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

Table 4

Test results of different IMFs input signals.

Working condition	IMF1-IMF3	IMF1-IMF5	IMF1-IMF6	IMF1-IMF4
A->B	94.17%	96.67%	97.83%	97.83%
A->C	93.50%	95.33%	96.50%	97.33%
A->D	94.00%	96.83%	97.33%	96.50%
B->A	92.83%	94.83%	95.83%	96.67%
B->C	95.83%	98.33%	99.33%	100%
B->D	94.67%	97.83%	97.50%	97.17%
C->A	92.33%	94.33%	94.83%	96.50%
C->B	92.83%	98.83%	97.33%	97.33%
C->D	94.83%	97.17%	98.67%	99.50%
D->A	93.17%	95.83%	96.33%	97.17%
D->B	94.50%	96.17%	98.00%	97.33%
D->C	95.17%	96.67%	98.83%	99.83%
Average	93.99%	96.18%	97.36%	98.17%