Plastic Rotation vs. Plastic Curvature

[Deniz]

Hi Silvia,
It is a well known property of opensees that you can easily monitor curvature at a section designated at integration point of a forcebeamcolumn element, in addition, plastic deformation at the ends in terms of rotation. However, it is not possible to relate "plasticDeformation (output is rotation)" with "section curvature (i.e. at the ends)" of a member undergoing plastic deformation, for natural axis of the section is not constant anymore.

My question is that is there simple way to record plastic deformation in terms of curvature or vice versa. I would also welcome any suggestions to solve this simple problem without going deep into stress strain relation of the section. You would appreciate that it becomes very difficult while running many geometric sections at a time.

Thank you.

Deniz Utkutug
Middle East Technical University
Ankara, Turkiye

[silvia]

this is an interesting modeling problem.
i would record curvatures and then decide over what hinge length to integrate it.
if your model is simple, you could think about using the displacement-based element and use many of them....

[Deniz]

Silvia,

Thank you for your fast response.

From your reply I can see that there is no implemented record command argument which monitors only plastic state curvature (or full curvature output from "deformation" flag).

Integration may be a good approach, however, it seems that it is not feasible utilizing it in a complex model with "nonlinearBeamColumn" elements. In addition, controlling output may become extremely difficult when you have to deal with a massive post process job.

I will be very happy if you post any update information related with this very subject in the future.

Best regards.

Note for FMK.
This may be a new add-on for "record" command to make things easy, i.e. plasticCurvature argument.

Deniz Utkutug
Middle East Technical University
Ankara, Turkiye

[silvia]

there is a plasticRotation recorder, but i'm not too convinced by it.
the deformation recorder does record curvature.
remember, the element does a numerical-integration scheme.
the problem is that there is not much data out there on the actual distribution of curvature, so you have to be careful as to what you do and assume.