OpenSeesPy Reliability Analysis and Parameter problem.

[erkanbicici]

Previously, in tcl, I was adding parameters in the following way.

randomVariable $rndmVar1 lognormal  -mean -3.503  -stdv 0.5254
randomVariable $rndmVar2 lognormal  -mean $fy  -stdv 3.225

parameter $param1 randomVariable $rndmVar1 element 10 fc
parameter $param2 randomVariable $rndmVar2 element 10 fy

However, now in python, the following code is not working. I have been working on it for a long time but I could not figure what I am missing. Any help will be great.

ops.randomVariable(1, "lognormal", "-mean", 20, "-stdv", 0.52)
ops.randomVariable(2, "lognormal", "-mean", 420, "-stdv", 0.52)

ops.parameter(1,"randomVariable",1, "element",1, "material",Conctag,'fc')
ops.parameter(2,"randomVariable",2, "element",1, "material",Steeltag, 'fy')

The code gives the following error. Even I have check the source code, I dont understand the problem.

Parameter::addComponent 1 -- no objects were able to identify parameter material 1 fc 
Parameter::addComponent 2 -- no objects were able to identify parameter material 3 fy 

Thank you

[mhscott]

Can you provide a short working script that reproduces the error? The commands you show in Tcl and Python are not consistent with each other.

[abtin139]

However, now in python, the following code is not working. I have been working on it for a long time but I could not figure what I am missing. Any help will be great.

[mhscott]

Please post the code

[abtin139]

i sent my code to your email

[mhscott]

I have not received an email. Please post the script here, in this thread.

[abtin139]

Sure. I want to know how to do reliability analysis with opensees? Because I enter the randomVariable command, but I can't get an answer for any variable or even in beam modeling. For example, with the Latin Hypercube Sampling simulation method, what is the shape of this concrete structure? In general, if it is an example, I need help in this field.
It's write in spyder(python 3.11).
reliability.zip

[mhscott]

I'm looking for a simple script, like 20 lines of code, that demonstrates the error. I am not going to look through multiple files in the zip you posted. I only want to know the error so that it can be fixed.

[abtin139]

i'm sorry.

from openseespy.opensees import * 
from scipy.stats import norm 
wipe()
model('basic', '-ndm', 2, '-ndf', 3)
node(1,0,0)
node(2,3,0)
geomTransf('Linear', 1)
meanR = 2000; sigR = 135  # for E_Steel
randomVariable(200, 'normal', '-mean', 2000, '-stdv', 135)
element('elasticBeamColumn', 1, 1,2, 0.08, 2e11, 0.001, 1)
fix(1,1,1,1 ) 
timeSeries('Linear', 1 )
pattern('Plain', 1 , 1 )
load(2, 0,-5e6,0)

[mhscott]

This code does not produce an error on the latest version of OpenSeesPy (3.5.1.12).

[abtin139]

Therefore, for more complex models, there is no longer a need to connect opensees to MATLAB, and with this library (openseespy), the Latin Hypercube Sampling (LHS) or mcmc or mc method can be implemented?
thank you so much.