基于EEMD和Teager能量算子的行星齿轮箱故障特征提取研究  

Research on Fault Feature Extraction of the Planetary Gear Box Based on EEMD and Teager Energy Operator

作　　者：李伟 LI Wei(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4期109-113,119,共6页Noise and Vibration Control

基　　金：2019年度铜陵学院校级科研资助项目(2019tlxy31)。

摘　　要：针对行星齿轮箱振动信号故障特征提取困难、幅值和频率调制明显等问题,提出一种基于集合经验模态分解(EEMD)和Teager能量算子的行星齿轮箱故障特征提取方法。新方法首先利用EEMD分解样本信号得到若干本证模量函数(IMF),计算各IMF与原始信号的相关系数和欧式距离,筛选出能够表征原始信号特征的分量。其次,计算筛选后各IMF分量的Teager能量算子,将计算结果进行信号重构。最后,提取重构信号模糊熵及其对应概率密度函数的波形指标,作为基本特征参数进行故障诊断。将所提方法应用于行星齿轮箱实验数据分析,并与常用特征提取方法对比,结果表明:所提出的新方法不仅能有效区分行星齿轮箱故障类型和程度,且识别率优于所对比的方法。A fault feature extraction method for planetary gearboxes based on ensemble empirical mode decomposition(EEMD)and Teager energy operator is proposed to overcome the difficulties of fault feature extraction and obvious amplitude and frequency modulation.Firstly,several intrinsic modal functions(IMFs)are obtained by decomposing sample signals with EEMD.The correlation coefficients and Euclidean distances between the IMFs and original signals are calculated,and the components that can characterize the characteristics of original signals are screened out.Secondly,the Teager energy operators of the screened IMF components are calculated and the calculation results are reconstructed.Finally,the waveform index of the reconstructed signal and its corresponding probability density function are extracted as the basic parameters for fault diagnosis.The proposed method is applied to the analysis of experimental data of the planetary gearbox.The results show that the proposed fault feature extraction method can not only effectively distinguish the type and degree of the planetary gearbox fault,but also has better recognition rate than the common feature extraction methods.

关 键 词：故障诊断 行星齿轮箱 EEMD TEAGER能量算子 模糊熵 波形指标 

分 类 号：O422.6[理学—声学]