A Hybrid Model Based on Ensemble Empirical Mode Decomposition and Fruit Fly Optimization Algorithm for Wind Speed Forecasting

Table 8

The typical results of the hybrid model and the results of the other models for the four seasons.

Case	Errors	Persistence model	ARIMA model	EEMD-FOARBF	EEMD-FOAGRNN	EEMD-FOASVR	Hybrid model
Spring	MAE	0.7741	0.7285	0.3675	0.5690	0.3692	0.0976
	RMSE	0.9023	0.8769	0.4714	0.7505	0.4783	0.1308
	IA	0.8638	0.8684	0.9647	0.9019	0.9617	0.9973
Summer	MAE	0.7208	0.7111	0.4312	0.5280	0.3940	0.1032
	RMSE	0.8589	0.8615	0.5287	0.6472	0.4920	0.1280
	IA	0.8716	0.8682	0.9374	0.8965	0.9496	0.9964
Fall	MAE	0.6708	0.7879	0.6917	0.4197	0.3169	0.1113
	RMSE	0.8585	1.0181	1.0098	0.6322	0.4604	0.1453
	IA	0.9554	0.9326	0.9294	0.9732	0.9874	0.9987
Winter	MAE	0.7833	0.7017	0.6117	0.6211	0.4171	0.0875
	RMSE	1.0450	0.9779	0.7548	0.7955	0.5301	0.1164
	IA	0.9098	0.9133	0.9399	0.9264	0.9749	0.9988
Average	MAE	0.7373	0.7323	0.5255	0.5345	0.3743	0.0999
	RMSE	0.9162	0.9336	0.6912	0.7064	0.4902	0.1301
	IA	0.9002	0.8956	0.9429	0.9245	0.9684	0.9978