Correlation analysis of structural response and ground motion intensity measures

Abstract

Estimation of losses due to earthquake-shaking induced building damage is an important step in Performance Based Earthquake Engineering (PBEE). The implementation of PBEE depends on the ability to predict Engineering Demand Parameters (EDPs) from ground motion Intensity Measures (IMs). The response of a building during an earthquake is governed by the complexities in the ground motion that is governed by the seismic source, the Earth structure, and local site conditions. This thesis aims at evaluating the correlation between building response and intensity measures. In the first part of my study, I compute building responses from large sets of recorded ground motions (up to 2500) by considering Reinforced Concrete (RC) frames with different natural periods (0.1-1.6s). I then calculate several IMs from recorded ground motions in the near field region of moderate-to-large earthquakes (𝑀𝑤 ≥ 5). For the ground-motion waveforms, I perform time-history analysis to evaluate building response and extract the engineering demand parameters in terms of interstory drift. From the response of buildings subjected to ground motion records, I infer that interstory drift ratio strongly correlates with spectral measures of ground motion intensity (correlation of 0.85). The results presented in this study help to identify optimal IMs in the context of PBEE. Recorded ground motions may often be insufficient to study the effects of near field ground motion complexity on building response because of limited data. Therefore, earthquake engineers increasingly rely on simulated ground motions based on numerical modeling approaches that include realistic earthquake sources and medium. The complexities of these models contribute to the ground motions and will have potential impacts on the building responses. Correspondingly, in the second part of my thesisI simulate broadband ground motions (frequency range 0 – 10 Hz) for (𝑀𝑤 = 6.7) Tottori Earthquake (2000) for 30 different source realizations. I perform building response analysis for a RC frame (natural period T = 0.96s) subjected to simulated ground motions. From the analysis, I observe and quantify the dominant source complexities on the interstory drift.