Application of sample entropy and Fractional Fourier transform in the fault diagnosis of rolling bearings

A new method for fault feature extractions based on the sample entropy and fractional Fourier transform was presented. The core of the new method was to map the original data with poor separability into an appropriate fractional space firstly. Then the sample entropies of the transformed data before and after the fractional Fourier transformation with an appropriate order were computed and compared so as to fullfil the fault feature extraction. The results show the new method can enhance the separability of different failure modes, and discriminate the normal, inner ring fault, outer ring fault and roller fault signals distinctly. © 2017, Editorial Office of Journal of Vibration and Shock. All right reserved.