load - displacement hybrid control loading system

[dadadahe]

Dear STKO

I am a beginner of STKO software.

I want to carry out hysteresis analysis on this model to get hysteresis loops.

And I want to adopt a load - displacement hybrid control loading system. First, load control is adopted. When the load reaches 0.4Py, 0.6Py, 0.8Py and 1.0Py, the cycle is repeated once respectively, where Py is the yield load and its value is 120.75KN. When the model yields, displacement control is adopted, and the Δy value is 36mm. Each stage is repeated twice. The specific loading system is shown in the attachments.

Can you teach me how to do load-displacement hybrid control loading, and I would appreciate it if you could give me an example or do it directly on my model.

I attach the model in case you need it. Thank you in advance for your support!

[STKO Team]

First I had to fix few things in your model:

• You mixed some 2D entities with 3D ones. I changed everything to 3D. It will lead to better post-processing if you would like to see stresses and strain in fiber sections. (Also, keep in mind that if you go for a 2D model, the Z axis is always in the out-of-plane direction, and OpenSees does not consider the local axes on a 2D beam). So if you don't want to have those troubles, just go for a full 3D model (it's easier)
• You don't need 2 interactions for the Hinge material (Rz) and the EqualDOF. You can apply both the ZeroLengthElement and the EDOF condition on the same Interaction.
• Since your ZeroLengthElement provides a material only to the Rz DOF, the EqualDOF should connect ALL other DOFs! Otherwise your system will be singular

Regarding the Mixed Load-Disp controlled analysis.
What you did is correct for the Disp-Controlled analysis.

to do a mixed load (first) and disp (after) control, you need to:

• Create another Path timeSeries with the load multipliers of the load cycles (from 0 to 1) with a total duration of 1 pseudo-second.
• Create a load = Py
• Create a load pattern with the previously mentioned timeSeries and load
• Create a load control analysis (after this you can set the loadConst, and reset pseudotime to 0)
• You can use the loadConst without removing the previous load just because the last point of your load cycle is 0! So it would make no difference
• now you can go on with the steps you just did (i.e. loadPattern and analysis for disp control)

IN this model I just fixed the issues I found.
You can try to implement the mixed load-disp approach and let us know if you have issues

3-2.zip

(155.47 KiB) Downloaded 170 times

[dadadahe]

Dear stko team
Thank you very much for your reply. Your support has greatly helped me learn to use stko. According to your suggestion, I try to do a mixed load (first) and disp (after) control. However, some problems were encountered:
1. How to realize the load control first and then the displacement control? How to set the sequence in stko.
2. About the unit of load value in load control, do I need to convert it to N · m
3. If you can, could you help me modify my model directly.
Thank you again and look forward to your reply

[STKO Team]

This is how the 2-staged analysis should be done:

3-2-new.zip

(167.78 KiB) Downloaded 166 times

The first analysis, which is load-controlled will not converge now because the strength of your hinge (179000 N*mm) is less than the maximum Py value (120750 N) time the distance from the hinge (1050 mm).

The imposed moment on the hinge will then be 120750*1050=126787500 N*mm (126.7875 kN*m), which is way larger than your hinge strength 179000 N*mm = 0.179 kN*m.
Are you sure about that value? I think you have to multiply it by 1000...

[dadadahe]

Thank you very much for your help and wish you a happy life!

[ValentinaSB]

You are welcome!
Let us know if you have troubles

Best, Valentina

[dadadahe]

Dear stko team,

Why is the displacement value of the control point not zero after the end of the load control stage. I want to make the displacement of the control point zero before starting the displacement control stage. Could you help me modify it,

Thank you very much and look forward to your reply.

[STKO Team]

This is a corrected version of your model:

3-8 mixed control.zip

(168 KiB) Downloaded 162 times

There were few errors:

• The Path time series for the load-control has a wrong duration. It should end at time=1, same as the duration of your first analysis

  

  linear time series is short.png (29.31 KiB) Viewed 2489 times

• The first analysis should be executed before adding the second laod pattern (sorry! this was my mistake in the file I sent you before)

  

  analysis order.png (11.56 KiB) Viewed 2489 times

Now you will still see a non-zero displacement at the end of your first load cycle:

residual disp.png (19.92 KiB) Viewed 2489 times

This is perfectly normal: Your first analysis imposes (via the Path time series) a zero load at the end. But due to the accumulated plastic displacement, the displacement will not be zero at the end.

hyst.png (22.48 KiB) Viewed 2489 times