Periodic Dirichlet boundary conditions #418
Conversation
Codecov Report
@@ Coverage Diff @@
## master #418 +/- ##
==========================================
- Coverage 91.32% 87.53% -3.79%
==========================================
Files 22 22
Lines 3043 3177 +134
==========================================
+ Hits 2779 2781 +2
- Misses 264 396 +132
Continue to review full report at Codecov.
|
There was a problem hiding this comment.
I added some short comment. I did not really check the _add! - function :P
It would also be nice if you can inhomogeneities to the PeriodicConstraint via a function (similarly to how Dirichlet does it). I think you are enforcing [|um|] = 0, but I am enforcing [|u|] = [|uM|]
fredrikekre
changed the title
Done. |
fredrikekre
changed the title
| # ``\boldsymbol{u}^\mu`` to be periodic, and ``\boldsymbol{t}`` anti-periodic across the | ||
| # RVE. Similarly as for Dirichlet boundary conditions, Eq. ``(1\mathrm{b})`` is directly |
There was a problem hiding this comment.
"and \boldsymbol{t} anti-periodic across the RVE"
Is this really true that we enforce, doesn't this just happen to be the case if the structure is periodic?
There was a problem hiding this comment.
Good question. I believe it is actually just an assumption, but never enforced, which is used to get rid of the second equation.
There was a problem hiding this comment.
I suppose it must hold true for the integral equation due to equilibrium, but I'm not sure about the point-wise anti-periodicity...
There was a problem hiding this comment.
Here are my notes from this -- looks like it is assumed to be pointwise anti-periodic in the derivation, although it is never enforced.
There was a problem hiding this comment.
It looks like we assume anti-periodicity, and by inserting that assumption in equation 9 it is effectively enforced. Without this assumption, the problem doesn't have sufficient bc because the traction would be unknown for both boundaries. At least that is how it all makes more sense to me now...
fredrikekre
force-pushed
the
fe/periodic
branch
from
4773aa8
to
2b531a5
Compare
No description provided.