Hi, silvia,
I've just conducted a nonlinear fiber section analysis, in order to test for the Arc-Length Integrator.
Compared with the results by Displacement Integrator, two curves fits well in the ascending branch... but in the descending branch, the curve with Arc-Length flip-flops, as the one with Displacement Integrator runs smoothly...I think that the differences may be contribute to the selection method in arc-length method...so, what's the selection criterion for the two roots in arc-length method? and can arc-length method be used to control the direction of displacement increment??
BTW: is there any preconditioning method in OpenSees when the condition number of the stiffness matrix becomes very large??
about Arc-Length Method
Moderators: silvia, selimgunay, Moderators
Re: about Arc-Length Method
Hello,
Can you post your tcl files? I also want to see the difference. Thank you very much.
[quote="Xiang Wang"]Hi, silvia,
I've just conducted a nonlinear fiber section analysis, in order to test for the Arc-Length Integrator.
Compared with the results by Displacement Integrator, two curves fits well in the ascending branch... but in the descending branch, the curve with Arc-Length flip-flops, as the one with Displacement Integrator runs smoothly...I think that the differences may be contribute to the selection method in arc-length method...so, what's the selection criterion for the two roots in arc-length method? and can arc-length method be used to control the direction of displacement increment??
BTW: is there any preconditioning method in OpenSees when the condition number of the stiffness matrix becomes very large??[/quote]
Can you post your tcl files? I also want to see the difference. Thank you very much.
[quote="Xiang Wang"]Hi, silvia,
I've just conducted a nonlinear fiber section analysis, in order to test for the Arc-Length Integrator.
Compared with the results by Displacement Integrator, two curves fits well in the ascending branch... but in the descending branch, the curve with Arc-Length flip-flops, as the one with Displacement Integrator runs smoothly...I think that the differences may be contribute to the selection method in arc-length method...so, what's the selection criterion for the two roots in arc-length method? and can arc-length method be used to control the direction of displacement increment??
BTW: is there any preconditioning method in OpenSees when the condition number of the stiffness matrix becomes very large??[/quote]