IDA curve

[sadraddin]

Hello huftte or support team
I have some questions about IDA curve
Does seismostruct create multi-record IDA curve? if yes, How I can do that?
Does seismostruct create capacity curve and indicate collapse point? or How I know that my structure is failed in IDA?
I am analyzing 12 story RC building. so, when I run my model with IDA analysis it takes a long time may be 1 week to complete. Is there any way to reduce the time?
Does seismostruct analyze RC shear wall structure for the same 12 story without any beam and column?
Thanks for your help

[huffte]

Happy New Year, sadraddin.

I believe the best way I have found to perform IDA is to get your initial file with the first accelerogram as the specified dynamic loading. Then you can modify that file, change the loading to your second accelerogram, save under a new name, etc. . You could batch execute and use the batch utility to export results from all files into Excel, where you would do the post-processing, averaging of responses to the various records, and so on.

I have not personally tried the batch utility with IDA, so I'll be curious to learn how you fare in this endeavor.

As to your question regarding what defines failure - again, that is up to you. Do you have a drift limit you have to meet? Is material strain your failure criteria? Is collapse your criteria?

For reducing the time, there are options, but you will have to decide if they are acceptable or not:
(1) restrain all DOF's that are of no interest
(2) maximize your time step while maintaining accuracy
(3) in Project Settings --> Elements, uncheck 'Carry out Stress Recovery' - again, you'll have to decide if this is acceptable for your particular situation
(4) use infrmFB (force-based inelastic elements) rather than infrmDB (displacement-based inelastic elements) where possible and do not discretize the elements
(5) it's only a guess, but if you could zero out all element masses and create your own lumped masses (only), that might speed things up; distributed masses can slow things down a bit with some software
(6) suppress all real time plotting if running without the batch utility
(7) relax the convergence criteria - again, you have to determine how much is acceptable
(8) if you are running a 3D model with bidirectional loading, this is not an option, but could you simplify to a 2D model?
(9) use 'super-elements' by defining an artificial element having the properties of an entire floor in a single member. This can obviously be tricky when torsional effects are present, but you may wish to look into it.

As for the shear wall structure with no beams and columns, I think that would be your decision as well. If you can assign the appropriate properties to capture the shear wall behavior with vertical frame elements and lumped masses only, then that should be a relatively simple model.

Best of luck sadraddin.

[sadraddin]

Thanks huftte for your great responses
Happy new year

[sadraddin]

Hello huftte
Could you tell me how to enter height of shear wall in Seismstruct v.7?
Thanks

[huffte]

I am afraid that I do not understand your question, sadraddin. The height of an element is simply the distance between element ends? If you mean the thickness, then it depends on how you are modeling the shear wall. Are the dimensions such that both shear and flexure deformations are significant? Then you may have to come up with effective properties and dimensions to accurately model the structure.

[sadraddin]

I mean story height for shear wall because it does not exist in seismostruct. for example, I have shear wall 20 ft width and 1 ft thickness and story height (shear wall height) is 10 ft. So, how I can enter this 10 ft?
Thanks for your help

[sadraddin]

Thank you huftte I got it

[sadraddin]

Hello huftte
Is there any criteria to choose IDA scale factor range?

Thanks

[huffte]

It depends on what you wish to accomplish, sadraddin. Maybe you have a suite of records scaled to 1.0 times your 'design' loading and you wish to investigate 5 increments up to that loading. Then your factors would be .2, .4, .6, .8, 1.0. Or maybe you wish to investigate what happens if the 'design' load is exceeded by 50%. Then your scale factors might be .3, .6, .9, 1.2, 1.5. Or maybe you aren't scaling records to a 'design' loading at all and want to look at the inelastic response to various unmodified records. Then you might have to try different values for the maximum scale factor on IDA for each record in order to get the maximum inelastic response you desire.

So it really is up to you sadraddin, and it is an important decision. You really need to think about what you are trying to accomplish with your analysis and set the factors accordingly.

Best of luck sadraddin.

[sadraddin]

Thanks huftte do you have any reference or paper talking about this incrementation?
I want to know the range of my incrementation for example from 0.1 to 2 because I am doing master thesis on IDA so I have to determine the range of IDA scaling.
Thanks your responses are very helpful for me I really appreciate that