Data-Based Health State Analysis for the Axle of High Speed Train

Abstract

Taking into account that the present popular methods, such as the judgement of the axle failure based on temperature threshold, and the early warning of axle based on real-time temperature analysis, cannot analyze the changing of performance trends, a health state analysis method for the axle of high-speed train based on long-term temperature monitoring data is proposed in this paper, which including the following main steps: (1) Preprocessing of the original data to correct the singular zero value and complement the missing values, (2) Smoothing of the processed data in order to automatically extract the beginning and end points of every temperature rising stage of axles, (3) Establishment of the calculation method of temperature rising rate, and evaluating the health sate of axles based on the temperature rising rate. Finally, the proposed method is validated based on the data from a test line, the results demonstrate the effectiveness and practicability of the method.

DOI: 10.1109/CIS.2015.115

1 Figure or Table

Cite this paper

@article{Xie2015DataBasedHS, title={Data-Based Health State Analysis for the Axle of High Speed Train}, author={Guo Xie and Minying Ye and Xinhong Hei and Jinwei Zhao and Fucai Qian}, journal={2015 11th International Conference on Computational Intelligence and Security (CIS)}, year={2015}, pages={454-457} }