Application of information entropy in fault diagnosis of high speed train wheel set

Abstract

Keywords: information entropy; high speed train; wheel set; fault diagnosis Abstract. Because of the characteristics of the high speed train and its operating environment, the high speed train is in an unstable state. As a very important part of the high speed train, the train wheel set is also influenced by other external factors which are very complicated. In this paper, the vibration characteristics of a number of key components of the train wheel assembly are studied by using the theory of information entropy, and the diagnosis of the faults of the wheel is completed by using the basic concepts and rules of D-S evidence theory. In the end, the paper takes the wheel set failure of CRH3 high speed train as an example. Through theoretical research and experimental results, this paper also shows that the information entropy feature extraction and fusion method can realize the quantitative characterization and system operation state evaluation of multi-source and multi-level signal characteristics. It is also proved that the method can effectively reduce the uncertainty of fault diagnosis and improve the accuracy of fault diagnosis.