Journal of Engineering and Applied Science

Table 3 Goodness-of-fit results of SVR−BES versus other forecasting models

From: Support vector regression-bald eagle search optimizer-based hybrid approach for short-term wind power forecasting

Method	SVR hyper-parameters			Error criterion
Method	C	γ	ε	R	RMSE	MAE	MAPE (%)
DT	-	-	-	0.8817	1.7395	1.2772	9.6428
RF	-	-	-	0.9308	1.3396	1.0107	7.6304
SVR	10	5	0.001	0.9107	1.4924	1.1141	8.4110
SVR–GWO	521.3222	1.175882	0.00199	0.9396	1.3202	1.0056	7.5920
SVR–MRFO	132.6669	1.083281	0.01984	0.9443	1.2837	0.9655	7.2895
SVR−BES	14.8385	0.828621	0.09595	0.9457	1.2598	0.9374	7.0774

Back to article page