Strain/stress at gauss points of (in-)elastic problems in weak formulation

23

Very appreciate your suggestions.
Ming

24

Hi @bleyerj, one more question about this elasto-plastic model. What should be the right and effective way to get the nodal reaction forces along each boundary (especially on the side with DirichletBC)?

A*u.vector() # result with only the applied external force

I used the way for elastic model as below:

aa = fe.inner(sigma(epsilon(u_t)), epsilon(v))*dxm
AA = fe.assemble(aa)
RF.vector()[:] = AA*u.vector()

Iā€™m not sure if it is correct. Thanks.

UPDATED maybe this is more reasonable:

# f_int = F_ext - res
f_int = fe.inner(epsilon(v), as_3D_tensor(sig))*dxm
RF.vector()[:] = fe.assemble(f_int)

please let me know your comments.

25

Hi,
I would be assembling directly the residual rather than internal forces. You would then have a vector which is zero everywhere except on the DirichletBC where it would give nodal reactions