时频分析方法在冲击故障早期诊断中的应用研究

Applications of Time Frequency Analysis Based on Wavelet Packet to Incipient Impulse Faults Diagnosis

冲击故障的振动响应信号是时变的 ,用传统的诊断方法一般难以进行早期发现和诊断。本文基于小波包对时变信号的多分辨分解和重构能力以及 CONE时频分布对时信号的优良描述特性 ,发展了基于小波包的时频分析方法。经试验和现场监测数据的分析和验证 ,说明该方法能够从很强的基础振动信号中提取清晰直观的冲击故障特征 。

Vibration response signal of impulse fault is generally time variant. It is difficult to early detect and diagnose by means of traditional methods. Basing on the capability of decompsition and reconstruction of wavelet packet to time variant and the better characteristics of Cone time frequency distribution to time variant signal, a time frequency analysis method based on wavelet packet is developed in this paper. It is shown by experiments that the clear cut and visual feature of impulsive signal can be successfully extracted from strong basic vibration, and the early impulsive fault can be effectively diagnosed.