Add compute tag for elastic strain #2833
Conversation
c7248aa to
9708804
Compare
nikwit
left a comment
There was a problem hiding this comment.
looks good to me. Do I understand correctly that nabla here just refers to a purely spatial gradient and not a covariant derivative?
| return displacement; | ||
| } | ||
|
|
||
| template <size_t Dim> |
There was a problem hiding this comment.
a short comment that this is
|
@nikwit It can actually be a covariant derivative, though we don't currently solve elasticity systems on a curved geometry. But for the XCTS solves we'll have a strain that's the symmetrized covariant derivative of the shift vector: |
|
@nilsleiffischer Yes, I see. But the way it is calculated here with |
|
That's right, to support a curved geometry we would have to add an overload here that takes a metric and christoffel symbols. |
|
ok, thanks :) One more thing: the affine map in the unit test here would be an edge case where the jacobian is constant over the domain (is independent of coordinates) if I understand that correctly. But I assume thats fine as we are not testing the calculation of the partial derivatives itself. |
|
Yep that's how I see it, too |
9708804 to
a298d25
Compare
The strain won't be explicitly solved for anymore with the upcoming second-order DG operator, so this compute tag is used to observe it and dependent quantities such as the potential energy.
a298d25 to
ab372dc
Compare
Proposed changes
The strain won't be explicitly solved for anymore with the upcoming second-order DG operator, so this compute tag is used to observe it and dependent quantities such as the potential energy.
Upgrade instructions
Code review checklist
make docto generate the documentation locally intoBUILD_DIR/docs/html.Then open
index.html.code review guide.
bugfixormajor new featureif appropriate.Further comments