Short-Term Reactive Power Forecasting Based on Real Power Demand Using Holt-Winters’ Model Ensemble by Global Flower Pollination Algorithm for Microgrid

Table 8

SARIMA fit model with respect to reactive power time series data.

Parameter	Coefficient	Std. error			0.025	0.975
ar.L1	-1.1571	0.264	-5.25	0.00	-1.974	-0.940
ar.L2	-0.4891	0.137	-3.56	0.00	-0.758	-0.220
ma.L1	-0.0046	2751.87	-1.67	1.00	-5393.57	5393.56
ma.L2	-0.9954	2739.04	-0.00	1.00	-5369.41	5397.42
ma.S.L4	-0.9122	8.393	-0.109	0.913	-7.36	15.53
ma.S.L8	-0.0863	0.783	-0.110	0.912	-1.62	1.44
Sigma 2	0.2848	783.49	0.000	1.00	-1535.33	1535.90