谐波小波与不同尺度下轴心轨迹分析

Harmonic Wavelets and Orbits at Different Scale

根据谐波小波良好的正交性、频域内的“盒形”谱特性和简单明了的时、频域数学表达式,利用FFT算法实现不同尺度下振动信号的谐波小波分析,它摆脱了Mallat塔形算法的约束,保证了不同尺度下不同频道信号的采样频率和数据长度与原始信号相同。因而,可以实现不同尺度下转子轴心轨迹分析,为回转机械具有非平稳、非线性响应类故障的深化识别。

According to the good featrues of Harmonic Wavelets(Hws),such as orthogonality,box-like spectrum and simple formula in time of frequency domain,this paper adopted FFT algorithm to represent vibration signal by Hws.Harmonic Wavelet transform(HWT)breaks away Mallat algorithm,and ensures every Hws at different scale to have the same sampling frequency and data length as origial.So it can be used to analysis orbits at different scale and recognize nonstationary and nonlinear faults.