Seismosoft | Earthquake Engineering Solutions

Dear SeismoSoft team,

I perform static nonlinear analysis of a two-story steel special concentrically braced frame. I have made two models - first one with concentrated plasticity in a plastic hinge in the middle of the brace, in the second one I use the SeismoStruct v.7.0.6 infrmFB element and a Minegotto-Pinto steel material model for modelling bracing elements. To provoke buckling, an initial camber is applied in the middle of the brace.
After performing the analysis with monotonically increasing lateral displacements I got the resultant capacity curves. Both curves matches quite well with one difference in the plastic range - there is no falling branch in the curve for the model with distributed plasticity, that indicates loss of structural strength and stiffness rather it shows overstrength with increasing displacement of the control node.
I explain this effect with the fact that by modelling concentrated plasticity in a plastic hinge when reaching the accepted fracture criteria of the element, the strength of the element drops to zero but the analysis goes on with stress redistribution. That is not the case when using distributed plasticity.
Could you give me some advice about simulating fracture of element with loss of strength, which doesn't cause the analysis to stop but leading to stress redistribution and drop of stiffness and strength of the structure in the case of distributed plasticity?

Best regards,
Raycheva

Hi Raycheva.

Did you set up performance criteria? It seems you should be able to specify the fracture strain as a performance criterion and change the residual strength option to "No residual strength". Note that the default for new performance criteria is "Keep residual strength".

That might get closer to the behavior you wish to model.

Best of luck Raycheva.

Hi Tim Huff,

Thanks for the reply. Yes, I did set up performance criteria with the option "No residual strength" but as soon as the criteria is reached in any of the braces, the analysis terminates with logout message "Max_Tol"and "Sol_prb". I thought of some modelling mistake that I have done and made another model of the two-story frame, removing one of the braces (as if it has fractured) with the option "Keep strength" when fracture criteria is reached, then there is no problem with the analysis. I think of the possibility of changing any options in the material model.

Best wishes,
Raycheva

OK. Maybe adjusting the convergence criteria could help. Is it possible that several fractures occur almost simultaneously in the mode - enough to render the frame unstable very soon after the initial fracture?

Just thinking out loud, Raycheva. You'll get there. Best of luck.

Thanks for the advice. Actually it is possible several fractures to occur almost simultaneously so I tried to set performance criteria just for one of the braces, not for all of them. Unfortunatelly the result was the same - analysis terminates with the same logout messages. I'm trying to adjust the convergence criteria but with no success by now.

Kind regards,
Raycheva