A Fault Diagnosis Method for Rolling Bearings Based on Feature Fusion of Multifractal Detrended Fluctuation Analysis and Alpha Stable Distribution

Table 5

The diagnosis results comparison of five methods.

Feature type	Diagnostic accuracy (%)
	1	2	3	4	5	6	7	Average
MFDFA features	100	80	100	90	70	90	70	85.7
ASD features	80	100	100	80	90	70	100	88.6
MFDFA + KPCA	100	100	100	90	70	100	90	92.9
ASD + KPCA	90	100	100	90	100	80	100	92.9
MFDFA + ASD + KPCA	100	100	100	90	100	100	100	98.6