Seismic fragility analysis method of building structures integrated with probabilistic seismic hazard analysis

The selection of an optimal earthquake intensity measure (IM) plays a pivotal role in the seismic fragility analysis of building structures. However, the determination of the most suitable earthquake IM remains a subject of debates within seismic engineering community. To address this challenge, a novel method was proposed in this study to produce the recurrence period-based seismic fragility curves by integrating with the probabilistic seismic hazard analysis (PSHA). Departing from the conventional approaches, the proposed method eliminates the reliance on specific earthquake IMs derived from the inputted earthquake ground motions to quantify earthquake intensity levels. Instead, it utilizes the recurrence period of seismic hazards which is obtained from the PSHA as an intermediate variable. Consequently, seismic fragility curves of building structures are expressed as the probabilities of structural seismic responses exceeding the predefined damage states across varying recurrence period of seismic hazards. In the proposed method, three major components are included. First, the PSHA is employed to generate uniform hazard spectra (UHS) at varying recurrence periods of seismic hazards, which subsequently serve as the basis of synthesizing artificial earthquake ground motions. Second, structural seismic response analysis is conducted using the generated artificial earthquake ground motions to predict the engineering demand parameters (EDPs). Third, structural reliability analysis is performed using the probability density evolution method (PDEM) to obtain the probability density function (PDF) of the EDPs. To validate the advantages of the proposed method, a case study was conducted on a twenty-story reinforced concrete (RC) frame structure to generate the recurrence period-based seismic fragility curves. For comparison, the traditional peak ground acceleration (PGA)-based and elastic response spectral acceleration at the fundamental period Sa(T₁)-based seismic fragility curves are evaluated. The investigation results indicated that the PGA-based seismic fragility curves tend to overestimate and the Sa(T₁)-based seismic fragility curves tend to underestimate the recurrence period of structural seismic responses reaching each damage state. The proposed method is beneficial for comparing the difference of seismic performance across a wide range of structural types and vibration periods.