Stochastic Analysis of Vehicle-Bridge Coupled Interaction and Uncertainty Bounds of Random Responses in Heavy Haul Railways

Abstract

The systematic running safety assessment of railway bridges presents lots of challenges, one of which is estimating the uncertainty bounds of the structural responses of bridges under vehicle loads with multisource randomness. In this study, a probability safety assessment method is proposed for evaluating the uncertainty bounds of random time-history responses for the stochastic train-bridge coupled system. First, a refined probabilistic model for the train-bridge coupled system (TBS) in heavy haul railway is established with the multi-excitations of random track irregularities, random vehicle loads and stochastic structural parameters. The probability density evolution method (PDEM) is employed to obtain the solution of the time-varying probability transferred between the stochastic excitations and the output of the dynamic responses. Then, to establish a rapid and straightforward approach for the systematic running safety assessment of the TBS, the quantiles of the probability distribution are used to estimate the time-history uncertainty bounds of random responses of interest distributed in real probability functions. Case studies by the field test and numerical simulation are presented to verify and investigate the accuracy and reliability of the proposed method. The results show that the quantiles of the probability distribution proposed are suitable for the systematic running safety assessment of the TBS.