Research Article
Deep Learning Enabled Fault Diagnosis Using Time-Frequency Image Analysis of Rolling Element Bearings
Table 29
Prediction accuracies for MFPT.
| | Model | 20 epochs with
early stopping | 30 epochs with no early stopping | No overlap |
| | Architecture 2 | 79.1% | 80.9% | 75.2% | | Architecture 1 | 82.9% | 75.1% | 75.1% | | Proposed CNN architecture | 96.4% | 93.8% | 87.3% |
|
|
