Research Article
Rolling Bearing Fault Diagnostic Method Based on VMD-AR Model and Random Forest Classifier
Table 2
Classification result for fault location problem using VMD-AR model and SVM classifier.
| Number of modes | Number of training samples | Number of testing samples | Classification accuracy (%) | Total accuracy (%) | Normal | IR fault | B fault | OR fault |
| 4 | 80 | 320 | 99 | 100 | 100 | 53.13 | 88.03 | 160 | 240 | 100 | 100 | 100 | 50.83 | 87.71 | 240 | 160 | 100 | 100 | 100 | 53.25 | 88.31 | 320 | 80 | 100 | 100 | 100 | 72.5 | 93.13 |
| 5 | 80 | 320 | 100 | 87 | 98.25 | 95 | 95.06 | 160 | 240 | 100 | 86.5 | 98.83 | 100 | 96.33 | 240 | 160 | 100 | 88.25 | 99.25 | 100 | 96.88 | 320 | 80 | 100 | 89.5 | 99 | 100 | 97.13 |
| 6 | 80 | 320 | 100 | 100 | 88.88 | 100 | 97.22 | 160 | 240 | 100 | 98 | 99.17 | 100 | 99.29 | 240 | 160 | 100 | 100 | 100 | 100 | 100 | 320 | 80 | 100 | 100 | 100 | 100 | 100 |
| 7 | 80 | 320 | 99.63 | 96.75 | 92.13 | 91 | 94.88 | 160 | 240 | 99.67 | 99.83 | 97.67 | 93.5 | 97.67 | 240 | 160 | 100 | 100 | 100 | 95.75 | 98.94 | 320 | 80 | 100 | 100 | 100 | 94.5 | 98.63 |
|
|