State Diagnosis of Elevator Control Transformer over Vibration Signal Based on MEA-BP Neural Network

Table 6

Diagnostic accuracy of the models for different operating states after 30 independent runs.

Algorithm	Normal			High-order harmonic state			Winding fault			Core fault
	Mean	Min	Max	Mean	Min	Max	Mean	Min	Max	Mean	Min	Max
WP-BPNN	0.843	0.833	0.875	0.841	0.832	0.877	0.837	0.830	0.874	0.841	0.829	0.877
WP-MEA-BPNN	0.877	0.869	0.884	0.874	0.862	0.883	0.871	0.863	0.884	0.876	0.835	0.885
EWP-BPNN	0.849	0.837	0.881	0.892	0.887	0.902	0.891	0.886	0.900	0.889	0.842	0.893
EWP-MEA-BPNN	0.910	0.905	0.915	0.908	0.904	0.913	0.907	0.903	0.912	0.905	0.899	0.917