Research Article
The Chaotic Prediction for Aero-Engine Performance Parameters Based on Nonlinear PLS Regression
Table 2
The optimal results after selecting independent variable by Bootstrap method based on NPLS.
| | Forecasting method | Embedding dimension (m) | Delay time (τ) | Subset numbers | NMSE |
| | Cubic spline interpolation | 9 | 3 | 5 | 0.6480 | | Uniform kernel | 28 | 6 | 4 | 0.5753 | | Epanechnikov kernel | 9 | 3 | 7 | 0.5917 | | Bi-weight kernel | 8 | 19 | 2 | 0.5606 | | Triweight kernel | 9 | 3 | 5 | 0.5786 | | Gaussian kernel | 3 | 10 | 4 | 0.6510 |
|
|
