Multi-body dynamics-based sensitivity analysis for a railway vehicle

Guofu Ding, Yong He, Yisheng Zou, Rong Li, Xiaojia Ma, Sheng-feng Qin

Research output: Contribution to journalArticlepeer-review

6 Citations (Scopus)


A railway vehicle is a complicated mechatronic system. Its mechanical structure and connections are key influences on its performance. When a railway vehicle runs at very high speeds its dynamic performance becomes of considerable interest. Therefore, the identification of the design factors that significantly impact on the vehicle's running behavior, such as safety, stability, comfort and reliability, is a key step towards vehicle design optimization. The optimal design of a railway vehicle is difficult to perform using existing methods due to the complexity inherent in rail/track interactions. Thus, performing a sensitivity analysis of a railway vehicle can be seen as a bridge between performance analysis and an optimal design. This paper presents a sensitivity model that can be used to analyze the performance parameters of a railway vehicle in terms of variation of its design parameters. First, the railway vehicle is described as a multi-body system that can be broken down into its component parts and then, the dynamic behavior of the system is investigated. To simplify the model, a spring/damper connection is used in the classic Hertz nonlinear elastic contact model to represent the force in the normal direction at the wheel/rail contact point. Finally, based on the adjoint variable method, the equations that constitute the sensitivity model of the railway vehicle are derived and analysis software is developed. A case study shows the reliability of the proposed approach.
Original languageEnglish
Pages (from-to)518-529
JournalProceedings of the Institution of Mechanical Engineers Part F: Journal of Rail and Rapid Transit
Issue number5
Publication statusPublished - Jul 2015


Dive into the research topics of 'Multi-body dynamics-based sensitivity analysis for a railway vehicle'. Together they form a unique fingerprint.

Cite this