A method of analysis for grinding wheel's unbalanced signal based on wavelet packet

There are a large quantity of noises and the broadband signals with other period composition in the vibration signal of grinding wheel, so it is important to extract signal with the same frequency of the grinding wheel in order to keep precise balance of the signal. The precision of the traditional methods where a band-pass filter has been 'used is not high, and the parameters must be re-designed to the different grinding wheel machines. By the decomposition of wavelet packet to the vibration signal of the grinding wheel, the good frequency characteristic can be obtained, which can provide an effective way to analyze the unbalanced signal with vibration around the unsteady state. An experiment is given in this paper, and it is concluded that the proposed method can extract the signal more precisely than the traditional methods.