Research On Strategy Design And Simulation Of Active Disturbance Rejection Control For Wind Turbine Active Power Control

Wind energy is anticipated to become a backbone of the future energy system and play a vital role in the "net zero emission" climate action.With the progressive development of the global wind energy industry,the active power output of variable-speed and variable-pitch wind turbines needs to respond to power commands from the wind farm controller.In order to ensure the accuracy,reliability and stability of wind turbine active power control performance,it is necessary to develop advanced control strategies and optimization techniques to achieve wind turbine active power reference tracking and fatigue load mitigation.Firstly,this paper outlines the working principle of wind turbines,starting from the structure and main classification methods of wind turbines.On this basis,through the mechanism analysis of aerodynamic,mechanical,pitch servo,electrical and other subsystems,a 3-input and 2-output mechanism model of wind turbine is derived,and the complexity and many uncertainties of the system are analyzed,laying the foundation for the design and optimization of the control system.Secondly,in view of the constant power control problem of variable-speed variable-pitch wind turbines in the high wind speed region,an error-based active disturbance rejection pitch controller is proposed,and the stability of the closedloop system is proved according to the singular perturbation theory.On the basis of retaining the advantages of the original coyote optimization algorithm,an improved coyote optimization algorithm combining Levy flight and sinusoidal chaotic mapping is proposed,and it is applied to the performance optimization problem of the pitch control system.The benchmark function test specified by the IEEE Congress on Evolutionary Computation verifies that the improved coyote optimization algorithm has strong optimization ability,fast convergence speed and high repeatability.In addition,simulation results based on the OpenFAST 1.5 MW wind turbine model show that the error-based active disturbance rejection pitch controller can maintain the constant power output of the unit with smaller fluctuations.The frequency domain analysis results of the closed-loop system illustrate that the tracking and disturbance rejection performance of the system has strong robustnessFinally,aiming at the active power reference tracking issue of variable-speed variable-pitch wind turbines,an active power regulation strategy dual-activating the pitch action and the rotor kinetic energy is proposed.A multivariable active disturbance rejection active power controller is designed,and the convergence result of the closed-loop system are proved by combining the comparison principle of the ordinary differential equation.On this basis,the active power reference tracking performance optimization problem is described as a multi-objective optimization problem,and a multi-objective coronavirus herd immunity optimization algorithm is proposed based on non-dominated ranking and elite retention strategy.Seven typical test functions are selected to verify that the proposed multi-objective coronavirus herd immunity optimization algorithm has superior Pareto front approximation capability.In addition,simulation results based on the OpenFAST 5.0 MW wind turbine model show that the proposed the multivariable active disturbance rejection active power controller can shorten the rise time and settling time of power response,and the proposed active power regulation strategy can improve the power tracking accuracy,shorten the aggregate pitch travel,and alleviate the fatigue load of the unit.