Rolling Bearing Diagnosis Based on Adaptive Probabilistic PCA and the Enhanced Morphological Filter

Table 2

Evaluation indicator.

Number	Name	Definition
F1	Feature energy factor
F2	Permutation entropy
F3	Envelope spectrum sparsity
F4	Amplitudes at the 3rd harmonics of the bearing fault frequency

Note. Y (f) is the Hilbert envelope spectrum of y (t), n is the length of the spectral, M is the number of fault frequencies , N is the length of the time signal, m is the embedding dimension, π is a permutation, τ is the time lag, A represents the amplitude of the signal, and f_o represents the fault frequency of the signal.