Seismic risk model for regional buildings that considers the influence of temperature and intensity measures

Abstract

Probability-based seismic risk and random ground motion theories are used to develop high-precision seismic risk functions and models of regional building portfolios. Seismic intensity measures and structural demand parameters are critical in developing seismic fragility models for building clusters. Earthquake damage data indicate that ambient temperature may influence structural fragility, but the effect of temperature on seismic risk is commonly ignored. A method for quantifying seismic intensity is proposed in this study and optimized using 450,000 acceleration records detected by 15 stations during the Luding earthquake in Sichuan, China, on September 5, 2022. Considering the influence of different temperature fields, the proposed seismic intensity quantification method was used to evaluate structural damage during two destructive earthquakes in Xinjiang, China (the Aktao earthquake on September 2, 2003 and the Zhaosu earthquake on December 1, 2003). Three structural empirical seismic risk matrices and cloud maps were developed, and a vulnerability membership curve was generated using fuzzy decision-making, a membership algorithm, and nonlinear regression. A structural seismic risk index function was proposed, and vulnerability cloud diagrams and curves based on actual structural seismic vulnerability datasets were generated. Together, these analyses demonstrated the impact of temperature and intensity measures on the seismic risk to regional buildings. A model was established to evaluate the seismic risk to typical structures and was validated using historical earthquake damage data from China. This method improves the accuracy of earthquake risk and vulnerability assessments for typical building clusters given the influence of temperature. The developed seismic risk prediction model enables the evaluation of vulnerability and resilience of typical structures under low and high temperatures.