Table 1.
Results for the SVR, Tf model and NNs, parallel form GE
| MSE (−) | Fit (%) | |||
|---|---|---|---|---|
| Training set | Test set | Training set | Test set | |
|
SVR: C = 100 γ = 2.79 × 10−2 |
4.6 × 10−4 | 9.1 × 10−4 | 90.93 | 86.96 |
| Tf of PID | – | 1.5 × 10−3 | – | 83.52 |
|
NNa 14-5-1 |
1.2 × 10−3 | 2.9 × 10−3 | 84.89 | 80.14 |
aWhere NN 14-5-1 means: 14 is the number of inputs; five is the number of hidden neurons; one is the number of outputs