转子突加不平衡声信号小波分解的特征分析

he sudden unbalance signal of rotor shows clear singularity which contains great fault characteristics. The alterable focus character of the wavelet analysis theory can process the signal locally in time and frequency at the same time. The acceleration signal and noise signal of the sudden unbalance is analyzed by this method to find the position and degree of its singularity. The characteristics of the acceleration signal and noise signal of the sudden unbalance decomposed by wavelet-packet are compared and studied. The results show that the noise signal has its special advantage in the fault diagnosis of rotating-machines. The wavelet-packet decomposition based on noise signal and acceleration signal can detect the characteristics of the sudden unbalance and the moment when the fault occurred.