I would like to share with you an observation.
I have set up a problem of linear static stress analysis, solid ring with internal pressure. The stresses and strains look fine, but the strain energy density seems to be computed incorrectly. If you're interested, check out the attached archives. If this is an actual error and has a been reported yet I could do that; if you have already reported this as an error please let me know.
PS: I attach to configurations: one free-floating (displacements wrong, but stresses correct), and one where I apply symmetry boundary conditions on one eighth of the entire geometry.
PPS: the name of the files is a bit of a misnomer. Originally I intended to apply pressure pointing inwards, the models in the archives have the pressure applied on the interior outwards. So, the ring is not compressed but rather under extension.
You did not mention why you thought the strain energy density calculations looked to be wrong, so I took a look at one of your models. The model, stress results, and strain results are axisymmetric, but the strain energy density results are not.
I agree that the strain energy density results appear to be wrong.
Simulation Mechanical user since Dec 1997