Table 1.
Number of combinations based on the number of features extracted.
Fig 1.
Experimental test rig.
Table 2.
Vibration data distribution.
Table 3.
Statistical features.
Fig 2.
(a) Skewness factor, (b) kurtosis factor, (c) crest factor, (d) shape factor, (e) impulse factor and (f) margin factor of all bearing conditions.
Fig 3.
The proposed feature selection algorithm (features A, B, C, D, E and F represent skewness factor, kurtosis factor, crest factor, shape factor, impulse factor and margin factor, respectively).
Table 4.
Training accuracy for the key combination of features (features A, B, C, D, E and F represent skewness factor, kurtosis factor, crest factor, shape factor, impulse factor and margin factor, respectively).
Fig 4.
Comparison of the testing accuracy (average of 10-fold cross-validation).
Table 5.
Cyclical assessment for the proposed WFS by 10-fold cross-validation.
Table 6.
Cyclical assessment for the MRMD by 10-fold cross-validation.