About a convergence error

Post Reply
BEngineer
Posts: 10
Joined: Tue May 11, 2021 10:33 am

About a convergence error

Post by BEngineer » Wed Jun 23, 2021 1:10 pm

Hello everyone,

I am currently having an error with my model.
I'm a new user. So, I had many errors until now, but this is the one I couldn't solve, yet.
Could anyone help me, please?

The attached file is my model...

Thank you!
Attachments
wall_c_fix_XYplane.rar
(107.54 KiB) Downloaded 13 times

STKO Team
Posts: 737
Joined: Tue Oct 29, 2019 8:45 am

Re: About a convergence error

Post by STKO Team » Tue Jun 29, 2021 1:53 pm

Here's a working version of your model:
wall_c_fix_XYplane.zip
(157.45 KiB) Downloaded 14 times
There were some issues with your model:
  • The Concrete07 material has a nice hysteretic loop, but it gets really unstable in certain circumstances, leading to unphysical results:
    Immagine 2021-06-29 151524.png
    Immagine 2021-06-29 151524.png (32.97 KiB) Viewed 178 times
  • In OpenSees with 2D models, the geomTransf does not use the user-defined Z vector (it assumes that the section local Z axis coincides with the global Z). So your custom orientation with Local Axes in STKO is useless. It will collapse all the fiber in 1 point, thus leading to a singular matrix.
    You can either re-create all your fiber sections in the opposite direction, or (as I did) turn your model into a 3D model, and add a Fix2D constraint (Uz = Rx = Ry = 0) all over the model.
  • After those changes your model runs but will show convergence issues due to a sudden drop in your Pushover curve. I solved it by changing some parameters in the analysis, most importantly the Krylov-Newton and the Adaptive time step.

BEngineer
Posts: 10
Joined: Tue May 11, 2021 10:33 am

Re: About a convergence error

Post by BEngineer » Tue Jun 29, 2021 4:06 pm

Dear STKO Team,

Thank you very much!
I really appreciate your help!

Post Reply