Research Article
A Hybrid Forecasting Model Based on Empirical Mode Decomposition and the Cuckoo Search Algorithm: A Case Study for Power Load
Table 3
Comparison of power load forecasting evaluation by using different methods between weekdays and weekends.
| | Model | Date | Evaluation criterion | | MAPE (%) | RMSE | MAE | GRA |
| | BPNN | Weekdays | 1.30 | 164.97 | 118.03 | 0.8654 | | Weekends | 1.23 | 148.35 | 101.40 | 0.8600 |
| | GABPNN | Weekdays | 1.17 | 154.94 | 106.05 | 0.8781 | | Weekends | 1.05 | 137.21 | 88.33 | 0.8779 |
| | WNN | Weekdays | 1.28 | 162.09 | 114.33 | 0.8688 | | Weekends | 1.27 | 162.83 | 108.65 | 0.8581 |
| | CSAWNN | Weekdays | 1.23 | 157.05 | 109.05 | 0.8737 | | Weekends | 1.13 | 138.86 | 94.96 | 0.8689 |
| | EMD-CSAWNN | Weekdays | 0.82 | 101.56 | 72.77 | 0.9087 | | Weekends | 0.68 | 87.00 | 55.19 | 0.9148 |
| | ARIMA | Weekdays | 1.48 | 184.23 | 128.98 | 0.8539 | | Weekends | 1.34 | 169.06 | 109.09 | 0.8548 |
|
|
