Analysis did not converge

[CesarDavid]

Any help with this model convergence.

[STKO Team]

Merge your model. All faces are separated

[CesarDavid]

Thank you very much. I did, but still no convergence.

[kesavapraba]

Hi, Please note the following changes that I made to your model, and it now runs smoothly.

1] It seems that pushover load definition has some issue. Time series (ciclica) for the pushover displacement already included a comparable displacement (~28 mm), however the spPrescribed load in the condition again multiplies it with -31. You are multiplying the actual displacement by 31! So, I changed it to -1. You can see the maximum displacement as around 30 from the experimental background curve.

2] Refine the mesh to improve the accuracy

3] Analysis settings (pushover): Increase the numIncr to atleast 5000 since the cyclic loading has many loops, hence increasing numIncr can help achieving good match between the experimental and numerical prediction.

[CesarDavid]

I really apreciate your help. Thanks

[kesavapraba]

You are welcome

[Gabriele.Neri]

Dear stko team,
I'm trying to do a parametric analysis, but I have some issues with the convergence.
I've tried to use different system solvers (Mumps, UmfPack), I've tried to increase the numIncr, without getting any results.. Can you help me?
I attached the error, the warning and the file I'm testing on.

Concerning the parametric analysis, the steps are the following:
- I created TCL files with the Write Input File command
- I opened a command shell in the directory in which I generated the TCL files
_I launched OpenSees with the number of processors = number of parameters, that are 30 in this case
(C:\OpenSees-Solvers\openseesmp.bat main.tcl 30)

Thanks in advance

[kesavapraba]

Hi, the issue is with forceBeamColumn element, it is not able to find force/displacement compatibility. I checked the BeamSectionProperty, you have used hinge option with nonlinear fiber model for the middle of the beam or column elements. I suggest this is not necessary unless you go for finite plastic hinge length model at the ends of the beam-column elements. Hence, I modified it to StandardIntergration type with 5 integration points, as you see in the attached figure. Now, the analysis converges.

[Gabriele.Neri]

Hi, the issue is with forceBeamColumn element, it is not able to find force/displacement compatibility. I checked the BeamSectionProperty, you have used hinge option with nonlinear fiber model for the middle of the beam or column elements. I suggest this is not necessary unless you go for finite plastic hinge length model at the ends of the beam-column elements. Hence, I modified it to StandardIntergration type with 5 integration points, as you see in the attached figure. Now, the analysis converges.

Hi,
Meanwhile thanks for teh support!
You are right, I set hinge option both for beam and column elements because I need to consider non linearity for the entire length of beam-column elements in the model.

Thanks in advance

[kesavapraba]

Ok, Yes, you can do it without hinge option as I suggested in my previous post. Thank you.