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

Table 3

The percentage of choosing action in different phase for the experiment with real weather data.

Interval (minutes)	Action (V)
	+5	+2	+0.5	−0.5	−2	−5
Early 0~25	24%	16%	8%	12%	12%	28%
Middle 100~125	16%	16%	12%	24%	24%	8%
Final 215~240	4%	16%	32%	32%	12%	4%