Domain Reduction Method Error " BAD TERMINATION"

[Zouatine]

Dear STKO Team,

I would like to kindly request your help regarding the following issue:

I attempted to replicate the model presented by Professor José Antonio Abell, as described on the following page:

https://opensees.github.io/OpenSeesDocu ... H5DRM.html

In my case, I also used the ground motions from the Santiago case (I was not able to attach the file here since it is approximately 1 GB). The mesh size is 1 m, as recommended on the webpage. However, when I run the analysis, I encounter the issue showed in the picture.

here is my model :

drm-w-bigstructure.zip

(2.71 MiB) Downloaded 243 times

Thank you,

[jaabell]

Hello.

Which version of OpenSees are you using that generates this error? Are you compiling your own version?

I'd suggest temporarily trying out one from before August 2024 to see if the error is still there.

Regards,
JAA

[Zouatine]

Thank you for your response ,

I actually tried all the versions compiled and available in the following reseachgate webpage :

https://www.researchgate.net/publicatio ... r_OpenSees

Which i guess vary from 3.7.0 to 3.7.2 and all of them gave the same error.

However, I got the idea to try it on a single processor, so no model divisions (using the 3.7.0 and 3.7.2 OpenSees versions available on the ResearchGate website). I changed the AnalysesCommand from parallel to a single processor + remesh. I noticed that the error has changed to the following:

New_Error.JPG (145.79 KiB) Viewed 5386 times

So now a different error appears, and the postprocessing file is missing. I could be wrong, but this made me question whether the GM dataset is correct for parallel analysis (since i am the santiago dataset from you website). I would appreciate it if you could share the compiled OpenSees (the one with OpenSeesmp.bat)versions that you used and give me an insight about my issue.

Thank you and waiting for your kind response,

[jaabell]

Hi, I'm going to need to update the dataset and the tutorial it seems. Recent changes might have broken this example. My time frame for this is this semester. I'm currently working with two PhD students on verifying the full ShakerMaker+H5DRM+OpenSees stack to work out all the issues. By the end of the year we should have some new on this regard.

On the other hand, I'm a Linux user, so I can't share an executable with you.

[Zouatine]

Thank you for your response, Professor.

Since I am currently working on a project and need to meet some deadlines, could you please suggest some ways for me to run the example on my own? Even a simplified or approximate version (not necessarily 100% correct) would be very helpful. So far, I have not been able to understand why two different errors appear when I use a single processor versus parallel processing.

Also, is there a way to run the DRM in STKO without inputs from ShakerMaker, for example, using a uniform input or multiple support excitation? maybe a temporary solution ?

Thank you,