基于均值Gnome熵和神经网络的轴承故障诊断  

Bearing Fault Diagnosis Method Based on Average Gnome Entropy and Neural Network

作　　者：朱熹 吕勇[1,2] 袁锐 吴利锋 ZHU Xi;LV Yong;YUAN Rui;WU Li-feng(Key Laboratory of Metallurgical Equipment and Control Technology,Ministry of Education,Wuhan University of Science and Technology,Wuhan 430081,China;Hubei Key Laboratory of Mechanical Transmission and Manufacturing Engineering,Wuhan University of Science and Technology,Wuhan 430081,China)

机构地区：[1]武汉科技大学冶金装备及其控制教育部重点实验室,武汉430081 [2]武汉科技大学机械传动与制造工程湖北省重点实验室,武汉430081

出　　处：《组合机床与自动化加工技术》2022年第8期67-70,74,共5页Modular Machine Tool & Automatic Manufacturing Technique

基　　金：国家自然科学基金(51875416);湖北省自然科学基金(2020CFA033)。

摘　　要：针对Gnome熵(gnome entropy, GEn)作为特征值的轴承故障分类精度不高的问题,提出均值Gnome熵(average gnome entropy, AGEn)的方法,有效解决GEn的负值和高维情况下熵值衰减等问题,并提高轴承故障分类的准确度。GEn的出现有效解决了传统的熵对滚动轴承故障诊断都有着参数选择的问题,AGEn继承了GEn这一特性,并能在多维条件下保持熵值稳定。利用AGEn对振动信号的动态特征进行量化,得到特征集后,将特征集输入BP神经网络模型,可以对内圈、外圈和滚动体故障的滚动轴承进行故障识别。通过轴承故障诊断试验台轴承实验信号的成功应用,验证了该方法的有效性。Aiming at the problem that the bearing fault classification accuracy of gnome entropy(GEn)as the characteristic value is not high,the average gnome entropy(AGEn)method is proposed to effectively solve the negative value of GEn and the entropy attenuation in the case of high dimensionality.And improve the accuracy of bearing fault classification.The emergence of GEn effectively solves the traditional entropy problem of parameter selection for rolling bearing fault diagnosis.AGEn inherits the characteristics of GEn and can keep the entropy stable under multi-dimensional conditions.Use AGEn to quantify the dynamic characteristics of the vibration signal.After the feature set is obtained,the feature set is input into the BP neural network model,which can identify the failure of the inner ring,outer ring and rolling element of the rolling bearing.The effectiveness of the method is verified by the successful application of bearing experimental signals on the bearing fault diagnosis test bench.

关 键 词：滚动轴承 Gnome熵 故障诊断 故障分类 BP神经网络 

分 类 号：TH133.3[机械工程—机械制造及自动化] TG659[金属学及工艺—金属切削加工及机床]