An Integrated Health Condition Detection Method for Rotating Machinery Using Refined Composite Multivariate Multiscale Amplitude-Aware Permutation Entropy

Table 4

The diagnostic results gained by combining diverse methods with four classifiers.

Diverse classifiers	The testing accuracy of classifiers with diverse approaches (%)					Average accuracy (%)
	RCmvMAAPE (%)	RCmvMPE (%)	mvMAAPE (%)	RCmvMSE (%)	RCMAAPE (%)
ELM	97.04	95.93	88.52	94.44	85.19	92.22
SVM	95.56	94.07	87.04	92.96	83.33	90.59
KELM	98.52	96.30	89.26	94.81	85.92	92.96
WOA-KELM	100	99.26	91.48	95.93	88.89	95.11
Average accuracy (%)	97.78	96.39	89.07	94.54	85.83	—