基于EMD分解和Levy-SSA-BP神经网络的齿轮故障诊断  

Gear Fault Diagnosis Based on EMD Decomposition and Levy-SSA-BP Neural Network

作　　者：徐婧雯 杨平[2] 阴晓俊[1] Xu Jingwen;Yang Ping;Yin Xiaojun(Shenyang Academy of Instrumentation Science Co.,Ltd.,Shenyang 110043,China;School of Information Science and Engineering,Shenyang University of Technology,Shenyang 110870,China)

机构地区：[1]沈阳仪表科学研究院有限公司,辽宁沈阳110043 [2]沈阳工业大学信息科学与工程学院,辽宁沈阳110870

出　　处：《机械传动》2024年第5期152-157,共6页Journal of Mechanical Transmission

基　　金：中国航空发动机集团应用创新项目(630010504)。

摘　　要：为解决齿轮磨损早期故障诊断问题,提出了一种基于经验模态分解(Empirical Mode Decomposition,EMD)和算法优化反向传播(Back Propagation,BP)神经网络的故障诊断方法。首先,将声发射信号进行经验模态分解,得到本征模函数(Intrinsic Mode Function,IMF);其次,计算各IMF分量与原始信号的相关系数,并对各个分量进行特征提取,构成特征矩阵;最后,将特征矩阵放入经过Levy飞行和麻雀搜索算法优化后的BP神经网络中进行识别。对比BP神经网络和麻雀搜索算法优化后的神经网络,本文提出的算法准确率更高,且对轻微磨损故障的识别能力更好,可以用于早期齿轮故障诊断。To solve the problem of early diagnosis of gear wear,this study proposes a fault diagnosis method based on empirical mode decomposition(EMD)and algorithm optimization of the back propagation(BP)neural network.Fristly,EMD is used to decompose acoustic emission signals,obtaining a series of intrinsic mode function(IMF).Secondly,calculating the correlation coefficient of each IMF with the original signal,and the feature extractions of each component are carried out to form a feature matrix.Finally,the feature matrix is put into the BP neural network optimized by Levy flight and the sparrow search algorithm for identification.Comparing the BP neural network and the neural network optimized by the sparrow search algorithm,the algorithm proposed in this study has a higher accuracy rate,and the ability to identify minor wear faults is better,which can be used in early gear fault diagnosis.

关 键 词：齿轮箱 声发射 故障诊断 BP神经网络 麻雀搜索算法 Levy飞行 经验模态分解 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程] TH132.41[自动化与计算机技术—控制科学与工程]