OOP Micromodeling of Masonry Wall

[kesavapraba]

Dear Sir/Madam
Can you please help in figuring out the issue with the attached model? I followed the same procedure as explained in the video tutorial by Dr. Massimo in the following link:
https://www.youtube.com/watch?v=m65mJq6Uce8
The issue is that I could not get the same pressure vs displacement plot. Also, that automatic pop of the Gnuplot does not appear in my windows PC. Thank you

[STKO Team]

That was done with a very old version of STKO and the custom script for plotting the graph and doing the incrementation was not working with how the DisplacementControl has been changed in the latest version.

Here is how it can be done now.
You have adaptive time step directly in the Analsyis command, and the monitor tool for real.time plot!

Side note: I had to change the value of pressure. At the time, I used a value such that the total pressure*area would give a unit value (1 N), so that I could use the pseudoTime as total reaction, for an easy plot.

Now this is not necessary anymore because we have the plot tool, and it is not going to work anyway becasue the DisplacementControl code in OpenSees has changed, and if you put a very low reference load with penalty constraint handler, it can give convergence issues

oop_urm_wall.zip

(1.5 MiB) Downloaded 324 times

[kesavapraba]

Dear STKO Team
Thanks a lot for your reply and support. I can now understand the process.

[STKO Team]

You're welcome!

[kesavapraba]

Dear Sir/Madam
I would like to invite your comments on the following results obtained using the above model. The model used MITC4 element. In addition to MITC4 shell, I also tried ASDEAShellQ4 element. Please consider the following graphs.

Graph1-> ASDEAShellQ4 element with Corotational option
Graph2-> MITC4 element with -updateBasis checked in
Graph3->ASDEAShellQ4 element with Linear option
Graph4-> MITC4 element without -updateBasis checked in

I use ASDEAShellQ4 for my study where I need to apply both the in-plane and out-of-plane loading and capture the respective responses. I understand ASDEAShellQ4 has some advantages over other elements. However, for the model in the discussion, MITC4 without -updateBasis seems to perform well. Kindly share your comments. Also, which shell element would be accurate for my case and why? I am interested in capturing both peak strength and displacement.

Thank you

[STKO Team]

However, for the model in the discussion, MITC4 without -updateBasis seems to perform well.

It performs fairly well, but if I got it correctly, it should be Graph4, the one that is overestimating the force?

Also, they only huge difference you see here, is in the post peak regime, (after a displacement = 5) because then strain softening appears, and the shellMITC4 has a weird stiffening effect due to the full integration of the drilling DOF.

Also, the drop of the experimental curve you see after displacement = 30, I don't think it is a load drop, just a reversal

[kesavapraba]

Thank you for the reply. Yes, my mistake, the graph 3 and 4 should interchanged. Please read the figure as per the description below.

Graph1-> ASDEAShellQ4 element with Corotational option
Graph2-> MITC4 element with -updateBasis checked in
Graph4->ASDEAShellQ4 element with Linear option
Graph3-> MITC4 element without -updateBasis checked in

My interest was to understand why AsdeaShellQ4 overestimates the peak strength.

[STKO Team]

Actually it's quite hard to explain...

The original implementation of the ShellMITC4 had a full integration for the drilling DOF (rotational dof about the shell normal axis).
This full integration leads to overstiffening of membrane behavior and prevents strain localization. As a small workaround, we implemented a scaleFactor for drilling, which by default is set by STKO to 0.01. Note however that in OpenSees this is not available.

Probably the scale factor I used was to low, so I had to slightly overestimate the fracture energy which in that experimental test was not given, if I remember properly.

After some time from that webinar, we implemented the ASDShellQ4 which uses a correct reduced integration for the Drilling DOF, so no scale factor is necessary...

Try this, use the MITC4 with the default scaleDrilling set to 1 (this is what the OpenSees original implementation uses).

[kesavapraba]

Thank you for the details. It's fine.

[STKO Team]

You are welcome!