Simulation and study of maximum power point tracking for rim-driven tidal current energy power generation systems

This paper presents a maximum power point tracking (MPPT) control algorithm based on an intelligent reinforcement learning. The proposed model-free Q-learning algorithm realizes the online learning of the control algorithm of the tidal power generation system by updating the action values stored in the Q-table. By learning the optimal rotor speed-output power curve, the algorithm fits the optimal generator curve and applies the optimal Pe−ωr curve to the optimal control method of the tidal power generation systems.