Nonlinear dynamic analysis of structures requires strong ground motion time histories (accelerograms) as input. The inherent scarcity of recorded ground motions for specific conditions (magnitude, distance, region, and site) makes utilization of alternatives unavoidable; such alternatives include simulated records and modified real records. There are many simulation methods available, but there is often a preference to using stochastic simulation methods, if justifiable, due to the ease with which many records can be simulated in a generic way. However, there are concerns that such simulated records may not produce similar nonlinear response in structures as real records (or modified real records) due to the lack of realistic phasing and other record characteristics, including peaks and troughs effects and response spectral shape effects. This study investigates peak nonlinear responses of inelastic single-degree-of-freedom systems with different hysteretic characteristics subjected to sets of stochastically simulated records, lightly modified real records, and scaled-real records; the former two of which were proposed by Atkinson and Macias (2009) as representative of expected ground motions for Cascadia subduction earthquakes of M 8.5 at Vancouver, Victoria, and Seattle.

We conclude that (1) the peak nonlinear responses due to the modified records and the scaled-real records are similar if the peaks and troughs effects and response spectral shape effects are taken into account adequately in the choice of scaling factors for the real records; and (2) the peak nonlinear responses due to the simulated and modified records are similar. These findings, though obtained based on limited sets of ground-motion records, are in agreement with previous studies and highlight the need for judicious choices in cases involving the scaling of records. The results also suggest that stochastically simulated records may be an appropriate way to capture overall response potential for both linear and nonlinear structural systems, at least over a range of periods from 0.1 to 2 sec.

You do not have access to this content, please speak to your institutional administrator if you feel you should have access.