example problems

jefferyr · Post by **jefferyr** » Thu Jul 10, 2008 9:37 am

Hello,

I am having problems simply executing part of the online (i.e. current) example 9. I'm executing the following:

puts " -------------------- "
puts " f. RC Section: Circular"
source Ex9f.build.RCSection.Circ3D.tcl
source Ex9.analyze.MomentCurvature3D.tcl

puts " -------------------- "

Both the 1.7.5 and 2.0.0 Mac versions and 2.0.0 Windows version of Opensees won't execute the example as downloaded from the website. Have you seen these convergence errors?

Thanks

--------------------
f. RC Section: Circular
WARNING: CTestEnergyIncr::test() - failed to converge
after: 10 iterations
NewtnRaphson::solveCurrentStep() -the ConvergenceTest object failed in test()
StaticAnalysis::analyze() - the Algorithm failed at iteration: 22 with domain at load factor 7700.1
OpenSees > analyze failed, returned: -3 error flag
WARNING: CTestEnergyIncr::test() - failed to converge
after: 10 iterations
NewtnRaphson::solveCurrentStep() -the ConvergenceTest object failed in test()
StaticAnalysis::analyze() - the Algorithm failed at iteration: 0 with domain at load factor 7700.1
OpenSees > analyze failed, returned: -3 error flag
WARNING: CTestEnergyIncr::test() - failed to converge
after: 10 iterations
NewtnRaphson::solveCurrentStep() -the ConvergenceTest object failed in test()
StaticAnalysis::analyze() - the Algorithm failed at iteration: 0 with domain at load factor 65013.3
OpenSees > analyze failed, returned: -3 error flag
WARNING: CTestEnergyIncr::test() - failed to converge
after: 10 iterations
NewtnRaphson::solveCurrentStep() -the ConvergenceTest object failed in test()
StaticAnalysis::analyze() - the Algorithm failed at iteration: 0 with domain at load factor 11378.2
OpenSees > analyze failed, returned: -3 error flag
DONE Pushover analysis: CtrlNode 1002, dof 6, Curv=0.0102 /inch
--------------------

silvia · Post by **silvia** » Thu Jul 10, 2008 8:22 pm

that was the point of the example.
it ran into convergence problems and it solved them to completion "DONE!"
look at the script and look at the results.

jefferyr · Post by **jefferyr** » Fri Jul 11, 2008 11:39 am

Now that I look at it more closely, you're right. I was sure that I had an example of non-convergence... sorry to have cried wolf too soon.

Jeff

WhiteNoise · Post by **WhiteNoise** » Fri Jan 09, 2009 8:52 am

[quote="silvia"]that was the point of the example.
it ran into convergence problems and it solved them to completion "DONE!"
look at the script and look at the results.[/quote]

Sorry, I didn't get the point.
As I understood, examples9 are non-convergence cases, if so, how can I fix the problem?
I did look in the scripts but I can't figure out the solution.
Could you please give me an advise?

silvia · Post by **silvia** » Fri Jan 09, 2009 8:58 am

the power of opensees is the ability to change your solution strategies if one doesn't work. that's what the example does. the variable ok tells you whether the analysis converged. if it didn't, it modifies the analysis and manages to get over that point and moves on.

WhiteNoise · Post by **WhiteNoise** » Fri Jan 09, 2009 9:24 am

[quote="silvia"]the power of opensees is the ability to change your solution strategies if one doesn't work. that's what the example does. the variable ok tells you whether the analysis converged. if it didn't, it modifies the analysis and manages to get over that point and moves on.[/quote]

This implies I'm free to change the rest of the parameters involved, such as the tolerance, iterations number, etc... I see.
Thank you very much.

silvia · Post by **silvia** » Fri Jan 09, 2009 9:37 am

yes, as shown in the examples.
i would change the tolerance as a last resort.