asdeasoft.net

Dear STKO Team,

I am facing convergence problems with the analysis of a rectangular concrete wall. The modeling was made based on the RC Shear Wall video posted on your YouTube channel with some adjustments for this specific model but I think the settings for the analysis are not quite good.

Attached to this message, please find the model.
Please help.

Thank you.

Happy New Year.

Best Regards.

Dear user,

We had a look at your file and here is a working copy of your original file: Here are some changes I made:

• Your model was singular because the top node (control node, where you applied the displacement history) was not attached to the main wall. I added an interaction between the top node (master) and the wall top-edge (slaves), using a rigidLink boundary condition
• I increased the number of time steps. You are using the IMPLEX option in the DamageTC3D model. The IMPLEX algorithm makes the model always converging, but it requires small time steps.
• I changed your cyclic timeSeries. It was incorrectly starting from t = 1. However in your previous vertical analysis you reset the time to 0, so the cyclic timeSeries should start from 0.
• In order to impose a constant dU (displacement increment) for every dT, I changed the abscissas of your cyclic timeSeries, otherwise you waste too much computational time in the first cycles that are pretty small.
• Then you maximum imposed displacement is small. The imposed displacement is 1, and your timeSeries had a maximum of 0.01. So you where imposing a maximum of 0.01 mm. I increased it so that the maximum displacement is 10 mm.
• I also made the total duration of the timeSeries = 1 second (just for clarity, it is a static analysis to the real-time does not matter)
• I also added the monitor tool, that is useful for monitoring your analysis in real-time

Here is the output of your model:

Thank you very much. Yet, the monitor does not run. I get the attached message and the analysis does not go further. If I delete the monitor it is ok.

We fixed this problem in version 1.1.5. What version are you using?
Try to download the latest one.

1.1.5

Yeah you are right,
We fixed it in version 2.0.0 that will be released soon (not in version 1.1.5)

In the meanwhile, here's the fixed "monitor.py" file: You can un-zip it and copy it in
"C:\Program Files\STKO\external_solvers\opensees\analysis_steps\Misc_commands"
replacing the existing one

Let us know if it works.

Everything works fine now.

Thank you very much.