Short-term fault prediction of mechanical rotating parts on the basis of fuzzy-grey optimising method

This paper presents a multidimensional fault characteristic parameter model for rotating parts for the purpose of mechanical fault diagnosis technology. Specifically, a fuzzy-grey optimising prediction method is presented to foretell short-term faults. The method is able to set up a prediction model with as few as four data and also can effectively handle the non-linearity of prediction data. The optimising factor τ*τ* is determined to make the prediction data closest to the original data. Taking a rolling bearing of an oil-line pump as an example to forecast its fault characteristic parameters and comparing with GM(1,1), the experiment results show that ideal effects of precision examination have been obtained, and the method proves feasible and practical.