摘要
针对滚动轴承早期信号微弱导致故障特征难以提取和故障诊断准确率不高的问题,提出了一种基于改进麻雀搜索算法-变分模态分解(ISSA-VMD)和样本熵(SE)的滚动轴承早期故障特征提取方法。首先,在轴承早期故障诊断过程中,模态分解个数和惩罚因子的选择对变分模态分解(VMD)的分解效果有着很大的影响,为消除人为选择参数的影响,将麻雀搜索算法(SSA)优化为改进麻雀搜索算法(ISSA),利用ISSA参数优化后的VMD方法对信号进行了分解;然后,计算了敏感固有模态函数(IMF)分量的样本熵,构成了特征向量;最后,将特征向量作为支持向量机(SVM)的输入,进行了滚动轴承早期故障类型的识别。研究结果表明:ISSA-VMD+样本熵特征提取模型的故障诊断准确率为98.3%,与SSA-VMD+样本熵、灰狼优化算法(GWO)-VMD+样本熵、鲸鱼优化算法(WOA)-VMD+样本熵、传统VMD+样本熵、经验模态分解(EMD)+样本熵等特征提取模型相比,故障诊断准确率分别提高了3.3%、6.6%、5%、3.3%、5%;该模型可以准确地提取故障特征,提高故障诊断准确率。
In order to solve the problems that the early signal of rolling bearing was weak which made it difficult to extract fault features,and the accuracy of fault diagnosis was low,a feature extraction method based on improved sparrow search algorithm-variational modal decomposition(ISSA-VMD)and sample entropy(SE)was proposed.First of all,in the process of bearing early fault diagnosis,the choice of the number of modal decomposition and penalty factor had a great influence on the decomposition effect of VMD;in order to eliminate the influence of artificial selection of parameters,sparrow search algorithm(SSA)was optimized to improved sparrow search algorithm(ISSA).The signals were decomposed by the VMD after the parameter optimized by ISSA.Then,the sample entropy of the sensitive intrinsic mode function(IMF)components was calculated to form the eigenvector.Finally,the feature vector was used as the input of support vector machine(SVM)for fault type identification of early fault of rolling bearing.The experimental results show that the fault diagnosis accuracy of ISSA-VMD+sample entropy feature extraction model is 98.3%.Comparing with SSA-VMD+sample entropy,grey wolf optimizer(GWO)-VMD+sample entropy,whale optimization algorithm(WOA)-VMD+sample entropy,traditional VMD+sample entropy and empirical mode decomposition(EMD)+sample entropy feature extraction models,the fault diagnosis accuracy of the model is respectively improved by 3.3%,6.6%,5%,3.3%and 5%.The model can accurately extract fault features and improve the fault diagnosis accuracy.
作者
刘玉明
刘自然
王鹏博
LIU Yuming;LIU Ziran;WANG Pengbo(School of Mechanical and Electrical Engineering,Henan University of Technology,Zhengzhou 450001,China)
出处
《机电工程》
CAS
北大核心
2023年第9期1426-1432,共7页
Journal of Mechanical & Electrical Engineering
基金
河南省自然科学基金资助项目(182300410234)。
关键词
轴承早期故障
故障特征提取
改进麻雀搜索算法-变分模态分解
样本熵
支持向量机
经验模态分解
early fault of rolling bearing
fault feature extraction
improved sparrow search algorithm-variational modal decomposition(ISSA-VMD)
sample entropy(SE)
support vector machine(SVM)
empirical mode decomposition(EMD)
