基于EMD近似熵和TWSVM的齿轮箱故障诊断  

Gearbox Fault Diagnosis Based on EMD-approximate Entropy and TWSVM

作　　者：张曹[1] 陈珺[1] 刘飞[1] 

机构地区：[1]江南大学物联网工程学院,江苏无锡214122

出　　处：《煤矿机械》2017年第4期142-145,共4页Coal Mine Machinery

基　　金：国家自然科学基金项目(NSFC61403167)

摘　　要：针对齿轮箱振动信号的非平稳和非线性特征,给出了一种基于经验模态分解(EMD)近似熵和双子支持向量机(TWSVM)的齿轮箱故障诊断方法。对不同类型的齿轮信号进行EMD分解,得到若干个具有不同时间尺度的本征模函数(IMF)分量,再对IMF分量使用近似熵求解,得到一组特征向量,最后将其输入到TWSVM分类器中进行故障诊断。仿真实验表明,该方法能有效地提取故障特征,不同的故障类型表现出不同的故障信息;与传统SVM相比,TWSVM的计算时间更短,分类效果更好。A gearbox fault diagnosis method based on empirical mode decomposition （EMD） approximate entropy and twin support vector machine （TWSVM） is presented for the gear box vibration signals with the non-stationary and nonlinear characteristics. Firstly, the different types of gearbox signal are decomposed into several IMF components with different time scales by EMD. Then the approximate entropies of the IMF components are calculated as the eigenvector matrix. Finally,the eigenvector matrix is input into TWSVM classifier to identify the type of faults. The simulation experiment results show that this method can effectively extract the fault features, and the different fault types show different fault information. Compared with the traditional SVM, TWSVM has a shorter calculation time and a better classification performance.

关 键 词：经验模态分解 近似熵 双子支持向量机 齿轮箱故障诊断 

分 类 号：TH132.46[机械工程—机械制造及自动化] TP277.3[自动化与计算机技术—检测技术与自动化装置]