Short-Term Wind Speed Prediction Using EEMD-LSSVM Model

Table 6

Performance comparison of different models for forecasting wind speed (Case Two).

Prediction horizon	Indicators	LSSVM	ARIMA	EMD-LSSVM	BP	EEMD-ARIMA	EEMD-LSSVM
One-step	RMSE	0.4599	0.4719	0.2573	0.4783	0.1681	0.1666
	MAPE	11.96%	11.77%	5.315%	12.60%	4.395%	3.803%
	MAE	0.3437	0.3467	0.1825	0.3574	0.1202	0.1178
	CC	0.9790	0.9777	0.9934	0.9772	0.9972	0.9972
	SSE	30.4624	32.0723	9.5337	32.9460	4.0695	3.9988
	SDE	0.4597	0.4704	0.2572	0.4783	0.1686	0.1664
Two-step	RMSE	0.7006	0.7294	0.3611	0.7222	0.3166	0.2288
	MAPE	18.69%	18.91%	7.926%	18.78%	7.043%	5.905%
	MAE	0.5253	0.5435	0.2723	0.5353	0.2223	0.1786
	CC	0.9493	0.9454	0.9869	0.9461	0.9911	0.9947
	SSE	70.1998	76.0851	18.6418	74.5792	14.3304	7.4848
	SDE	0.7015	0.7250	0.3557	0.7217	0.3146	0.2282
Three-step	RMSE	0.8728	0.9032	0.4525	0.8907	0.3688	0.3023
	MAPE	23.26%	23.50%	9.964%	22.77%	9.770%	7.153%
	MAE	0.6491	0.6754	0.3459	0.6569	0.2820	0.2290
	CC	0.9183	0.9142	0.9794	0.9157	0.9860	0.9906
	SSE	108.1625	115.8408	29.0710	112.6429	19.3140	12.9783
	SDE	0.8789	0.8942	0.4418	0.8906	0.3695	0.3018