Sensitivity and Reliability Analyses in to Actively Controlled Structures Under Earthquake

Abstract

This study aims for a probabilistic assessment of active seismic control of structures in the presence of uncertainties of the structural model and the stochastic model of the ground acceleration. First, sensitivity analyses based on the Borgonovo sensitivity index are conducted to measure the effect of uncertainty sources on the maximum and root mean square (RMS) of the main structural responses in different seismic hazard regions. Then, the reliability analyses are conducted using Monte Carlo simulation (MCS) and importance sampling (IS) methods to the estimation of failure probabilities and reliability indices of the structures in different seismic hazard regions, force capacity of actuators, and seismic performance levels. Numerical studies carried out on a 10-story structure show that the uncertainties in peak ground acceleration (PGA) and story stiffness have the most effects on the seismic responses of the structure. The effect of the considered uncertainties is negligible on the maximum demand control force. The results also show that equipping the structure with an active control system results in a significant reduction in the failure probabilities of the structure. The target reliability index for the uncontrolled structure in life safety seismic performance level may not be satisfied in high and very high seismic hazard regions, it can however be easily dealt with, using the proposed technique, in the corresponding controlled structures. By increasing the control force capacity, an increment is observed in the reliability index. A high and suitable reliability index is given for both controlled and uncontrolled in the collapse prevention seismic performance level.