Table 2.
Accuracy of status recognition at different currents when fixed parameters are used. (a) When the learning rate is 0.01, the number of hidden layers is 4, and the dropout rate is 0.15. (b) When the learning rate is 0.02, the number of hidden layers is 5, and the dropout rate is 0.25.
| Current (A) | Accuracy (%) |
|---|---|
| (a) | |
| 1000 | 20.18 |
| 1500 | 22.00 |
| 2000 | 92.65 |
| 2500 | 93.75 |
| 3000 | 97.66 |
| 3500 | 99.21 |
| (b) | |
| 1000 | 95.05 |
| 1500 | 96.88 |
| 2000 | 96.35 |
| 2500 | 82.68 |
| 3000 | 79.56 |
| 3500 | 84.38 |