摘要
针对滚动轴承故障诊断存在的故障程度难以区分、早期故障不易发现、故障诊断精度低等问题,这里提出了一种基于小波包变换(Wavelet Packet Transform,WPT)、主成分分析(Principal Component Analysis,PCA)与支持向量机(Support Vector Machine,SVM)相结合的滚动轴承故障程度诊断方法。该方法首先对原始信号进行小波包分解,然后对分解后的信号进行重构,计算重构信号能量作为特征值;随后,运用主成分分析对特征向量进行降维,将降维后的特征输入支持向量机,完成故障模型的训练与测试。这里主要分析了累计贡献率、小波包分解层数、母小波类型对故障诊断成功率的影响。实验结果表明此方法可以有效地识别不同故障位置的故障程度。
Difficulties are encountered when fault degree is evaluated,initial failure is distinguished and fault is diagnosed acurately for rolling bearing.we propose a fault degree diagnosis method for rolling bearing by making use of wavelet packet transform(WPT),principal component analysis(PCA)and support vector machine(SVM).We decompose the original signal by wavelet packet first.After that,we reconstruct the decomposed signal and calculate its’energy as the eigenvalue.We utilize the principal component analysis to reduce the feature vector dimensions.Finally,we put the dimensioned feature into the support vector machine to establish the fault diagnosis model,which is improved by training and testing.We evaluate the effect of cumulative contribution rate,wavelet packet decomposition layer and mother wavelet type on the success rate of fault diagnosis.Experimental results show that this method can effectively identify the fault degree of different fault locations.
作者
王帅星
黄茜
王晓笋
巫世晶
WANG Shuai-xing;HUANG Xi;WANG Xiao-sun;WU Shi-jing(School of Power and Mechanical Engineering,Wuhan University,Hubei Wuhan 430072,China)
出处
《机械设计与制造》
北大核心
2022年第4期5-9,共5页
Machinery Design & Manufacture
关键词
故障程度
小波包变换
主成分分析
支持向量机
Fault Degree
Wavelet Packet Transform
Principal Component Analysis
Support Vector Machine
