Wind power efficiency is hindered by wake effects, especially at high wind speeds, increasing uncertainty and turbine loads. Traditional yaw control mitigates this but raises safety risks. To address this, we propose an MRL framework integrating Model Predictive Control (MPC) and Reinforcement Learning (RL) to enhance power generation, reduce loads, and ensure safety. FAST.Farm simulations confirm its effectiveness over RL alone.

Reinforcement Learning,MPC,Wake effects,Wind turbine load