Simplified Machine Diagnosis Techniques under Impact Vibration Using Higher Order Cumulants with the Comparison of Three Cases

  •  Kazuhiro Takeyasu    


Among many amplitude parameters, Kurtosis (4-th normalized moment of probability density function) is recognized to be the sensitive good parameter for machine diagnosis. On the other hand, a new method of machine diagnosis can be considered utilizing the higher order cumulants which have the characteristics that cumulants more than 3rd order are 0 under Gaussian distribution. Cumulants are stated in combination with the same order moment and the moments under that order. Simple calculation method is required on the maintenance site. Furthermore, the absolute deterioration factor such as Bicoherence would be much easier to handle because it takes the value of 1.0 under the normal condition and tends to be 0 when damages increase. In this paper, nth normalized cumulant is considered so as to intensify the sensitivity of diagnosis. Also, the simplified calculation method for this new parameter by impact vibration is introduced. Furthermore, the absolute deterioration factor is introduced. Three cases in which the rolling elements number is nine, twelve and sixteen are examined and compared. The new calculation method is examined whether it is a sensitive good parameter or not. Compared with the results obtained so far, the new method shows fairly good results.

This work is licensed under a Creative Commons Attribution 4.0 License.