Motor bearing fault diagnosis based on wavelet packet decomposition of instantaneous power

This paper gives a new method to resolve the problem of extracting fault characteristics in electromotor. It is proved that the fault information is plentiful in instantaneous power by theory illation, and the fundamental component transforms to the DC component which can be filtrated easily; analyze the signal via wavelet packet decomposition, the root mean square of the node coefficients and it´s change rate used as the symptom of bearing fault is calculated. Simulation result accords with the theory illation, which indicates that this method can be used in bearing fault diagnosis.