A study on safety assessment of cable-stayed bridges based on stochastic finite element analysis and reliability analysis

Abstract

It is more reasonable to evaluate the effect of uncertainty included in actual structures using the reliability assessment method. However, using the reliability assessment method to evaluate complicated structures, such as cable-stayed bridges whose limit state function is given as a nonlinear function, was very difficult. Therefore, in this study the structural response characteristics are investigated quantitatively using Stochastic Finite Element Analysis (SFEA) according to the external uncertainty of random variables included in cable-stayed bridges, and Reliability Analysis (RA) considering external uncertainty is performed more efficiently using the result of SFEA. For this approach, after performing the initial shape analysis of the cable-stayed bridge, improved Stochastic Finite Element Method Program (SFEMP) is developed for executing nonlinear SFEA by considering correlation of random variables. The feasibility of the developed program was investigated by comparing the analysis results of beam-cable structures with those of the verified program using Direct Monte Carlo Simulation (DMCS). The effect of each random variable on the structural response was examined by performing sensitivity analysis. The safety index and failure probability according to the correlation was investigated by applying RA that considered the correlation of the major random variables. The critical percentile distribution of the RA result is re-evaluated quantitatively considering the effect of the internal uncertainty.