基于小波包和EM聚类的采煤机齿轮故障诊断  

Fault Diagnosis of Shearer Gear Based on Wavelet Packet and EM Clustering

作　　者：张志刚[1] 陈巧云 马俊[1] Zhang Zhigang;Chen Qiaoyun;Ma Jun(Jiaozuo University,Jiaozuo 454000,China)

机构地区：[1]焦作大学,河南焦作454000

出　　处：《煤矿机械》2020年第9期183-186,共4页Coal Mine Machinery

基　　金：河南省科技厅软科学项目(182400410073)。

摘　　要：针对采煤机齿轮故障振动信号难以准确获取故障特征的问题,提出了一种利用小波包结合高斯混合EM聚类的齿轮故障诊断方法。首先对故障信号进行小波包分解和重构,得到其高频率尺度下的能量值,然后以此作为故障样本属性,结合高斯混合EM聚类算法建立故障模型数据库,最后将实时信号与故障库对比进行分类诊断。实验仿真结果表明,该方法对齿轮的几种典型故障表现出了良好的诊断能力,且可以实现采煤机不停机在线诊断,对提高采煤机故障诊断智能化水平具有较高的参考价值。In order to solve the problem that it is difficult to get the fault features of gear fault vibration signals of shearer accurately,a fault diagnosis method based on wavelet packet and Gaussian mixture EM clustering was proposed.Firstly,the fault signal was decomposed and reconstructed by wavelet packet to get its energy value in high frequency scale,then it was used as the fault sample attribute and the fault model database was established by combining the Gauss mixture EM clustering algorithm.Finally,the real-time signal was compared with the fault database to realize classification and diagnosis.The experiment simulation results show that this method has a good diagnostic ability for several typical faults of gear and can realize the on-line diagnosis of shearer without shutdown,which has a high reference value for improving the intelligent level of shearer fault diagnosis.

关 键 词：齿轮 故障诊断 小波包 高斯混合模型 EM聚类 

分 类 号：TP206.3[自动化与计算机技术—检测技术与自动化装置] TD421.6[自动化与计算机技术—控制科学与工程]