June 22, 2019 at 1:41 pm #2482
How can we find the location of plastic hinges in OpenSees?
Thanks!June 22, 2019 at 4:30 pm #2490
It depends on how the plastic hinges have been modelled. Please upload a simple file to let us check.June 22, 2019 at 4:39 pm #2491
Thank you so much for your attention.
I have attached a 2-story frame under pushover analysis.
Attachments:You must be logged in to view attached files.June 22, 2019 at 5:00 pm #2498
In the attached file, hinges are modelled inside the beam element (dispBeamColumn), so they cannot be displayed since they formed into the fiber element.June 22, 2019 at 5:02 pm #2499
Yes, so what is your suggestion?June 22, 2019 at 6:20 pm #2501
If you can switch solver, use the one included in Designer and keep working with hinges.
If you like to maintain the use of Opensees, use separate elements for hinges.June 23, 2019 at 12:35 am #2503
Thank you for the comments. Is there any video that shows how we can do a nonlinear analysis using your product?June 23, 2019 at 9:14 am #2506
Not yet, we will add it soon.June 23, 2019 at 4:45 pm #2512
In my opinion, to monitor the formation of plastic hinges I found the most accurate way is to check the strain value in each section.
As you know, in OpenSees we mesh the section to several fibers.
Using the below command we can get the values of strain in each fiber for the integration points (usually number of integration points is 5; so we have 5 sections).
ecorder Element -file $dataDir/StressStrain51_Sec1.out -ele 51 section 1 fiber $fiberCoordinateZ $fiberCoordinateY stressStrain
ecorder Element -file $dataDir/StressStrain51_Sec5.out -ele 51 section 5 fiber $fiberCoordinateZ $fiberCoordinateY stressStrain
We should extract the maximum strain among the fibers of each section. Then, we can compare it with the yield strain (e.g. the attached photo show this ration in 5 stations (integration points) for each element).
This technique needs to generate many many output files (above command). But I do not know that is NextFEM able to do such post-processing or not?
Attachments:You must be logged in to view attached files.June 23, 2019 at 4:53 pm #2514
In the following of my comments, I would like to add that recently I made a Matlab code for generation of the mentioned “recorder” command. For a 2-bay and 2-story frame, 8000 commands were generated (total number of fibers = 8000).
The volume of files was over 14 GB!
I think if NextFEM is able to do such post-processing it would be great!June 23, 2019 at 10:28 pm #2515
NextFem Designer at the moment supports the import of simple datasets containing a single integer value for each node or element for each step. That was supposed to work in conjunction with the procedure009.tcl file given with our program (look for the checkProc parameter to be passed to the functions). The current capability seems quite limiting with respect to the use you want to make of it. We cannot however exclude a wider support for additional datasets capabilities in the future.
As a side note you may want to reduce the number of fibers monitored to those where yield is likely to occur, like the top and tips of the flanges or the midpoint of the web if shear also plays a big role.
You must be logged in to reply to this topic.