I have made a non-adaptive pushover analysis on a 3-levels steel frame. The class of my elements are infrmFB and I used as material stl_mp. If I run the analysis without including the geometric nonlinear effects, the pushover curve is very widespread and the analysis continues until the end, but if I include the geometric nonlinear effects, the analysis stops very soon and it does not finish. Is that big difference normal? If yes, why?
Thanks to everyone that will help me.
geometric nonlinear effects in pushover analysis
Re: geometric nonlinear effects in pushover analysis
There really is no such thing as "normal" in this type of analysis. A tall slender structre would certainly have tremendous differences in response results depending upon whether geometric nonlinearities are included or disregarded.
Is your frame a moment frame? Such structures can be very susceptible to predicted changes in response due to geometric nonlinear effects. This is not limited to moment frames, but braced frames are typically much stiffer and less susceptible.
Best of luck vinniec.
Is your frame a moment frame? Such structures can be very susceptible to predicted changes in response due to geometric nonlinear effects. This is not limited to moment frames, but braced frames are typically much stiffer and less susceptible.
Best of luck vinniec.
Tim Huff
Re: geometric nonlinear effects in pushover analysis
Hi hufftle, thanks for your help. Anyway I solved the problem, maybe. In my structural model (MRF) I assigned a diaphragm at every floor and it didn't work as I told in the previous message. So I removed the diaphragms and now it work very well, I mean that the curve that the program provides as output is perfect! Why couldn't I model rigid diaphragms in this case? Should I use another type of constraint?
-
seismosoft
- Posts: 1316
- Joined: 06 Jul 2007, 04:55
Re: geometric nonlinear effects in pushover analysis
vinniec,
If one disables geometric nonlinearity, then P-Delta effects are not considered, hence it may indeed happen that the pushover curve no longer features a post-peak softening behaviour.
If one removes a diaphragm constrain, then the beams became more flexible, and the pushover curve may indeed feature a more softening behaviour. In real buildings, however, slabs do exist, and they do axially constrain beams, increasing their flexural stiffness and strength, thus augmenting also the overall capacity of the frame.
In the Verification Report installed with the program (available also from http://www.seismosoft.com/en/SStructDocumentation.aspx), it is shown how models featuring diaphragm constraints reproduced well the experimental response of 3D frames.
Seismosoft Support
If one disables geometric nonlinearity, then P-Delta effects are not considered, hence it may indeed happen that the pushover curve no longer features a post-peak softening behaviour.
If one removes a diaphragm constrain, then the beams became more flexible, and the pushover curve may indeed feature a more softening behaviour. In real buildings, however, slabs do exist, and they do axially constrain beams, increasing their flexural stiffness and strength, thus augmenting also the overall capacity of the frame.
In the Verification Report installed with the program (available also from http://www.seismosoft.com/en/SStructDocumentation.aspx), it is shown how models featuring diaphragm constraints reproduced well the experimental response of 3D frames.
Seismosoft Support