Nonlinear Modal Time History

[huffte]

In some situations - namely when all inelastic behavior is limited to link elements - nonlinear modal time history analysis can be much more efficient compared to direct integration time history analysis.

I am not completely well versed in the accuracy of one method compared to the other, but believe they are likely comparable for the type of problem I describe.

It would be useful to incorporate an option to do nonlinear modal time history analysis in soem cases I believe.

[seismosoft]

Dear Huffte,

Yes, you are correct in your observations, and hence yours is certainly a pertinent suggestion, which we will surely add to our developments list.

We note, however, that our top priority in terms of expanding the analytical capabilities of SeismoStruct is introducing first Response Spectrum Analysis, and then complete Nonlinear Static Procedures.

Many thanks for the good suggestions,

Seismosoft Support

[huffte]

It seems that in the current version of SeismoStruct one may not define several time history loadings and run a Dynamic Time History Analysis fro each case independently. It would be a valuable tool for me to be able to do this. In other words, to have a Kobe loading and a Loma Prieta loading defined in the same file, run the analysis, and immediately compare results for the 2 cases.

Regardless, I still fin SeismoStruct a most valuable tool in my research.

Apologies if this is already implemented, but I have searched the help system and can't confirm that it is.

[seismosoft]

Dear huffte,

Thank you for your always pertinent and well-informed comments and feedback.

No, SeismoStruct does not feature a 'Restart' feature that would allow results from previous runs to be saved, and then a new analysis to be started afterwards. The non-negligible complications, in terms of coding and data/memory management, associated to the introduction of such feature are currently hard to justify, not only because we feel there to be other slightly more critical/pressing priorities, but also because a user can always run two separately analyses and compare the results afterwards, outside of the program.

The only thing you could perhaps considering doing, as a workaround, especially if the models are not too large, would be to duplicate, within the same *.spf file, the structure you're modelling, and apply to this other structure the second accelerogram.

Thanks again for the feedback and kind words,

Seismosoft Support

[huffte]

Do you have a tentative schedule for incorporating these items in SeismoStruct?

1. Response Spectrum Analysis
2. Mass participation ratios in Eigenvalue Analysis

Thanks very much.

[seismosoft]

1) first half of 2012
2) already available in current version for translational mass DOFs, available within this year for rotational mass DOFs

Seismosoft Support