F. Honarvar et Hr. Martin, NEW STATISTICAL MOMENTS FOR DIAGNOSTICS OF ROLLING ELEMENT BEARINGS, Journal of manufacturing science and engineering, 119(3), 1997, pp. 425-432
Statistical moment analysis has proven to be a very effective techniqu
e for diagnosis of rolling element bearings. The fourth normalized cen
tral statistical moment, kurtosis, has been the major parameter in thi
s method. In this paper it will be shown that the third normalized sta
tistical moment can be as effective as kurtosis if the data is initial
ly rectified. The advantage of this moment over the traditional kurtos
is value is its lesser susceptibility to spurious vibrations, which is
considered to be one of the shortcomings of higher statistical moment
s including kurtosis. The sensitivity of this moment to changes of loa
d and speed is also less than kurtosis. The proposed method can also b
e applied to higher odd statistical moments.