A Chebyshev interval computational framework for propagating parameter uncertainty in train-track-bridge systems

Abstract

The dynamic behavior of the train-track-bridge system (TTBS) under uncertain conditions has significant implications for the safety, reliability, and design of high-speed railways. However, precise probability distribution information based on a large number of samples is often lacking in practical engineering scenarios, so it is more appropriate to consider the uncertain parameters as unknown but bounded non-probabilistic interval variables rather than random variables assuming probability distributions. This study investigates the impact of interval uncertain parameters on the dynamic response of TTBS. Employing the finite element method, the dynamic analysis model of high-speed train-track-bridge system was established and the non-invasive Chebyshev interval analysis method was used to compute the boundary of the system's interval dynamic responses. Numerical results show that even in scenarios with high uncertainty levels and multiple parameters, the proposed method can reduce the computational effort while maintaining high accuracy. This study provides a novel framework for quantifying parameter uncertainty for TTBS, which offers practical insights for safety assessment and design optimization of high-speed rail systems operating on bridges under uncertain conditions.