Research Article
Integrating Feature Engineering with Deep Learning to Conduct Diagnostic and Predictive Analytics for Turbofan Engines
Table 6
Identified key sensors using stepwise regression and machine learning.
| Features | SR (%) | MARS (%) | RF (%) | XGB (%) | Average (%) |
| Sensor_11 | 76.64 | 35.1 | 25.71 | 7.02 | 36.12 | Sensor_9 | 10.01 | 9.78 | 6.3 | 49.52 | 18.9 | Sensor_12 | 6.77 | 12.3 | 13.99 | 34.26 | 16.83 | Sensor_4 | 2.59 | 22.33 | 22.06 | 7.49 | 13.62 | Sensor_7 | 1.47 | 6.44 | 7.81 | 0.23 | 3.99 | Sensor_15 | 0.76 | 3.66 | 5.01 | 0.2 | 2.41 | Sensor_21 | 0.61 | 5.33 | 2.8 | 0.09 | 2.21 | Sensor_20 | 0.3 | 1.4 | 6.18 | | 1.97 | Sensor_2 | 0.36 | 1.01 | 2.76 | | 1.03 | Sensor_17 | 0.21 | 0.64 | 3.19 | | 1.01 | Sensor_14 | 0.07 | 0.53 | 1.52 | 0.43 | 0.64 | Sensor_3 | 0.18 | 0.19 | 1.57 | | 0.49 | Sensor_13 | | | 0.63 | 0.67 | 0.33 | Sensor_8 | 0.01 | 0.49 | 0.46 | 0.09 | 0.26 | Sensor_6 | 0.02 | 0.79 | | | 0.2 |
|
|