A Reinforcement Learning-Based Maximum Power Point Tracking Method for Photovoltaic Array

Table 4

Comparison of efficiency factor for the three comparing methods.

Learning phase (min)	Methods
	Open-circuit voltage	P&O	RLMPPT
Early 10:00~10:25	86.9%	83.5%	89.2%
Middle 11:40~12:05	87.5%	97.8%	99.4%
Final 13:35~14:00	87.8%	97.7%	99.3%
Overall 10:00~14:00	87.6%	95.4%	98.5%