Stochastic analysis model for vehicle-track coupled systems subject to earthquakes and track random irregularities

This paper devotes to develop a computational model for stochastic analysis and reliability assessment of vehicle-track systems subject to earthquakes and track random irregularities. In this model, the earthquake is expressed as non-stationary random process simulated by spectral representation and random function, and the track random irregularities with ergodic properties on amplitudes, wavelengths and probabilities are characterized by a track irregularity probabilistic model, and then the number theoretical method (NTM) is applied to effectively select representative samples of earthquakes and track random irregularities. Furthermore, a vehicle-track coupled model is presented to obtain the dynamic responses of vehicle-track systems due to the earthquakes and track random irregularities at time-domain, and the probability density evolution method (PDEM) is introduced to describe the evolutionary process of probability from excitation input to response output by assuming the vehicle-track system as a probabilistic conservative system, which lays the foundation on reliability assessment of vehicle-track systems. The effectiveness of the proposed model is validated by comparing to the results of Monte-Carlo method from statistical viewpoint. As an illustrative example, the random vibrations of a high-speed railway vehicle running on the track slabs excited by lateral seismic waves and track random irregularities are analyzed, from which some significant conclusions can be drawn, e.g., track irregularities will additionally promote the dynamic influence of earthquakes especially on maximum values and dispersion degree of responses; the characteristic frequencies or frequency ranges respectively governed by earthquakes and track random irregularities are greatly different, moreover, the lateral seismic waves will dominate or even change the characteristic frequencies of system responses of some lateral dynamic indices at low frequency. (C) 2017 Elsevier Ltd. All rights reserved.