The OpenSees Community

Hi all,
I am looking to model a SDOF structure to perform time history analysis against earthquake loading. My primary objective is to model the structure in such a way that its global force-displacement hysteresis response should follow a specific uniaxial material model (flagshape uniaxial material model). Secondly, I need to carry out the analysis with different period structures. That means I should have flexibility in varying section properties in order to adjust the structures period accordingly. I have couple of options to do so with fibre sections and non-linear beam column elements. But issue with using fibre section is that the global force displacement response of the structure may not necessarily follow the material model I am using since the global response in this way will be a combination of responses at different element level.
Any lead how I can model the structure to meet my desired objectives is highly appreciated!

Cheers,
Mereen

mhscott wrote: ↑Wed Nov 01, 2023 3:01 am This error can be due to so many things. Make a minimal example that replicates the issue.

import openseespy.opensees as ops
import math
import matplotlib.pyplot as plt
import numpy as np

A = 1.0
E = 2.*10**11
Fy = 3.*10**10
Nsteps = 1000
Px = 0.0
Py = 0.0
mass = 1266514796.0*4

# set modelbuilder
ops.wipe()
ops.model('basic', '-ndm', 2, '-ndf', 2)

# create nodes
ops.node(1, 0.0, 0.0)
ops.node(2, 1.0, 0.0)

# set boundary condition
ops.fix(1, 1, 1)
ops.fix(2, 0, 1)
ops.mass(2, mass, 0.0)
ops.uniaxialMaterial('FlagShape', 1, E , Fy, 0.0, 0.25)
ops.element("Truss",1,1,2,A,1)
ops.timeSeries("Linear", 1)
# create a plain load pattern
ops.pattern("Plain", 1, 1)
ops.load(2, Px, Py)

ops.system("ProfileSPD")
ops.numberer("Plain")
ops.constraints("Plain")
ops.integrator("LoadControl", 1.0/Nsteps)
ops.algorithm("Newton")
ops.test('NormUnbalance',1e-8, 10)
ops.analysis("Static")
ops.loadConst('-time', 0.0)

ops.timeSeries('Path', 2, '-dt', 0.01, '-filePath', 'Northridge Beverly.dat', '-factor', 9.81)
ops.pattern('UniformExcitation', 2, 2,'-accel', 2)

freq = ops.eigen('-fullGenLapack', 1)[0]**0.5
dampRatio = 0.05
Kcomm = (2*dampRatio)/freq
ops.rayleigh(0., 0., 0. , Kcomm)
period = (2*(math.pi))/freq

ops.wipeAnalysis()
ops.constraints('Plain')
ops.numberer('Plain')
ops.system('SparseSYM')
#ops.test('NormUnbalance',1e-8, 10)
ops.test('NormDispIncr', 1e-5, 1000)
ops.algorithm('KrylovNewton')
ops.integrator('Newmark', 0.5, 0.25)
ops.analysis('Transient')
ops.analyze(2999, 0.01)
ops.wipe()

running this example is popping up this error. When I try to fix the node two in Y direction it appears. I am not sure whether the applied earthquake load is in Y direction that's why or what. But running the same with static analysis and having load at Y direction doesn't cause this error.

mhscott wrote: ↑Thu May 18, 2023 6:38 am Track down the line of code that causes the error

Hi mhscott is there any updates regarding this? I am also having the same error and the error line correspond to ops.analyze(10) command.

Hey everyone!
I'm currently working on modelling a cantilever column to perform some time history analysis. I'm interested in the force displacement hysteresis behaviour of my model, and I want it to follow a predefined uniaxial material model globally.
I cannot simply use a fibre section and assign the material model because material model is based on a stress strain law and the global force displacement response may not exhibit the same behaviour because in this way the global response will be an aggregated response of the different fibre materials. I am looking for an approach by which my all structure should behave as my assigned material model.
Any comment is highly appreciated.
Regards
Mereen

Hello everyone.

I am currently using the flagshape material model in OpenSeespy. However, I cannot find the user command line in OpenSeespy material library. Currently, I am using the command ops.uniaxialMaterial('FlagShape', tag, E, Fy, Eh ,beta). the command line is working well but the issue is, from my force displacement graph, I am getting a rounded shape at the transition point from elastic to post elastic. However, based on my knowledge for the said model the transition should be smooth. I have also tried the same model in opensees tcl , the shapes of the graphs are smooth at the transition point. I am wondering if there may be an optional command that may control this transition? Here is the link to the source code of the material https://github.com/OpenSees/OpenSees/bl ... terial.cpp .

MereenBaloch wrote: ↑Mon Oct 16, 2023 5:03 pm

mhscott wrote: ↑Mon Oct 16, 2023 3:18 pm FlagShape material is from Chin-Long: https://github.com/OpenSees/OpenSees/bl ... terial.cpp

Is it the unaxial hardening material in the openseespy doc? Because I am unable to find it in user documentation.

Thanks Prof. I got this. I was wondering about the reference command like of the model.

mhscott wrote: ↑Mon Oct 16, 2023 3:18 pm FlagShape material is from Chin-Long: https://github.com/OpenSees/OpenSees/bl ... terial.cpp

Is it the unaxial hardening material in the openseespy doc? Because I am unable to find it in user documentation.

mhscott wrote: ↑Mon Oct 16, 2023 3:58 am The FlagShape material is available in OpenSeesPy. Make sure you are using the latest version.

I can see the the self centering unaxial material but it has certain limitations of parameter beta. The model I just mentioned where value of beta = 1 corresponds to beta = 2.

mhscott wrote: ↑Sun Oct 15, 2023 7:52 am You have to get the actual source code for the model.

Could you please have a look on this. I can find the source code on opensees repo. The code is from ching-long lee and the lastest version of opensees do not have this model. Help will be highly appreciated. Thanks
https://github.com/OpenSees/OpenSees/bl ... rial.h#L40

mhscott wrote: ↑Sun Oct 15, 2023 7:52 am You have to get the actual source code for the model.

Thanks Prof. I have got the source code. How I can implement the code to openseespy?

Hello,
I have an opensees.exe file which contains a specific material model I want to use for my analysis in openseespy. My Question is how I can run the opensees.exe file in python? Or if can not is their a way to access the source code of the opensees.exe file which contains my desired material model extract from their and implement it to openseespy?