Table 11 Performance comparison among base models and proposed model.
From: Developing a hybrid time-series artificial intelligence model to forecast energy use in buildings
Model/method | Performance indices | Improvement by the SAMFOR | ||||||
|---|---|---|---|---|---|---|---|---|
RMSE (kWh) | MAE (kWh) | MAPE (%) | R | RMSE (kWh) | MAE (kWh) | MAPE (%) | R | |
SARIMA | 44.08 | 36.94 | 58.19 | 0.806 | 42.3 | 31.9 | 49.6 | 0.108 |
SVR | 11.70 | 5.78 | 9.89 | 0.909 | 9.9 | 0.8 | 0.3 | 0.005 |
SRIMA-SVR | 13.91 | 9.83 | 10.42 | 0.902 | 12.1 | 4.8 | 0.9 | 0.012 |
RF | 8.79 | 5.12 | 10.18 | 0.903 | 7.0 | 0.1 | 0.6 | 0.011 |
Proposed SAMFOR | 1.77 | 5.02 | 9.56 | 0.914 | Â | Â | Â | Â |
