Empirical vulnerability estimation models considering updating the structural earthquake damage database

Abstract

The seismic vulnerability of regional buildings is a fundamental index to effectively predict and estimate the seismic resilience of urban and rural areas. The structural empirical vulnerability database can be used to develop and generate relatively accurate multidimensional vulnerability models. With the development of instrument intensity and the updating of macroseismic intensity scales, the structural vulnerability assessment model established using the traditional regional structural empirical vulnerability database shows features of low accuracy. Therefore, using innovative methods to update the conventional vulnerability estimation model has become a fundamental scientific problem to be urgently solved. In this study, the mixed effects of instrument intensity and macrointensity and their evaluation scale were sufficiently considered, and the traditional vulnerability level quantification scale was improved. The structural empirical vulnerability database of three typical earthquakes in China (Wenchuan in 2008, Yushu in 2010, and Jiuzhaigou in 2017) has been comprehensively updated, and the multiscale statistical and computational model has been developed. A nonlinear regression model based on multidimensional vulnerability quantification parameters was proposed. A comparison model for predicting the actual seismic vulnerability of four types of buildings considering typical earthquakes was developed. The traditional vulnerability index algorithm was updated, and the point cloud and strip domain quantitative model based on the updated empirical vulnerability database (27,394 buildings) was established.