Empirical seismic vulnerability models for building clusters considering hybrid intensity measures. (1st June 2023)

• Record Type: Journal Article
• Authors: Li, Si-Qi; Gardoni, Paolo
• Is Part Of: Journal of building engineering. Volume 68(2023)
• Publication Date: 2023-06-01
• Subjects: Hybrid intensity measures -- Seismic resilience of buildings -- Empirical seismic vulnerability -- Fragility update model -- Vulnerability point-cloud and strip model of building clusters; Building -- Periodicals; 690.05
• DOI: 10.1016/j.jobe.2023.106130
• Languages: English
• ISSNs: 2352-7102

Abstract

Abstract: Seismic risk and vulnerability assessments of buildings are essential to quantify the vulnerability of large-scale urban and rural areas. Regional risk analysis needs a relationship model between ground motion parameters (macroseismic intensities) and structural seismic demands. However, the current practice mainly considers only the impact of horizontal seismic demands, while often neglecting the vertical seismic demands. Additionally, the selection of the macroseismic intensities traditionally depends on the chosen empirical damage assessment method. However, different macroseismic intensities even for the same ground motion might lead to different results. This paper comprehensively considers the impact of multidirectional earthquake sequences, updates the instrument intensity calculation model, and proposes a quantitative method considering hybrid intensity measures to assess the vulnerability of building clusters. The traditional nonlinear regression model is modified and improved using risk analysis, reliability, Bayesian, and probabilist methods. The updated vulnerability model of typical building clusters is then developed based on the hybrid intensity measures. The field inspection database of 57 typical destructive earthquakes in Yunnan, China, is used to update the prediction models for the seismic vulnerability of buildings. Highlights: A vulnerability and resilience assessment model based on hybrid intensity measures is proposed. A lognormal distribution damage assessment model based on hybrid intensity measures is proposed. A vulnerability point cloud and strip assessment model based on the mean seismic damage index is proposed.