Multidomain Feature Fusion Network for Fault Diagnosis of Rolling Machinery

Table 5

The average performance in different domains.

Input	Model	Average accuracy
		CWRU (%)	Jiangnan (%)	Paderborn (%)
TF	WDCNN	99.78	99.74	99.73
	1D-CNN	99.81	99.80	99.77
FF	SDAE	99.69	99.71	99.78
	1D-CNN	99.75	99.74	99.72
TFF	VGG16	99.84	99.85	99.82
	2D-CNN	99.87	99.86	99.84
TF + FF + TFF	MFFN	99.95	99.92	99.91