Short-Term Wind Speed Forecasting Using the Data Processing Approach and the Support Vector Machine Model Optimized by the Improved Cuckoo Search Parameter Estimation Algorithm

Table 2

Forecasting error results with different forecasting horizons.

Horizon	Wind turbine	Error type	EMD-SVM	EMD-GA-SVM	EMD-CS-SVM	EMD-FA-SVM	EMD-SDCS-SVM	Horizon	Wind turbine	Error type	EMD-SVM	EMD-GA-SVM	EMD-CS-SVM	EMD-FA-SVM	EMD-SDCS-SVM
1-step-ahead	#1	MSE	0.6798	0.6798	0.6798	0.6460	0.6382	4-step-ahead	#1	MSE	0.9262	0.9262	0.9262	1.5420	0.7883
		MAPE	4.6937	4.6937	4.6937	4.924	4.4052			MAPE	6.3854	6.3854	6.3854	7.103%	5.5685
	#2	MSE	0.8628	0.8226	0.8534	0.9262	0.7959		#2	MSE	1.256	1.0524	1.1339	0.9147	1.011
		MAPE	6.0093	5.8184	5.9602	5.327	5.597			MAPE	8.2016	7.279	7.8073	6.197	6.8343
	#3	MSE	0.8256	0.7907	0.8304	0.9553	0.803		#3	MSE	1.0914	0.9676	1.1172	1.1185	1.0714
		MAPE	5.5431	5.2876	5.5765	6.234	5.2886			MAPE	7.2059	6.4716	7.3004	6.897	6.8158
	#4	MSE	0.8657	0.8514	0.9065	0.7401	0.8135		#4	MSE	1.1689	1.1094	1.3419	1.0754	1.0149
		MAPE	5.5708	5.5101	5.8196	5.493	5.1737			MAPE	6.8888	6.526	7.7831	6.548%	6.0562
2-step-ahead	#1	MSE	0.7707	0.7707	0.7707	0.9932	0.7184	6-step-ahead	#1	MSE	0.9982	0.9982	0.9982	1.8679	0.8342
		MAPE	5.3944	5.3944	5.3944	5.006	5.0049			MAPE	6.9568	6.9568	6.9568	9.435	5.898
	#2	MSE	1.0091	0.9242	0.9806	0.7437	0.8885		#2	MSE	1.4805	1.0856	1.2166	2.2042	1.0627
		MAPE	6.8987	6.4877	6.7938	5.892	6.0704			MAPE	8.9118	7.6722	8.3732	8.378%	7.0088
	#3	MSE	0.8876	0.8259	0.8963	1.2746	0.8674		#3	MSE	1.1505	0.9688	1.1615	1.7628	1.0857
		MAPE	5.9299	5.5404	5.9339	7.069	5.5753			MAPE	7.8968	6.643	7.9065	6.910	6.949
	#4	MSE	0.9916	0.9558	1.0946	1.0891	0.8804		#4	MSE	1.2147	1.0537	1.5362	2.8053	0.981
		MAPE	6.2854	5.8933	6.8798	6.587	5.4089			MAPE	7.3785	6.423	8.9868	8.532	6.1465