Optimization of Wind Turbine Speed Reference Based on Deep Reinforcement Learning

To improve the wind energy capture performance of large-power wind turbines under variable wind speeds, this paper proposes an intelligent control strategy optimization method based on Deep Reinforcement Learning (DRL). Aiming at the lagging response of traditional MPPT control in large-inertia wind turbines and the difficulty in balancing power capture efficiency with mechanical fatigue loss, this paper proposes a decision-making framework based on deep reinforcement learning. By designing a multi-objective reward function that takes into account both maximizing power generation and suppressing mechanical loads, the optimal rotational speed reference sequence that the wind turbine can track is obtained. The simulation results show that: Compared with the traditional tip speed ratio method, the proposed strategy can increase the average power capture efficiency by 2.3% under variable wind speed conditions, and at the same time reduce the fatigue load of the transmission shaft by more than 16.6%.