Time dependent Dirichlet and Neumann BCs#1380
Conversation
|
Jenkins: OGS-6/Linux-PRs failed: https://svn.ufz.de:8443/job/OGS-6/job/Linux-PRs/2748/ |
chleh
force-pushed
the
time-dep-dirichlet
branch
from
f554024 to
cd3b527
Compare
|
Jenkins: OGS-6/Linux-PRs failed: https://svn.ufz.de:8443/job/OGS-6/job/Linux-PRs/2764/ |
chleh
force-pushed
the
time-dep-dirichlet
branch
from
cd3b527 to
288578f
Compare
| initializeDirichletBCs(_boundary_conditions, _dirichlet_bcs); | ||
| // for each BC there will be storage for Dirichlet BC, which by default is | ||
| // empty and has to be filled by the respective BC object if needed. | ||
| _dirichlet_bcs.resize(_boundary_conditions.size()); |
There was a problem hiding this comment.
Maybe I misunderstand the comment, but resize sets the _dirichlet_bcs size, so !_dirichlet_bcs.empty()
There was a problem hiding this comment.
Each BC can be Dirichlet (or "essential") or natural, or in the future maybe even both, so each of the BCs need some storage for it. As you wrote, this implies that _dirichlet_bcs is not empty. However, the memory overhead for that should be rather small, because only some empty std::vectors are created.
endJunction
force-pushed
the
time-dep-dirichlet
branch
from
d24ebd7 to
6e696b7
Compare
|
I've rebase it and included the mechanics and the thermal conductivity processes. |
endJunction
force-pushed
the
time-dep-dirichlet
branch
from
61df58e to
1320173
Compare
|
👍 |
| // convert mesh node ids to global index for the given component | ||
| bc_values.ids.reserve(bc_values.ids.size() + _mesh_node_ids.size()); | ||
| bc_values.values.reserve(bc_values.values.size() + _mesh_node_ids.size()); | ||
| for (auto id : _mesh_node_ids) |
endJunction
force-pushed
the
time-dep-dirichlet
branch
from
1320173 to
1fec357
Compare
endJunction
left a comment
endJunction
left a comment
There was a problem hiding this comment.
Testing new PR review feature of github.
|
Jenkins: OGS-6/Gui/Gui-Linux-PRs failed: https://svn.ufz.de:8443/job/OGS-6/job/Gui/job/Gui-Linux-PRs/2824/ |
endJunction
left a comment
endJunction
left a comment
There was a problem hiding this comment.
That's interesting feature.
endJunction
force-pushed
the
time-dep-dirichlet
branch
from
1fec357 to
ab7dd7b
Compare
endJunction
force-pushed
the
time-dep-dirichlet
branch
from
ab7dd7b to
3382ffe
Compare
The original preTimestep is calling boundary conditions' preTimestep now.
|
OpenGeoSys development has been moved to GitLab. |
6 participants
Time dependence is implemented using curves, i.e.
f(t, x) := c(t) * p(x), wherecis a curve andpis a parameter.The Parameter interface has been extended s.t. parameters now can tell if they are time dependent.