Ibarra Curve vs. Pinched Asymmetric Curve

[rashid]

Hi Seismosoft Team,

1. If I need to model a structure with its members exhibiting pinched hysteretic response as well as strength degradation post the ultimate strength, then what advantages does the Ibarra model has over the new pinched asymmetric curve or vice versa (I'm aware that the Ibarra model has cyclic as well as in-cycle degradation)?

2. What could possibly be the limitations of the pinched asymmetric curve given that it does not have a degrading branch in the backbone curve ? The curve can simulate degradation using the ductility and energy based factors, but what role could a degrading branch have played, in simulating degradation, if there were one ?

Thanking you in anticipation.

Muhammad Rashid

[seismosoft]

Hello rashid,
We really cannot comment on the merits and the drawbacks of each model, and the choice depends on your actual needs (e.g. structural model, pushover or cyclic analysis). We feel that one of the Ibarra-Medina-Krawinkler curves, which are general and versatile, should be ok for most of the cases, however we do not really know if the fact that for instance the pinched curve does not possess a descending branch in the backbone curve affect your results or not.
Seismosoft Support

[rashid]

Hi Seismosoft team,

Thank you for your kind response. I've another query regarding the Ibarra curve.

The Ibarra curve implemented is the modified IMK model which has the additional rotation parameter of Theta U in addition to Theta PC. Theta U is to account for the ductile tearing of steel sections, which leads to complete loss of strength, as opposed to Theta PC which is for gradual reduction in strength due to local buckling. Seismostruct does not accept values of Theta U greater than Theta PC and therefore cannot account for ductile tearing before Theta PC.

Theta U does not have an empirical prediction equation as Theta PC. The only recommendations for Theta U are 6-7% of drift based on the references below. I've one design scenario where Theta PC comes out to be 7%, which is the same as Theta U and therefore does not work in Seismostruct.

Do you think imposing a limitation on Theta U being greater than Theta PC is correct ? The second reference below also states that Theta U can also be used for any connection failures which leads to complete loss of strength ( i.e. in the post capping branch or even in the strain hardening branch). However this limitation in Seismostruct does not allow modeling such failures. Please correct me if I'm wrong in any way.

https://ascelibrary.org/doi/10.1061/%28 ... 1X.0000376

https://datacenterhub.org/resources/948 ... 9-2008.pdf

[Fanis_Moschas]

Dear User,
Thank you for your comment. Indeed in the IMK model a θu lower than θpc could be used to model complete loss of strength before the system reaches the residual branch (for example in the case of brittle failure of connections).
We will consider removing the limitation you referred to in a future release of Seismostruct.

[rashid]

Hi Seismosoft team,

Is there any plan to release some verification examples for the Pinched Asymmetric Curve to demonstrate its capabilities for various pinched and degrading behaviors ?

Best regards,
Rashid

[seismosoft]

Not really, but you can find several examples on the implementation of the link element in the verification report (and the relevant files that are included in the installation), as well as in the models database in our website:
https://seismosoft.com/sample-download
Seismosoft Support