Page 1 of 1
Cyclic analysis-Extracting the wrong hysteresis curve
Posted: Sat Jan 27, 2024 1:28 pm
by Edmond2000
Dear STKO team:
I have built a 2D joint model and performed a cyclic analysis, this model converges but I can't extract the correct hysteresis curve(RFy-Uy) at the loading point. I've double-checked the model but can't solve the problem. The curve I extracted and model is in the attachment. It is important to me, looking forward to your reply. Thank you.
Best wishes
Re: Cyclic analysis-Extracting the wrong hysteresis curve
Posted: Sat Jan 27, 2024 2:42 pm
by kesavapraba
Hi, Please note the followings in the model:
[1] Reaction force should extracted from the restrained nodes (fixed along Y DOFs), see the encircled nodes in the picture attached, this is where the reaction forces should be extracted and summed in your case.
[2] In the model, I have added monitor custom command to track the force-disp in live.
Please look at the revised model attached herewith.
Re: Cyclic analysis-Extracting the wrong hysteresis curve
Posted: Sat Jan 27, 2024 3:32 pm
by Edmond2000
Thanks very much for your timely reply. I just calculated this model again and I saw the correct curve in the monitor, but in post processor, I extracted RFy-Uy curve as your reply and get wrong result as shown in the attachments.
Re: Cyclic analysis-Extracting the wrong hysteresis curve
Posted: Sat Jan 27, 2024 4:16 pm
by kesavapraba
Hi, the displacement and reactions come from different nodes, you cann't use the same plot to extract them. So extract them separately and use spreadsheet to plot the curves. In other words, to extract RFy, select left and bottom nodes, and to extract Uy, use only the right node.
Alternatively, the force-displacement plot data (from the monitor) can be located as text file "Monitor_LoadDisp.out" in the same folder.
Re: Cyclic analysis-Extracting the wrong hysteresis curve
Posted: Sat Jan 27, 2024 4:34 pm
by Edmond2000
kesavapraba wrote: ↑Sat Jan 27, 2024 4:16 pm
Hi, the displacement and reactions come from different nodes, you cann't use the same plot to extract them. So extract them separately and use spreadsheet to plot the curves. In other words, to extract RFy, select left and bottom nodes, and to extract Uy, use only the right node.
Alternatively, the force-displacement plot data (from the monitor) can be located as text file "Monitor_LoadDisp.out" in the same folder.
Got it. But I have one last question, why is the order of magnitude of the reaction force so small, only about 0.1N, which is not realistic, I am using the SI units(mm-t).
Re: Cyclic analysis-Extracting the wrong hysteresis curve
Posted: Sun Jan 28, 2024 8:23 am
by kesavapraba
Hi, Looking at the deformed shape, there seems to be some inconsistency with the material definition for the joint2d. Please check how the force-disp (or moment-rot curve) is determined for the joint.
Regarding unit system, you can use any consistent unit system, for example, since you use N-mm, moment of 100 kN-m should be entered as 100e6. Please follow the same unit consistency for other definitions.
Re: Cyclic analysis-Extracting the wrong hysteresis curve
Posted: Sun Jan 28, 2024 9:21 am
by Edmond2000
Thank you so much for your assistance, I really appreciate it!
Re: Cyclic analysis-Extracting the wrong hysteresis curve
Posted: Sun Jan 28, 2024 10:00 am
by kesavapraba
You are welcome