Hysteresis response
Hysteresis response
We have modeled a pier in FB elements. Mander et al model for RC element with lumped mass atop is considered. To include P-Delta, we sub-divide the member into four elements. For verification, we perform dynamic analyses (earthquake acceleration history) once again without including geometric nonliterary. We observe, that the hysteretic response, i.e., base shear vs. tip displacement w.r.t. fixed base changes with number of elements.However, global response remain same. I suspect that damping has some influence here what we consider as initial stiffness proportional. We set damping to zero and then we find a good match regardless of the element subdivision. Please explain and guide my point.
-
seismosoft
- Posts: 1193
- Joined: 06 Jul 2007, 04:55
Re: Hysteresis response
Dear Rana,
In general, small differences between the models are expected, especially in simple ones, because of the different number of (inelastic) integration sections. In your 1-element model you have for instance 4 integration sections, in a 4-element model you have 4x4=16 integration sections. In simple cantilevers this would mean that in the first case you concentrate inelasticity in one section (the bottom one), in the latter case you might have 2 or even 3 locations with inelasticity. Damping should also change things (in the first case damping forces are applied at the top and the bottom of the member, in the second case all along the member height).
Note also that you can include P-delta effects in your models simply by checking the corresponding checkbox in the Project Settings dialog box (Analysis tab), no need to change the discretization of the structure for that.
Seismosoft Support
In general, small differences between the models are expected, especially in simple ones, because of the different number of (inelastic) integration sections. In your 1-element model you have for instance 4 integration sections, in a 4-element model you have 4x4=16 integration sections. In simple cantilevers this would mean that in the first case you concentrate inelasticity in one section (the bottom one), in the latter case you might have 2 or even 3 locations with inelasticity. Damping should also change things (in the first case damping forces are applied at the top and the bottom of the member, in the second case all along the member height).
Note also that you can include P-delta effects in your models simply by checking the corresponding checkbox in the Project Settings dialog box (Analysis tab), no need to change the discretization of the structure for that.
Seismosoft Support