Active Fault‐Tolerant <scp>MPPT</scp> Control for Variable‐Speed Wind Turbine Systems Using Reinforcement Learning

ABSTRACT In the present paper, a reinforcement learning (RL)‐based fault‐tolerant control scheme is developed for variable‐speed wind turbines in the event of actuator faults and disturbance torque. The designed controller law is an aggregate of two sub‐controllers: an equivalent controller designed to achieve the maximum wind power capture and a RL‐based robust controller designed to mitigate the impacts of actuator faults and disturbance torque. In the development of the RL controller, the estimated fault term from a designed observer is utilized to create an adequate cost function so that the disturbance rejection and fault‐tolerant control problem can be transformed into an optimal control problem. Then, by using the online policy iteration algorithm to minimize the designed cost function, the optimal control can be derived. The stability of the closed‐loop system is guaranteed via the Lyapunov theorem. Simulation results demonstrate the effectiveness of the proposed method compared to some existing results.