基于VMD的行星齿轮箱故障特征提取新方法  

A New Method for Fault Feature Extraction of Planetary Gearboxes Based on VMD

作　　者：李伟 罗成 LI Wei;LUO Cheng(School of Mechanical Engineering,Tongling University,Tongling 244000,Anhui,China)

机构地区：[1]铜陵学院机械工程学院,安徽铜陵244000

出　　处：《噪声与振动控制》2020年第3期94-99,147,共7页Noise and Vibration Control

基　　金：安徽省大学生创新训练资助项目(201810383195)。

摘　　要：针对行星齿轮箱振动信号故障特征提取困难的问题,提出一种基于变分模态分解的行星齿轮箱故障特征提取方法。首先利用变分模态分解(VMD)算法对样本信号进行分解,得到若干本征模态函数(imf)。然后,计算各分量与样本信号之间的相关系数和欧氏距离,筛选出表征样本信号特征的有效分量,并计算其Teager能量算子,将计算结果进行重构。最后,针对多尺度模糊熵对信号局部差异不够敏感,提取重构信号的多尺度模糊熵和多尺度能量作为基本参数,进行参数融合构成新指标。将其应用于行星齿轮箱太阳轮和行星轴承故障分析,结果表明:新方法既可以区分行星齿轮箱太阳轮不同故障类型,又能有效识别行星轴承不同位置故障。另外,与现有方法对比,新方法区分效果更好。A method for fault feature extraction of planetary gearboxes based on variational mode decomposition(VMD)is proposed.Firstly,the VMD algorithm is used to decompose the sample signals,and some intrinsic mode functions(IMF)are obtained.Then,the correlation coefficient and Euclidean distance between each component and the sample signal are calculated,and the effective components that represent the characteristics of the sample signal are screened out.And the Teager energy operator is calculated.The calculation results are reconstructed.Finally,considering that the multi-scale fuzzy entropy is not sensitive enough to the local difference of signals,the multi-scale fuzzy entropy and multi-scale energy of the reconstructed signals are extracted as the basic parameters.And the parameters are combined to form a new index.The new method is applied to the fault diagnosis of the sun gear and the planet bearings of the planetary gearbox.The results show that the new method can not only distinguish the different fault types of the sun gear of the planetary gearbox,but also effectively identify the faults in different positions of the sun gear of the planetary gearbox.In addition,the new method is more effective than the existing methods,.

关 键 词：故障诊断 变分模态分解 TEAGER能量算子 多尺度模糊熵 行星齿轮箱 

分 类 号：O422.6[理学—声学]