基于广义精细复合多尺度散布熵的机车轮对轴承智能诊断方法  

An Intelligent Diagnosis Method of Locomotive Wheelset Bearings Based on Generalized Refined Composite Multiscale Dispersion Entropy

作　　者：陆毅 LU Yi(Key Laboratory of Conveyance and Equipment of the Ministry of Education of China,East China Jiaotong University,Nanchang 330013,China)

机构地区：[1]华东交通大学载运工具与装备教育部重点实验室,南昌330013

出　　处：《机械设计与研究》2022年第4期119-124,137,共7页Machine Design And Research

基　　金：国家自然科学基金(51665013);江西省自然科学基金(20212BAB204007)资助项目;江西省教育厅科学技术研究项目(GJJ200616)。

摘　　要：针对机车轮对轴承单一与复合故障在内的不同健康状况的识别问题,引入一种基于精细复合多尺度散布熵改进的非线性动力学分析方法—广义精细复合多尺度散布熵。该方法解决了熵值波动大、计算不准确的问题,在计算过程中能获取更多有效信息。将之与灰狼算法优化的支持向量机结合,提出了一种机车轮对轴承智能诊断方法。为验证其效果,本文采用南昌铁路局实际机车轮对轴承数据进行实验,得到结论:所提方法识别准确率明显高于多尺度散布熵与精细复合多尺度散布熵的方法,而且能精确地识别复合故障以及不同程度故障,具有较大实际意义。In order to identify different health conditions including single and compound faults of locomotive wheelset bearings,a nonlinear dynamic analysis method based on refined composite multiscale dispersion entropy is introduced.This method solves the problems of high entropy fluctuation and inaccurate calculation,and can obtain more effective information in the process of calculation.Combined with support vector machine optimized by gray Wolf algorithm,an intelligent diagnosis method of locomotive wheelset bearing is proposed.In order to verify its effectiveness,this paper uses the actual locomotive wheelset bearing data of Nanchang Railway Bureau to carry out experiments.It is concluded that the recognition accuracy of the proposed method is significantly higher than that of the multiscale dispersion entropy method and the refined composite multiscale dispersion entropy method.The proposed method can accurately identify complex faults and different degrees of faults,which has practical significance.

关 键 词：轮对轴承 广义精细复合多尺度散布熵 灰狼算法 支持向量机 故障诊断 

分 类 号：TH13[机械工程—机械制造及自动化] TH17