   - test_diffusion still is in an intermediate state until the remaining
     2p pressure models are moved to new data structure



git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7434 2fb0f335-1f38-0410-981e-8018bf24f1b0

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7432 2fb0f335-1f38-0410-981e-8018bf24f1b0

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7431 2fb0f335-1f38-0410-981e-8018bf24f1b0

   - clean-up in 2p models
   - added possibility to fix a pressure somewhere in the domain



git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7430 2fb0f335-1f38-0410-981e-8018bf24f1b0

little docu improvements
cleanup of fvpressurecompositional, first steps towards general method for volume derivatives
cleanup of vector multiplications (concerning s=n*(u*n)) in pressure model

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7429 2fb0f335-1f38-0410-981e-8018bf24f1b0

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7426 2fb0f335-1f38-0410-981e-8018bf24f1b0

changes in the 2p2c/2p2cni model


git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7425 2fb0f335-1f38-0410-981e-8018bf24f1b0

removed computeOutflowValues in 2p2cnilocalresidual


git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7421 2fb0f335-1f38-0410-981e-8018bf24f1b0

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7420 2fb0f335-1f38-0410-981e-8018bf24f1b0

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7419 2fb0f335-1f38-0410-981e-8018bf24f1b0

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7417 2fb0f335-1f38-0410-981e-8018bf24f1b0

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7416 2fb0f335-1f38-0410-981e-8018bf24f1b0

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7415 2fb0f335-1f38-0410-981e-8018bf24f1b0

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7414 2fb0f335-1f38-0410-981e-8018bf24f1b0

   - update vector is now resized in the update() function of the
     transport model



git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7413 2fb0f335-1f38-0410-981e-8018bf24f1b0

structure



git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7412 2fb0f335-1f38-0410-981e-8018bf24f1b0

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7411 2fb0f335-1f38-0410-981e-8018bf24f1b0

   - made iterative impet work again!



git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7410 2fb0f335-1f38-0410-981e-8018bf24f1b0

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7409 2fb0f335-1f38-0410-981e-8018bf24f1b0

   - made assembling of fluxes only from one side working
   - some clean-up



git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7408 2fb0f335-1f38-0410-981e-8018bf24f1b0

Further simplified the evaluation of the outflow boundary conditions in the 2p2c and 2p2cni model. Added bool onBoundary as argument to computeFlux(). Then, the fluxVariables object that is generated in a correct way, depending whether we are on a boundary or not and computeOutflowValues() becomes superfluous (it did the same as computeFlux(), just with a BoundaryVariables object and evaluation at the boundary instead of the integration point of the sub-control-volume face).
Removed the <FaceType> template parameter in the fluxvariables, it is not required anymore. The face is also not needed any further as argument, it should to be called via the new face() method, which returns either a boundaryFace or a inner SCV face.

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7403 2fb0f335-1f38-0410-981e-8018bf24f1b0

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7398 2fb0f335-1f38-0410-981e-8018bf24f1b0

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7397 2fb0f335-1f38-0410-981e-8018bf24f1b0

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7396 2fb0f335-1f38-0410-981e-8018bf24f1b0

Now, the only reason for having BoundaryVariables in addition to
FluxVariables is that a different datatype triggers a different value of
the parameter onBoundary in the corresponding constructors. 


git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7395 2fb0f335-1f38-0410-981e-8018bf24f1b0

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7391 2fb0f335-1f38-0410-981e-8018bf24f1b0

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7390 2fb0f335-1f38-0410-981e-8018bf24f1b0

BoundaryFace. Conceptually, a BoundaryFace IS a SubControlVolumeFace.
This will greatly simplify a common treatment of Flux- and
BoundaryVariables. 


git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7389 2fb0f335-1f38-0410-981e-8018bf24f1b0

Renamed overloaded matrixHeatFlux() method (boxspatialparameters.hh) at outflow boundaries: boundaryMatrixHeatFlux() is called instead (from 2p2cnifluxvariables), using the face type as template parameter. In principle this method can be used for both, inner SCV faces and boundary faces; on the long run it would be nice to employ just one single method which handles the heat flux for both face types.

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7378 2fb0f335-1f38-0410-981e-8018bf24f1b0

well, for everything except the decoupled stuff. Thanks goes to
Christoph for pointing this out.

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7376 2fb0f335-1f38-0410-981e-8018bf24f1b0

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7375 2fb0f335-1f38-0410-981e-8018bf24f1b0

both matrix and RHS vector have to be protected to being acessible by the multiphysics assemble() method

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7374 2fb0f335-1f38-0410-981e-8018bf24f1b0

consequently methods from standard model

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7372 2fb0f335-1f38-0410-981e-8018bf24f1b0

since this was done in a semi-automatic way compilation for stuff not
used by the non-decoupled tests may be non-functional. I spared the
decoupled stuff because it does not compile ATM so I did not have a
way verify that it still works...

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7371 2fb0f335-1f38-0410-981e-8018bf24f1b0

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7370 2fb0f335-1f38-0410-981e-8018bf24f1b0

boxspatialparameters: additional empty matrixHeatFlux() method, which uses the face as argument and can handle both, boundary faces and sub-control-volume faces; this is used for the calculation of outflow fluxes only so far and has to be specified in the spatial parameters, if required. 
2p2cnifluxvariables: additional call to matrixHeatFlux(), if we are on a boundary face; changed temperature gradient to a private object with a return function to make it accessible from outside.
boxlocalresidual: removed some superfluous asImp()s.
2p2cnilocalresidual: added computeOutflowValues() method for the evaluation of outflow conditions.
regularizedvangenuchten: corrected typo.

git-svn-id: svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/trunk@7368 2fb0f335-1f38-0410-981e-8018bf24f1b0