The effects of modeling uncertainties on the residual drift of steel structures under mainshock-aftershock sequences

Abstract

Uncertainty in parameters that define the behavior of yielding components in structures has been shown to affect the validity of predicted near-collapse behavior. After a large-magnitude seismic event, the effect of uncertainties on predicted seismic demands can be intensified by upcoming aftershocks. On the other hand, cumulated damage through seismic sequences should be quantified by demand parameters that are able to properly present the capacity loss of damaged buildings. Residual drifts, an important measure of inelastic deformations, have a significant role in post-mainshock assessment of structures, and of interest in this study is to quantify changes in the sensitivity of residual drifts to the uncertainty of modeling parameters during sequential seismic events. In addition, the relationship between residual drifts and peak transient drifts is investigated. The study is based on two prototype 3- and 9-story steel moment resisting frames subjected to two levels of mainshock intensities that are followed by aftershocks. 20 as-recorded seismic sequences recorded at far-field accelerographic stations are considered. Sensitivity analysis is performed using the one-at-a-time (OAT) method. It was found that seismic demands are more sensitive to strength modeling parameters and beam ductility modeling parameters, and these sensitivities increase by an average of 20% during aftershocks. As-recorded aftershocks increase residual drift demands up to 19 and 15% at risk-targeted maximum considered earthquake MCER level for the 3- and 9-story frames, respectively, while aftershocks increase the dispersion of peak drift demands considerably.