Research Article
Casing Vibration Fault Diagnosis Based on Variational Mode Decomposition, Local Linear Embedding, and Support Vector Machine
Table 4
Classification results with different training data.
| | Training data | RBF | PNN | 1-v-r | H-SVMs | 1-v-r with PCA | H-SVMs with PCA |
| | Amplitude | 67.3% | 82.1% | 81.7% | 75.6% | 79.8% | 74.5% | | Frequency | 88.2% | 92.5% | 94.2% | 90.4% | 90.1% | 86.1% | | Amplitude + Frequency | 83.8% | 87.3% | 96.3% | 93.6% | 96.6% | 94.0% | | VMD-Envelope | 81.6% | 85.8% | 85.8% | 80.4% | 81.3% | 76.0% | | VMD-Envelope + Omega | 84.2% | 86.3% | 91.8% | 87.8% | 81.4% | 76.1% | | Amplitude + Frequency + VMD-Envelope | 87.2% | 88.3% | 96.4% | 94.8% | 95.6% | 94.5% | | Amplitude + Frequency + VMD-Envelope + Omega | 86.5% | 88.3% | 97.1% | 96.3% | 95.6% | 94.5% |
|
|
