Coupled RC Core Wall with MVLEM-3D

[STKO Team]

That's a warning that pops up when you don't have anything to record in a certain partition (in your case this happens because some of the regions are not present in all partitions).

However, it should no be an issue. When you open the partitioned database in the STKO's post processor, you will see some errors about the MPCO files with empty model that will be skipped, while STKO will keep only the ones with at least one node.
In this example, my region is available only in processor 6, for example:

VALID MPCO.png (292.65 KiB) Viewed 4478 times

[kolozvari]

Thank you for quick response. I re-partitioned the model, which solved the problem.

[STKO Team]

You're welcome

[kolozvari]

Hello, This is a follow up on your message below in this thread from some time ago.

Try this:
extract_data_from_selection_sets.zip


Put whatever you want in this custom function:
Custom1.png



And you will have an extra column after the ID of the Hinge with your custom name:
Custom2.png

I am trying to process coupling beams again, but I am running into the following issue (see below). I am modeling coupling beams with BeamWShearHinge elements and I am meshing it with 1 element assuming that the previous bug was fixed. I am using your API. I am defining a region with all coupling beam selection sets and selecting "include auto generated elements - selection". This used to work, but now I am getting this error. Can you please advise?

Thank you,

[STKO Team]

Make sure this error is not due to the previous error in the terminal:
It looks like one of the partitioned DB you loaded is invalid (or null).
Sometimes this happens when you have previously run an analysis with N processors, and the current analysis is run with M (< N) processors.
OpenSees will overwrite the first M MPCO databases, which will be incompatible with the old N-M ones.

If this is the case, just remove the N-M last MPCO files

[kolozvari]

Thank you for the quick response, but I can assure you this is not the case. I will send you the model via email. It would be great if you can take a look at it.

Thanks.

[STKO Team]

This is not the case, but the issue is similar. It's still due to the presence of incompatible MPCO files:

You are using an MPCORecorder of a subset (coupling beams) that is not available in all partitions (we are working on this point). This creates some MPCO files almost empty. This creates issues with results on gauss/fibers.
Now a simple workaround is to simply delete the empty MPCO Files. You can simply see the empty ones by looking at the file size. You will see that partitions 0,1,2 and 7 have a larger file size (those are the ones you have to keep).

You have another issue.
The region for the coupling beams has a wrong option for include_auto_generated_elements.
You must use option 2 (selection + auto generated).
Otherwise the logic for "embedded element " will not work

[kolozvari]

Thanks for the quick response. I am a bit confused because this used to work without having to manually delete MPCO recorders. This was with the old version of STKO where i had to split the beam i 2 to make it work.

Did something change in the meantime? Also, I tried the case when I record all strain and stress in the entire model, but I still get the same issue. If what are you are saying is true, wouldn't recording strain and stress in the entire model solve the issue of an empty MPCO?

Just to confirm, after every run I have ti manually delete files with small file size and then it will work? How do I know how small is small? Please provide me with some guidance.

Thank you.

[kolozvari]

Hello, I have done what you asked, and the API tool doesnt crash now. However, the output comes blank (see the image below). Please advise. I sent you the model and the API tool via email on Sept 28.

Looking forward to hearing from you. Thanks you.

[STKO Team]

I only changed, as explained in the previous reply, the region for the recorder of the coupling beams:

db_correct_region.png (39.21 KiB) Viewed 3839 times

When you load the MPCO databases STKO will tell you what are the invalid ones:

db_invalid.png (204.62 KiB) Viewed 3839 times

Delete them and reload the MPCO Database. Doing so it works just fine in my case using the recorder of the coupling-beams only.
Instead it will not work with the global recorder because you selected material.strains instead of material.strain (and the zero-Length does not support it)