Research Article
A Deep Domain-Adversarial Transfer Fault Diagnosis Method for Rolling Bearing Based on Ensemble Empirical Mode Decomposition
Table 7
Comparison of different models in diagnosis accuracy.
| | Working condition | SVM | CNN | TCA | JDA | EMBRNDNMD |
| | A->B | 67.25% | 78.33% | 80.38% | 91.75% | 97.83% | | A->C | 67.42% | 79.33% | 75.67% | 84.58% | 97.33% | | A->D | 67.37% | 78.17% | 76.25% | 85.31% | 97.50% | | B->A | 75.13% | 80.53% | 91.75% | 93.28% | 96.67% | | B->C | 62.38% | 78.87% | 74.63% | 91.27% | 100% | | B->D | 67.56% | 80.05% | 82.18% | 93.83% | 98.17% | | C->A | 65.37% | 71.08% | 84.37% | 91.08% | 96.50% | | C->B | 61.22% | 69.50% | 84.53% | 87.58% | 98.33% | | C->D | 67.51% | 79.33% | 82.75% | 91.32% | 99.50% | | D->A | 62.34% | 78.00% | 80.52% | 92.64% | 97.17% | | D->B | 64% | 78.83% | 86.78% | 91.16% | 98.33% | | D->C | 67.36% | 74.17% | 86.43% | 90.93% | 99.83% | | Average | 66.24% | 77.14% | 82.34% | 90.39% | 98.17% |
|
|
