An Experimental Investigation of FNN Model for Wind Speed Forecasting Using EEMD and CS

Table 1

The one-step ahead forecasting errors of the four models in three months.

Models		FNN model					EEMD-FNN model					CS-FNN model					CS-EEMD-FNN model
WTG^a		25	26	27	28	Mean	25	26	27	28	Mean	25	26	27	28	Mean	25	26	27	28	Mean
Jan.	MAE	0.84	0.90	1.11	0.89	0.94	0.63	0.60	0.77	0.62	0.66	0.60	0.63	0.79	0.62	0.66	0.46	0.46	0.64	0.50	0.52
	MAPE	7.47	10.66	9.86	8.47	9.12	5.74	6.88	7.19	6.03	6.46	5.46	6.97	7.34	6.12	6.47	4.31	5.08	5.93	4.77	5.02
	RMSE	1.15	1.14	2.14	1.23	1.42	0.85	0.80	1.09	0.83	0.89	0.80	0.84	1.22	0.83	0.92	0.64	0.62	0.91	0.68	0.71
Apr.	MAE	3.17	5.63	5.85	2.34	4.25	0.87	1.04	1.27	1.02	1.05	0.97	1.93	3.40	2.47	2.19	0.63	0.90	0.77	0.75	0.76
	MAPE	25.35	39.19	42.48	21.40	32.11	9.05	9.50	12.20	13.14	10.97	9.46	14.72	24.69	21.26	17.53	6.71	8.14	7.56	8.79	7.8
	RMSE	9.96	15.16	14.42	5.13	11.17	1.48	1.92	1.98	1.37	1.69	2.59	6.25	10.53	7.63	6.75	1.01	1.46	1.13	1.14	1.19
Jul.	MAE	5.19	9.9	8.1	4.12	6.83	0.91	1.78	2.00	1.42	1.53	4.49	1.29	1.95	2.66	2.60	0.99	1.58	1.46	0.99	1.25
	MAPE	53.46	89.05	72.84	51.48	66.71	14.80	24.52	22.87	30.87	23.27	45.98	18.31	20.80	36.34	30.36	14.63	20.86	18.63	17.78	17.98
	RMSE	12.24	23.43	19.76	10.14	16.39	1.77	2.86	3.62	2.09	2.59	40.57	2.77	6.00	7.41	14.19	1.21	2.90	2.57	1.76	2.11

Wind Turbine Generator.