A Deep Domain-Adversarial Transfer Fault Diagnosis Method for Rolling Bearing Based on Ensemble Empirical Mode Decomposition

Table 10

Comparison of different models in diagnosis accuracy.

Working condition	SVM	CNN	TCA	JDA	EMBRNDNMD
E->F	72.92%	81.67%	88.33%	94.17%	99.79%
E->G	69.17%	84.17%	86.67%	92.71%	98.96%
F->E	68.33%	80.62%	81.04%	91.04%	98.33%
F->G	66.46%	79.79%	79.79%	87.29%	97.51%
G->E	69.38%	74.79%	83.33%	93.33%	98.12%
G->F	69.58%	80.83%	86.67%	92.29%	98.54%
Average	69.31%	80.31%	84.31%	91.81%	98.54%