Map initial deformation to mesh
I need to compute the stress in a beam that has plastic deformation. I dont know what are the initial stresses so I need to compute them.
The way I see it is I would mesh the beam (with solid/shell elements) first. Then I would add the deformed geometry to the model. I would then deform the initial mesh so it looks like the deform geometry.
Other than adding spcd to faces/nodes, is there a bette way to do this? Can this be an optimization problem?
I can't share pictures but you could imagine a beam fixed at both ends and warped in the middle.
Thanks
Find more posts tagged with
Altair EmployeeAlso, I was thinking on deforming the beam in an explicit subcase and then take the deformed mesh and stresses as a start for the second analysis (nl static) in which I would load the top of the beam. Would it be a better way to do it?
Bill,
Are you intending on including the initial stress/strain on the model, or just the deformed shape.
You can export the deformed shape directly out of Hyperview as a solver deck.
Or you can map displacement results to a model. (attached video)
However, if you want to include the initial stress/strain, you should do a NL continuation step or transient setup.
https://help.altair.com/hwsolvers/os/topics/solvers/os/cntnlsub_sub_r.htm?zoom_highlight=CNTNLSUB
Bill,
Are you intending on including the initial stress/strain on the model, or just the deformed shape.
You can export the deformed shape directly out of Hyperview as a solver deck.
Or you can map displacement results to a model. (attached video)
However, if you want to include the initial stress/strain, you should do a NL continuation step or transient setup.
https://help.altair.com/hwsolvers/os/topics/solvers/os/cntnlsub_sub_r.htm?zoom_highlight=CNTNLSUB
I need to include the deformed shape and the stress/strain.
I think I will need to include the result from a previous step.
Also, I was thinking on deforming the beam in an explicit subcase and then take the deformed mesh and stresses as a start for the second analysis (nl static) in which I would load the top of the beam. Would it be a better way to do it?