Research Article
Estimation of Vehicle Dynamic Response from Track Irregularity Using Deep Learning Techniques
Table 3
Accuracy indices of the vehicle response of line 1 in the inspection-simulation dataset.
| Item | Output | RMSE | MAE | ρ | TIC |
| Vertical forces | Left force of axle 1 | 1745.1141 | 1211.5344 | 0.8810 | 0.0155 | Left force of axle 2 | 1796.8741 | 1239.4564 | 0.8825 | 0.0160 | Left force of axle 3 | 1770.2314 | 1238.6849 | 0.8775 | 0.0158 | Left force of axle 4 | 1836.0664 | 1275.5680 | 0.8773 | 0.0163 | Right force of axle 1 | 1746.3233 | 1204.2282 | 0.8819 | 0.0156 | Right force of axle 2 | 1783.9973 | 1221.8113 | 0.8831 | 0.0159 | Right force of axle 3 | 1775.3752 | 1232.1890 | 0.8770 | 0.0158 | Right force of axle 4 | 1821.2450 | 1259.2483 | 0.8783 | 0.0162 |
| Wheel unloading rate | Axle 1 | 0.0196 | 0.0140 | 0.7506 | 0.2237 | Axle 2 | 0.0197 | 0.0141 | 0.7380 | 0.2309 | Axle 3 | 0.0197 | 0.0142 | 0.7468 | 0.2261 | Axle 4 | 0.0199 | 0.0144 | 0.7330 | 0.2331 |
| Vertical car body acceleration | Front | 0.0132 | 0.0085 | 0.9689 | 0.1258 | Rear | 0.0135 | 0.0087 | 0.9678 | 0.1283 |
|
|