Research Article
Fault Line Selection Method Based on Transfer Learning Depthwise Separable Convolutional Neural Network
Table 6
Classification accuracy of each model under different noise conditions.
| Line selection model | SNR/dB | 30 | 35 | 40 | 45 |
| WPT + SVM | 84.21 | 85.03 | 85.43 | 86.68 | WPT + DT | 86.21 | 87.96 | 89.35 | 89.45 | WPT + RF | 87.06 | 88.78 | 89.84 | 90.16 | DSCNN | 99.20 | 99.28 | 99.28 | 99.33 |
|
|