基于小波包-包络分析的故障特征提取方法  

Approach to Extraction of Fault Features Based on Wavelet Packet and Envelope Analysis

作　　者：李辉[1] 宋智勇[1] 孙丰瑞[1] 

机构地区：[1]海军工程大学动力工程学院307教研室,武汉430033

出　　处：《振动．测试与诊断》2003年第4期291-294,共4页Journal of Vibration,Measurement & Diagnosis

摘　　要：针对齿轮、滚动轴承等的早期损伤类故障 ,提出将小波包分解作为包络分析的前置处理手段以提取振动信号的故障信息特征。在简述小波包基本原理的基础上 ,通过仿真信号 ,对振动信号的具体处理过程进行分析 ,并对可能遇到的问题 ,提出处理办法 ,然后应用于诊断实例。结果表明 ,该方法能够自由确定振动信号的频带划分并全面获得各频带内隐含的故障特征 ,且易于实现 。In order to diagnose the incipient faults of ball and roller bearings,gears and other machine elements by their vibration signals,an approach to the extraction of fault features from their vibrations using orthogonal decomposition of wavelet packets and envelope analysis is proposed,where the decomposition is used as the preprocessing of the envelope analysis.After a brief introduction to the foundamental principle of wavelet packets,an analysis of the concrete processing of a simulated vibration signal is made,from which several methods for solving the problems possibly encountered in the caurse of the processing are suggested.The results of a practical diagnosis indicate that in the approach the analysis bands of frequency of a signal can be divided according to your needs,and the fault features contained in each band of the signal can be completely exerated.Moreover,the approach developed in this work is easy to realize,and so has certain value in engineering applications.

关 键 词：小波包-包络分析 振动信号 故障诊断 特征提取 机械振动 

分 类 号：TH17[机械工程—机械制造及自动化] TN911.7[电子电信—通信与信息系统]