Hybrid fragility curve derivation of buildings based on post-earthquake reconnaissance data

Abstract

This study proposes a new hybrid method that uses both of post-earthquake reconnaissance data and numerical analysis results based on a finite element (FE) model. As the uncertainty of a capacity threshold for a structural damage state needs to be estimated carefully, in the proposed method, the probabilistic distribution parameters of capacity thresholds are evaluated based on post-earthquake reconnaissance data. Subsequently, the hybrid fragility curves were derived for several damage states using the updated distribution parameters of capacity thresholds. To illustrate the detailed process of the proposed hybrid method, it was applied to piloti-type reinforce concrete (RC) buildings which were affected by the 2017 Pohang earthquake, Korea. In the example, analytical fragility curves were derived first, and then hybrid fragility curves were obtained using the distribution parameters of capacity thresholds which were updated based on actual post-earthquake reconnaissance data about the Pohang city. The results showed that the seismic fragility estimates approached to the empirical failure probability at 0.27 g PGA, corresponding to the ground motion intensity of the Pohang earthquake. To verify the proposed method, hybrid fragility curves were derived with the hypothetical reconnaissance data sets created based on assumed distribution parameters with errors of 10% and 1%. As a result, it was identified that the distribution parameters accurately converged to the assumed parameters and the case of 1% error had better convergence than that of 10% error.