摘要

This paper presents some new statistical moments for the early detection of bearing failure. The third moment, skewness, of the rectified data or the fourth moment, kurtosis, of the unrectified data, has been the major statistical parameter for monitoring the condition of the rolling element bearing. A novel and unified description of normalized statistical moments is proposed. In the formula, the orders of the moments are real valued, providing more flexibility for field operation. It is also shown that the third rectified moment and the fourth moment are just two special cases of the present moment family. Then two new statistical moments are introduced. The results of the simulation and experimental tests show that the two new statistical parameters are preferred to the traditional third rectified moment and the fourth moment, respectively.