An Improved Optimal Torque Control Based on Estimated Wind Speed for Wind Turbines

Although wind power technology has achieved huge development in recent decades, how to improve energy production by introducing effective wind speed information is still a hot topic for wind turbine control communities. In this paper, an improved Optimal Torque Control (OTC) based on a novel wind speed estimation algorithm is proposed for large-scale wind turbines. First, a machine learning algorithm based wind speed estimation framework is developed, the input variables of which are selected by using Mutual Information (MI) index. Further, the optimization of parameters for the developed wind speed estimation model is accomplished by the Particle Swarm Optimization (PSO) algorithm and the smoothness of the estimated wind speed is guaranteed by a low-pass filter with appropriate bandwidth. Thereafter, through introducing the estimated tracking error of optimal rotor speed, an improved OTC scheme is designed without changing traditional OTC’s basic structure. Finally, the feasibility and effectiveness of the presented control strategy are verified and demonstrated with a 1.5MW wind turbine on FAST (Fatigue, Aerodynamics, Structures, and Turbulence) and Simulink platform.