Train Fault Diagnosis Based on Binary PELCD Sample Entropy with Full Vector Fusion
The service state of high-speed train for a long time operation will cause the deterioration of the performance of the key components of its bogie,and the breakdown of the safety events will cause serious economic losses and even casualties.In this paper,considering the characteristics of high-speed train vibration signals,the partial integrated local feature scale decomposition method is extended to the field of binary signal processing.At the same time,based on the theory of full vector spectrum,the information fusion of the same order component signals is carried out to obtain more complete data features,and the sample entropy features of the fused data are extracted to obtain the train fault features.The Grey Wolf optimization algorithm is used to optimize the parameters of support vector machine.Finally,the fault recognition rates under single fault condition,compound fault condition and component performance degradation are compared by experiments to verify the effectiveness and superiority of the proposed method.
fault diagnosisbinary partial ensemble local characteristic scale decompositionfull vector theorygrey wolf optimization algorithmsupport vector machine
国家科技期刊平台
NSTL
万方数据
