Citation

Li C, Luo S, Cole C, Spiryagin M. Veh. Syst. Dyn. 2017; 55(7): 1045-1070.

Copyright

(Copyright © 2017, Informa - Taylor and Francis Group)

DOI

10.1080/00423114.2017.1296963

PMID

unavailable

Abstract

Health monitoring systems with low-cost sensor networks and smart algorithms are always needed in both passenger trains and heavy haul trains due to the increasing need for reliability and safety in the railway industry. This paper focuses on an overview of existing approaches applied for railway vehicle on-board health monitoring systems. The approaches applied in the data measurement systems and the data analysis systems in railway on-board health monitoring systems are presented in this paper, including methodologies, theories and applications. The pros and cons of the various approaches are analysed to determine appropriate benchmarks for an effective and efficient railway vehicle on-board health monitoring system. According to this review, inertial sensors are the most popular due to their advantages of low cost, robustness and low power consumption. Linearisation methods are required for the model-based methods which would inevitably introduce error to the estimation results, and it is time-consuming to include all possible conditions in the pre-built database required for signal-based methods. Based on this review, future development trends in the design of new low-cost health monitoring systems for railway vehicles are discussed.

Language: en

Keywords

classification; fault detection and isolation; Health monitoring; model-based; on-board; signal-based