Seismic Capacity Estimate of RC Columns Considering Different Sources of Uncertainties

With the advent and development of performance-based earthquake engineering (PBEE), one of the effective means adopted by PBEE is the seismic fragility analysis that combines the seismic demand and capacity models. Tremendous efforts have been devoted to investigate the probabilistic estimate for the seismic demand, while few studies focus on the probabilistic capacity estimate for the widely used reinforced concrete (RC) columns. In order to facilitate the seismic capacity estimate of RC columns in practice, a novel probabilistic analysis method is proposed that can take into account the uncertainties inherent in the structures. This method incorporates the BP neural network, finite element analysis and Monte Carlo simulation, meanwhile the uniform design is performed to improve the quality of BPNN. Finally, the feasibility and accuracy of the proposed probabilistic method is demonstrated through a specific case study. The obtained results show that the capacity index of RC columns has a great variability, thus the random variables related to the structural seismic capacity cannot be neglected in the seismic fragility of RC columns.