基于自相关降噪和局域均值分解的轨道车辆轴箱轴承故障诊断方法  

Fault Diagnosis Method of Axle Box Bearing of High-Speed Train Based on Autocorrelation Noise Reduction and LMD

作　　者：宋冬利[1] 董俭雄 郑则君 江炘坤 SONG Dongli;DONG Jianxiong;ZHENG Zejun;JIANG Xinkun(State Key Laboratory of Traction Power,Southwest Jiaotong University,Chengdu 610031,China)

机构地区：[1]西南交通大学牵引动力国家重点实验室,成都610031

出　　处：《实验室研究与探索》2022年第12期63-67,共5页Research and Exploration In Laboratory

基　　金：国家重点研发计划(2019YFB1405401);西南交通大学学科交叉基础研究项目(2682021ZTPY099);湖南创新省份建设专项项目(2021GK4014)。

摘　　要：以轨道车辆轴箱轴承为研究对象,建立了基于振动监测的故障诊断模型,提出了基于自相关降噪和局域均值分解的轨道车辆轴箱轴承故障特征提取方法,用于轨道车辆轴箱轴承故障分类辨识。为分析该方法在轴箱轴承故障诊断中的有效性,参考轨道车辆轴箱轴承实际服役工况搭建了实验平台。进一步利用加速度传感器采集信号,将不同故障模式下的振动数据按照所构建方法的流程进行故障特征频率提取。实验结果表明,原始信号中的随机干扰噪声得到有效抑制,故障特征3倍频及以上频率成分被成功提取出来,验证了所构建方法用于轨道车辆轴箱轴承故障诊断的可行性。A fault diagnosis model based on vibration monitoring is proposed,and a fault feature extraction method based on autocorrelation noise reduction and local mean decomposition is established for fault classification and identification of rail vehicle axle box bearings.In order to analyze the effectiveness of this method in axle box bearing fault diagnosis,an experimental platform is built by considering the actual service conditions of the axle box bearing of rail vehicles.The vibration data under different fault modes are further extracted according to the flow of the method.The analysis of the experimental results shows that the random interference noise in the original signal is effectively suppressed,and the frequency components of 3 times and above of the fault feature are extracted,which verifies the feasibility of the proposed method for the fault diagnosis of the axle box bearing of rail vehicles.

关 键 词：轴箱轴承 故障诊断 局域均值分解 自相关降噪 

分 类 号：TH17[机械工程—机械制造及自动化]