基于LMD能量熵和支持向量机的齿轮箱故障诊断

Gearbox fault diagnosis based on LMD energy entropy and support vector machine

针对小样本情况下齿轮箱复合多种故障特征难以提取和分类的问题,提出了基于局部均值分解(LMD)能量熵和支持向量机(SVM)的故障诊断方法。首先利用LMD方法对采集的齿轮箱振动信号进行分解,得到有限个PF分量;然后根据不同故障下齿轮箱振动信号在频域区间内分布不均的特性,分析出PF分量能量在不同频域范围离散情况,即求出LMD能量熵;最后利用SVM多故障分类器对提取出的特征展开训练和测试,进行齿轮箱故障分类。实验结果显示,即使在小样本情况下,且同时存在非单一、多种齿轮箱故障时,基于LMD能量熵和SVM方法也可以对齿轮箱故障进行特征提取和精准分类,实现齿轮箱故障诊断。

Aiming at the problem that it is difficult to extract and classify multiple fault features of gearbox in the case of small samples,a fault diagnosis method based on local mean decomposition(LMD)energy entropy and support vector machine(SVM)is proposed.Firstly,the collected gearbox vibration signal is decomposed by LMD method to obtain finite pf components;Then,according to the uneven distribution of gearbox vibration signals in frequency domain under different faults,the dispersion of PF component energy in different frequency domain is analyzed,that is,the LMD energy entropy is calculated;Finally,SVM multi fault classifier is used to train and test the extracted features for gearbox fault classification.The experimental results show that even in the case of small samples and non single and multiple gearbox faults at the same time,the feature extraction and accurate classification of gearbox faults can be realized based on LMD energy entropy and SVM.