• CN: 11-2187/TH
  • ISSN: 0577-6686

Journal of Mechanical Engineering ›› 2026, Vol. 62 ›› Issue (16): 95-108.doi: 10.3901/JME.260626

Previous Articles    

Stochastic Analysis of Vehicle Ride Quality under the Correlation of Amplitude States of Track Irregularity

LUO Zhixiang1,2, GUO Bingbin1, ZHAO Yu2, ZHOU Shengtong2, XIAO Qian2, ZHOU Kai3, LI Hongguang4,   

  1. 1. The State Key Laboratory of Heavy-duty and Express High-power Electric Locomotive, CRRC Zhuzhou Locomotive Co., Ltd., Zhuzhou 412001;
    2. Key Laboratory of Railway Industry on Intelligent Operation and Maintenance for Locomotive and Vehicle, East China Jiaotong University, Nanchang 330013;
    3. College of Transportation, Tongji University, Shanghai 201804;
    4. The State Key Laboratory of Mechanical System and Vibration, Shanghai Jiao Tong University, Shanghai 200240
  • Received:2025-08-08 Revised:2026-02-07 Published:2026-10-14

Abstract: The amplitude scale factor (short for amplitude factor) of track irregularity can be used to describe the improvement or deterioration of the amplitude state of track irregularity. However, the amplitude states of alignment and longitudinal level irregularities in the left and right rails often exhibit random correlation. To this end, an approximate description method for the amplitude state of track irregularity based on typical irregularity samples and their amplitude factors is proposed, and the influence of the correlation between alignment and longitudinal level amplitude states on vehicle ride quality is investigated. Firstly, a vehicle dynamics simulation model under the excitation of typical irregularity samples is established. Secondly, the amplitude factors of alignment and longitudinal level irregularities are taken as random variables, and three Copula functions with different tail-dependency attributes are selected to characterize the correlation between these two variables. Then, to address the high computational cost of the vehicle simulation model, a stochastic analysis method and its implementation procedure based on adaptive sparse polynomial chaos expansion(PCE) and analysis of covariance(ANCOVA) sensitivity are presented under the constraint of limited experimental design samples. Finally, on the basis of model verification, the probability distribution characteristics and global sensitivity of vehicle ride quality indices under different Copula types and different correlation degrees are discussed in detail. The results show that:① An increase in the correlation degree between alignment and longitudinal level amplitude factors leads to an improvement in the mean ride quality, but the standard deviation increases, resulting in reduced robustness;② The Copula type and correlation degree significantly affect the higher-order moments (i.e., skewness and kurtosis) of the ride quality index, while having little influence on the lower-order moments;③ As the correlation degree increases, the sensitivity patterns of the three Copula types remain generally consistent. The longitudinal level amplitude factor is the dominant contributor to the variance of ride quality, but the variance contribution of the alignment amplitude factor increases with the correlation degree.

Key words: track irregularity, ride quality, Copula function, polynomial chaos expansion, ANCOVA sensitivity

CLC Number: