A Simple Hybrid Model for Short-Term Load Forecasting

Table 2

The parameters of the adaptively fitted double exponential smoothing models.

Day	Period of hourly forecast	Parameters
		Deterministic part		Fluctuation part
		𝛼	𝛾	𝛼	𝛾
January 15th	01.00 AM–04.00 AM	0.15	0.92	0.95	0.04
	05.00 AM–08.00 AM	0.3	1	0.91	0.58
	09.00 AM–12.00 PM	0.1	0.05	0.98	0.43
	12.00 PM–04.00 PM	0.3	0.65	0.34	0
	05.00 PM–08.00 PM	0.25	0.5	0.83	0.09
	09.00 PM–12.00 AM	0.3	1	0.14	1
April 11th	01.00 AM–04.00 AM	0.15	0.95	1	0.03
	05.00 AM–08.00 AM	0.3	1	0.33	0
	09.00 AM–12.00 PM	0.1	0.2	0.43	0.72
	12.00 PM–04.00 PM	0.2	1	0.29	0.29
	05.00 PM–08.00 PM	0.75	0.8	0.04	0.08
	09.00 PM–12.00 AM	0.1	0.95	0.13	0.95
July 14th	01.00 AM–04.00 AM	0.2	1	0.84	0.29
	05.00 AM–08.00 AM	0.25	1	0.61	1
	09.00 AM–12.00 PM	0.1	0.25	0.44	0.44
	12.00 PM–04.00 PM	0.8	0.15	0.62	0.98
	05.00 PM–08.00 PM	0.3	0.8	0.68	0.05
	09.00 PM–12.00 AM	0.6	0.5	0.13	0.99