Research Article
Fault Diagnosis Method of Check Valve Based on Multikernel Cost-Sensitive Extreme Learning Machine
Table 5
The data sample distribution by oversampling for check valve.
| Cost | Fault classes | Original samples | Processed samples |
| 1 | NC | 70 | 70 | NK | 20 | 70 | 1.5 | NC | 70 | 70 | NK | 20 | 105 | 2 | NC | 70 | 70 | NK | 20 | 140 | 2.5 | NC | 70 | 70 | NK | 20 | 175 | 3 | NC | 70 | 70 | NK | 20 | 210 | 3.5 | NC | 70 | 70 | NK | 20 | 245 | 4 | NC | 70 | 70 | NK | 20 | 280 | 4.5 | NC | 70 | 70 | NK | 20 | 315 | 5 | NC | 70 | 70 | NK | 20 | 350 | - - | - - | - - | - - | - - | - - | - - |
|
|