小波变换对转动机械振动信号的识别具有特有的效果。但小波包变换在工程应用中,存在频率混淆、幅值失真等缺陷,采用频率补偿算法,并选取了泵转子的动不平衡、弯+动不平衡等典型故障试验数据,通过分析,验证频率补偿算法能够提取振动信号的更多细节,可以识别出故障的特征频谱,能够有效地解决频率混淆、幅值失真等问题,为小波分析的工程应用奠定了一定基础。 Wavelet transform has the peculiar effect on the identification of rotating machinery vibration signals. However, there are some defects such as frequency confusion and amplitude distortion in the application of wavelet packet transform. The frequency compensation algorithm is adopted, and the typical fault test data of pump rotor such as dynamic unbalance and bending + dynamic unbalance are selected. By analyzing and verifying the frequency The compensation algorithm can extract more details of the vibration signal, identify the characteristic spectrum of the fault, and can effectively solve the problems of frequency aliasing and amplitude distortion, which lays a foundation for the engineering application of wavelet analysis.