Hello all,
I am going to be developing a new Timoshenko Beam Element that uses cubic-order Hermitian interpolations. The formulation is said to be super-convergent, i.e., it produces an exact analytical solution with a single finite element - much like the conventional Euler-Bernoulli beam element. It is also appropriate for dynamics problems using the same shape functions. The element does not lend itself to shear locking, and thus, requires no special treatment of the primary variables.
There are some particular challenges to this. For starters, how to properly implement the Hermite shape functions on the coupled displacements and rotations. If anyone (*cough* Swetha or Chandu *cough cough*) could chime in on this MOOSE groups thread:
https://groups.google.com/g/moose-users/c/NwQbXa75l9U, it would be a huge help!
--Chris