Apply Dirichlet bcs on the rotations (first derivatives) when using an Hermite FE

[stepi9]

Hi,
I'm trying to solve the Euler Bernoulli beam bending problem using firedrake and Hermite elements. The variational formulation is:
a=div(grad(u))EJdiv(grad(v))dx
L=fv*dx

However, I can't understand how to apply essential BCs on the rotations. If I use DirichletBC it only constraints the displacement u, not the associated rotation du/dx, which is the other DOF of the Hermite element.

Can anyone help me?

[wence-]

Is this in 1D? I think it must be because we don't support essential boundary conditions on Hermite elements in 2D.

In 1D, we support setting the vertex value $u$, but not its derivative $u_x$, on the boundary (this is implemented with a hack in DirichletBC`)

firedrake/firedrake/bcs.py

Lines 133 to 138 in c5cd37a

	def hermite_stride(bcnodes):
	if isinstance(self._function_space.finat_element, finat.Hermite) and \
	self._function_space.mesh().topological_dimension() == 1:
	return bcnodes[::2] # every second dof is the vertex value
	else:
	return bcnodes

So I think right now your best bet is to try and apply your BCs using Nitsche's method instead.