基于CEEMDAN-FastICA的滚动轴承故障特征提取  

Fault Feature Extraction of Rolling Bearing Based on CEEMDAN-FastICA

作　　者：刘兆亮 颜丙生[1] 刘春波[1] 聂士杰 宋宇宙 LIU Zhao-liang;YAN Bing-sheng;LIU Chun-bo;NIE Shi-jie;SONG Yu-zhou(School of Mechanical and Electrical Engineering,Henan University of Technology,Zhengzhou 450001,China)

机构地区：[1]河南工业大学机电工程学院,郑州450001

出　　处：《组合机床与自动化加工技术》2021年第3期61-65,共5页Modular Machine Tool & Automatic Manufacturing Technique

基　　金：国家自然科学基金项目资助(U1604134);河南省科技攻关项目(212102210327)。

摘　　要：针对轴承早期故障信号微弱、故障特征难以提取的问题,提出一种将完备集合经验模态分解(CEEMDAN)与快速独立分量分析(FastICA)相结合的故障特征提取方法。该方法首先利用CEEMDAN将轴承故障信号进行分解,得到一系列模态分量(IMF);然后依据峭度准则选取相应分量进行重构,引入虚拟噪声通道;最后利用FastICA对重构信号进行解混去噪,分离出源信号的最佳估计信号后进行包络谱分析进而提取故障特征频率。该方法通过LabVIEW软件平台进行编程实现。仿真信号和轴承故障实验信号的研究结果均表明该方法可明显降低噪声和调制成分干扰,突出故障特征频率成分。In order to solve the problem of weak early bearing fault signal and difficult to extract fault features,a fault feature extraction method is proposed,which combines the Complete Ensemble Empirical Mode Decomposition with Adaptive Noise(CEEMDAN)with Fast Independent Component Analysis(FastICA).Firstly,the bearing fault signal is decomposed by CEEMDAN to obtain a series of modal components(IMF);Then,the corresponding components are selected for reconstruction according to kurtosis criterion,and the virtual noise channel is introduced;Finally,the reconstructed signal is de-mixed and de-noising by FastICA,the best estimated signal of the source signal is separated,and then the envelope spectrum analysis is carried out to extract the fault characteristic frequency.This method is realized by LabVIEW software platform.The research results of simulation signal and bearing fault experiment signal show that this method can reduce noise and modulation component interference,and highlight fault characteristic frequency component.

关 键 词：完备集合经验模态分解 快速独立分量分析 故障特征提取 

分 类 号：TH162[机械工程—机械制造及自动化] TG506[金属学及工艺—金属切削加工及机床]