Coupled RC Core Wall with MVLEM-3D

[STKO Team]

The main difference for the out-of-plane behavior is that the ASDShellQ4 uses a general purpose thick shell formulation that works also in the thin shell limit (due to the MITC4 formulation that removes the well known transverse shear locking).

On the contrary, the DKGT/Q can be used only for thin plates. They do not consider transverse shear deformation at all.
(it's the same as for beams... ASDShellQ4=Timoshenko, DKGT/Q = Bernoulli)

That being said, that huge difference make sense only if the slab is quite thick. In this case the rigid diphragm assumption is not realistic because it assumes an infinite stiffness only for membrane actions.

What is the thickness of your slabs? Could you share your model?

[kolozvari]

Thank you for your response. The slabs are 0.2m thick (not very thick). Attached is the model I used for pushover, which has rigid diaphragms so I can easily apply pushover loads. For dynamic analysis I use essentially the same model without the rigid diaphragm.

Looking forward to hearing from you.

Thank you.

[STKO Team]

Try to discretize the mesh. You are modeling a slab with only 1 element between the beams. You cannot capture bending correctly.

[kolozvari]

Thank you for your suggestion. I discretized the model into more elements, but I am having a doubt about the results processing. I am using Monitor option to record results of the pushover analyses. However, when I discretize the mesh, it appears that base model nodes are not in the selection set that I use to record the base shear (see green markup in the screenshot below). When I try to select these nodes in the Mesh tab, I cannot. I am worried that the Monitor option is not recorded the entire base shear, but only base shear from some nodes. Please advise how to fix this issue.

Thanks

[kolozvari]

Hello, I just wanted to follow up on my previous question. Hope you can respond to me soon.

Thank you.

[STKO Team]

Thank you for your suggestion. I discretized the model into more elements, but I am having a doubt about the results processing. I am using Monitor option to record results of the pushover analyses. However, when I discretize the mesh, it appears that base model nodes are not in the selection set that I use to record the base shear (see green markup in the screenshot below). When I try to select these nodes in the Mesh tab, I cannot. I am worried that the Monitor option is not recorded the entire base shear, but only base shear from some nodes. Please advise how to fix this issue.

That's because you used VERTICES in your monitor selection sets.
Instead, you should select all the fixed EDGES, if you want to record reactions from all the mesh-nodes associated with the fixed edges.
Keep in mind the analogy between the STKO's CAD geometry and the associated Mesh (OpenSees's nodes and elements).

That's the strength point of geometry-based modeling. You select an EDGE, and when you refine the mesh, you will automatically get the associated mesh nodes, without worrying about their numbering

[kolozvari]

Thank you for your suggestion. I did what you said, but this does not appear to fix the issue. Please see screenshot below. I have added all edges to the selection set, but the meshed nodes still appear unselected when I choose "select" from the selection set.

Please advise.

[STKO Team]

I did what you said, but this does not appear to fix the issue. Please see screenshot below. I have added all edges to the selection set, but the meshed nodes still appear unselected when I choose "select" from the selection set.

Don't worry, that's just visualization:
If you add an edge to a selection set, in the mesh I will select (visually) only the edges

Look at this small example:
1 selection set with 1 edge selected:

geom.png (159.79 KiB) Viewed 257 times

In the mesh tab, I have that edge discretized with 2 elements and 3 nodes (1, 2, and 5). If I select the selection set, I only select the elements:

mesh.png (148.42 KiB) Viewed 257 times

But the monitor requires nodes, so it loops through the elements and gets a unique set of all those elements' nodes.
You can see it in the analysis_steps.tcl file (search for # Monitor Actor [<the id of your monitor analysis step>])

monitor_targets.png (85.95 KiB) Viewed 257 times

Finally, you can also use the selection set to programmatically do custom operations on mesh entities:
Make a region and ask for a Tcl list of the entities:

set_to_list.png (34.6 KiB) Viewed 257 times

Make a custom script to use the list:

use_list.png (22.76 KiB) Viewed 257 times

This is what you get:

custom.png (21 KiB) Viewed 257 times

[kolozvari]

Thank you for this thorough explanation. I just sent you a private message. I would be very grateful if you can take a look at it and let me know at your convenience.

Thank you.

[kolozvari]

Hello, I am getting the following error (see screenshot) when running one of my models and I am not sure what it means or how to fix it. I think that due to this error, one partition with results ends up being invalid so I cannot process my results. I have MPCO recorder for element selection sets only where I am saving coupling beam rotations. This is same model I sent you earlier via email regarding convergence errors, so you should be able to look at it.

Please advise.

Thank you.