Research Article
Discriminant Feature Distribution Analysis-Based Hybrid Feature Selection for Online Bearing Fault Diagnosis in Induction Motors
Table 4
Classification performance of the three different approaches for individual fault types of the different datasets.
| Dataset | Model | Inner | Outer | Roller | Inner + roller | Outer + inner | Outer + roller | Inner + outer + roller | Normal | Average performance |
| Dataset 1 | All features | 99.65 | 79.50 | 97.42 | 100.00 | 95.67 | 97.08 | 92.67 | 100.00 | 95.25 | Proposed model | 99.70 | 86.42 | 100.00 | 97.58 | 96.25 | 99.92 | 99.08 | 100.00 | 97.37 | Algorithm | 99.70 | 81.50 | 98.42 | 100.00 | 95.90 | 97.80 | 93.67 | 100.00 | 95.87 |
| Dataset 2 | All features | 98.53 | 90.50 | 98.75 | 98.25 | 92.83 | 96.08 | 97.67 | 100.00 | 96.58 | Proposed model | 98.75 | 95.33 | 100.00 | 100.00 | 99.50 | 97.92 | 98.75 | 100.00 | 98.78 | Algorithm | 98.75 | 91.40 | 98.75 | 98.95 | 93.83 | 96.25 | 97.70 | 100.00 | 96.95 |
| Dataset 3 | All features | 100.00 | 78.50 | 97.75 | 100.00 | 90.75 | 94.92 | 87.42 | 99.83 | 93.65 | Proposed model | 100.00 | 96.17 | 100.00 | 100.00 | 98.42 | 100.00 | 91.83 | 100.00 | 98.30 | Algorithm | 100.00 | 80.60 | 97.75 | 100.00 | 91.75 | 95.70 | 87.90 | 99.83 | 94.19 |
| Dataset 4 | All features | 100.00 | 89.00 | 96.58 | 100.00 | 97.83 | 99.08 | 96.50 | 100.00 | 97.38 | Proposed model | 100.00 | 96.42 | 100.00 | 100.00 | 99.92 | 99.92 | 98.33 | 100.00 | 99.32 | Algorithm | 100.00 | 89.95 | 97.00 | 100.00 | 98.00 | 99.15 | 96.50 | 100.00 | 97.58 |
| Dataset 5 | All features | 100.00 | 100.00 | 99.92 | 100.00 | 100.00 | 94.00 | 97.33 | 100.00 | 98.91 | Proposed model | 100.00 | 100.00 | 100.00 | 100.00 | 100.00 | 98.33 | 100.00 | 100.00 | 99.79 | Algorithm | 100.00 | 100.00 | 100.00 | 100.00 | 100.00 | 94.17 | 97.33 | 100.00 | 98.94 |
| Dataset 6 | All features | 100.00 | 99.67 | 100.00 | 100.00 | 100.00 | 100.00 | 100.00 | 100.00 | 99.96 | Proposed model | 100.00 | 100.00 | 100.00 | 100.00 | 100.00 | 100.00 | 100.00 | 100.00 | 100.00 | Algorithm | 100.00 | 99.75 | 100.00 | 100.00 | 100.00 | 100.00 | 100.00 | 100.00 | 99.97 |
| Dataset 7 | All features | 99.92 | 100.00 | 100.00 | 100.00 | 100.00 | 100.00 | 100.00 | 100.00 | 99.99 | Proposed model | 100.00 | 100.00 | 100.00 | 100.00 | 100.00 | 100.00 | 100.00 | 100.00 | 100.00 | Algorithm | 99.25 | 100.00 | 100.00 | 100.00 | 100.00 | 100.00 | 100.00 | 100.00 | 99.91 |
| Dataset 8 | All features | 100.00 | 100.00 | 100.00 | 100.00 | 100.00 | 100.00 | 100.00 | 100.00 | 100.00 | Proposed model | 100.00 | 100.00 | 100.00 | 100.00 | 100.00 | 100.00 | 100.00 | 100.00 | 100.00 | Algorithm | 100.00 | 100.00 | 100.00 | 100.00 | 100.00 | 100.00 | 100.00 | 100.00 | 100.00 |
| Dataset 9 | All features | 100.00 | 100.00 | 100.00 | 100.00 | 100.00 | 100.00 | 100.00 | 100.00 | 100.00 | Proposed model | 100.00 | 100.00 | 100.00 | 100.00 | 100.00 | 100.00 | 100.00 | 100.00 | 100.00 | Algorithm | 100.00 | 100.00 | 99.83 | 100.00 | 100.00 | 100.00 | 100.00 | 100.00 | 99.98 |
| Dataset 10 | All features | 100.00 | 100.00 | 100.00 | 100.00 | 100.00 | 100.00 | 100.00 | 100.00 | 100.00 | Proposed model | 100.00 | 100.00 | 100.00 | 100.00 | 100.00 | 100.00 | 100.00 | 100.00 | 100.00 | Algorithm | 100.00 | 100.00 | 100.00 | 100.00 | 99.00 | 100.00 | 100.00 | 100.00 | 99.88 |
|
|