Maximum Power Point Tracking Based on Reinforcement Learning in Photovoltaic System

Abstract

Maximum power point tracking (MPPT) technology is usually used in photovoltaic (PV) systems to extract the maximum power. Although the conventional MPPT techniques are easy to be implemented, they have to tune their control parameters by using trial-and-error method, which is not adaptive to different working conditions. Unlike the conventional MPPT techniques, the reinforcement learning-based MPPT (RL-MPPT) method has advantages of self-learning ability, which is better applicable performance under different weather conditions. To evaluate the RL-MPPT method, the simulations of Standard Test Conditions (STC) and varying irradiance conditions are performed.