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

Table 12

Comparative analysis of different optimization algorithms for cost function: .

Sample	Algo.	DRtn	Mean	E(DRtn)	Min
300	PSO	-1.08	-1.69	-1.32	-1.99
	GA	-1.03	-1.65	-1.39	-1.82
	FPA	-1.04	-1.63	-1.41	-1.99
450	PSO	2.54	1.85	-2.28	-0.22
	GA	-0.22	2.54	-0.22	1.85
	FPA	-2.28	2.54	0.47	-0.22
600	PSO	2.54	-2.28	0.47	0.47
	GA	-0.90	-0.22	-2.28	-0.90
	FPA	-1.59	-2.28	2.54	1.85
750	PSO	-1.59	0.47	-0.90	-0.90
	GA	-2.28	0.47	0.47	1.16
	FPA	1.85	1.16	0.47	0.47