基于SORT映射的IRCMFDE在旋转机械故障诊断中的应用  

IRCMFDE based on SORT mapping in fault diagnosis of rotating machinery

作　　者：王潞红 邹平吉[3] WANG Luhong;ZOU Pingji(School of Mechanical and Electrical Engineering,China University of Mining and Technology,Xuzhou 221116,China;Department of Mechatronics Engineering,Changzhi Vocational and Technical College,Changzhi 046000,China;Department of Personnel,Lanzhou Vocational and Technical College,Lanzhou 730070,China)

机构地区：[1]中国矿业大学机电工程学院,江苏徐州221116 [2]长治职业技术学院机械电子工程系,山西长治046000 [3]兰州职业技术学院人事处,甘肃兰州730070

出　　处：《机电工程》2024年第1期11-21,共11页Journal of Mechanical & Electrical Engineering

基　　金：山西省基础研究计划项目(20210302124485)。

摘　　要：针对旋转机械振动信号的强非线性和非平稳性,导致故障特征提取困难的问题,提出了一种基于SORT映射的改进精细复合多尺度波动散布熵(IRCMFDE)和蝙蝠算法优化的相关向量机(BA-RVM)的旋转机械故障诊断方法。首先,利用SORT映射函数替换了精细复合多尺度波动散布熵(RCMFDE)方法的正态累积分布函数,同时对RCMFDE方法的粗粒化方式进行了改进,提出了基于SORT映射的IRCMFDE方法;随后,利用IRCMFDE方法提取了旋转机械振动信号的故障特征,构造了故障特征集;最后,采用BA-RVM分类器对旋转机械的故障类型进行了智能化的识别和分类;将基于IRCMFDE和BA-RVM的故障诊断方法应用于滚动轴承、离心泵和齿轮箱的实验数据分析,并将其与现有故障诊断方法进行了对比分析。研究结果表明:基于IRCMFDE和BA-RVM的故障诊断方法能够有效地识别旋转机械的故障状态,识别准确率分别达到了100%、98%和99%,相比基于RCMFDE、精细复合多尺度熵、精细复合多尺度模糊熵、精细复合多尺度排列熵和精细复合多尺度散布熵的故障特征提取方法,该故障诊断方法的效率和平均识别准确率均优于对比方法,其更适合应用于旋转机械的在线实时故障监测。Aiming at the problem of difficult fault feature extraction caused by the strong nonlinearity and non-stationary nature of vibration signals of rotating machinery,a fault diagnosis method for rotating machinery based on SORT mapping based improved refined composite multiscale fluctuation dispersion entropy(IRCMFDE)method and bat algorithm optimized relevant vector machine(BA-RVM)was proposed.Firstly,the SORT mapping function was used to replace the normal cumulative distribution function of refined composite multiscale fluctuation dispersion entropy(RCMFDE),and the coarse-grained process of RCMFDE method was improved,thus an IRCMFDE method based on SORT mapping was proposed.Then,the fault features of rotating machinery vibration signals were extracted by IRCMFDE method,and the fault feature set was constructed.Finally,BA-RVM classifier was used to intelligently identify and classify the fault types of rotating machinery.The fault diagnosis method based on IRCMFDE and BA-RVM was applied to the experimental data analysis of rolling bearing,centrifugal pump and gear box,and it was compared with the existing fault diagnosis methods.The research results show that the fault diagnosis method based on IRCMFDE and BA-RVM can effectively identify the fault status of rotating machinery,with recognition accuracy rates of 100%,98%,and 99%respectively,and compared to fault feature extraction methods based on RCMFDE,refined composite multiscale sample entropy,refined composite multiscale fuzzy entropy,refined composite multiscale permutation entropy,and refined composite multiscale dispersion entropy,the efficiency and average recognition accuracy of this method are better than those of the comparison method,and it is more suitable for online real-time fault monitoring of rotating machinery.

关 键 词：改进精细复合多尺度波动散布熵 SORT映射 蝙蝠算法优化的相关向量机 旋转机械 故障分类识别 

分 类 号：TH133.3[机械工程—机械制造及自动化]