New model for energy equation of a porous solid

See merge request !1389

[md][boxlocalassembler] Only bind couplingContext and fvGeometry once

See merge request !1478

Fix/p reconstruction forchheimer

See merge request !1476

Fix calculation of L2 error in angeli test

See merge request !1411

Melanie Lipp authored 6 years ago and Timo Koch committed 6 years ago

In postTimeStep the L2 error is calculated using a difference to the analytical solution.
Without this commit, the L2 error is calculated using a difference between the numerical solution
at a certain time and the analytical solution one time step ahead.

With this commit, the numerical and analytical solution
used to calculate the L2 error should correspond to the same time.

[md][box][localassembler] apply Dirichlet Bcs in residual

Closes #642

See merge request !1472

Multidomain with the box scheme did not work with the absolute residual criterion since
the computed residuals were wrong (Dirichlet constraints were not incorporated).

Update version

See merge request !1471

Cleanup/poroelastic

See merge request !1467

finalize CHANGELOG.md for 3.0

See merge request !1468

[doc] Update build badge to new url

See merge request !1465

[cleanup] Replace property macros

See merge request !1463

[contributors] Add Kilian for 2018

See merge request !1461

Feature/multidomain with soldep params

See merge request !1399

Dennis Gläser authored 6 years ago and Timo Koch committed 6 years ago

The reference solution of this test needed to be updated anyway now that
the derivative computation for solution-dependent parameters is fixed.
We now use a more robust setup of the test with a fixed time step size
as well as a shorter end time to speedup the test.

Dennis Gläser authored 6 years ago and Timo Koch committed 6 years ago

If variables (e.g. flux vars cache) depend on variables of a coupled
domain, we have to reset them before jumping to the next element in
the coupling stencil. Otherwise subsequent residual computations are
wrong.

Dennis Gläser authored 6 years ago and Timo Koch committed 6 years ago

The flux variables cache was in a deflected state when entering
the coupling derivatives computation function. This commit restores
the undeflected state before handling derivatives.