旋转机械轴心轨迹信号的复数小波分析  

Rotating Machinery Orbit Analysis Using Complex Wavelet Transform

作　　者：申弢[1] 黄树红[1] 韩守木[1] 刘德昌[1] 

机构地区：[1]华中理工大学动力工程系,武汉430074

出　　处：《振动．测试与诊断》2000年第4期264-268,共5页Journal of Vibration,Measurement & Diagnosis

摘　　要：提出了一种利用复数小波变换分析转子振动轴心轨迹信号的方法。通过分析小波尺度与振动倍频间的关系 ,使得观察某一尺度上的小波幅值就相当于考察一定频率范围的周期谱图。对典型非线性故障轴心轨迹信号的分析揭示出 ,摩擦故障时高频振动在时 -频域上呈现出非均匀间歇现象 ,而轴裂纹时的高频振动是均匀间歇地出现 ,且周期与旋转基频相同。因此该方法可以直观地识别故障的细节性特征 ,有望形成一种新的、有效的监测诊断技术。A method is introduced for analyzing the axis orbit of a rotor by using the complex wavelet transform of a pair of vibration signals from two 90° positioned transducers atteched to the rotor. By the analysis of the relationship between scale parameter and vibration freguency, it can be seen that an examination of the amplitudes of the wavelet at a certain scale is equivalent to that of the periodogram at a certain range of frequency. The analysis of the orbits produced by two typical nonlinear faults, i.e. the friction and the crack, shows that the orbits contain different features for different faults. The high frequency components of the orbit due to the friction are unevenly spaced along the time axis in the time frequency domain, whereas those due to the crack are even spaced, and its period is equal to the rotating period of the rotor. It is expected that the method proposed in the paper can be developed into a new and effective technology for fault monitoring and diagnosis due to its capability for the identification of detailed fault features.

关 键 词：小波分析 轴心轨迹 特征提取 故障诊断 旋转机械 

分 类 号：TH133[机械工程—机械制造及自动化] TN911.6[电子电信—通信与信息系统]