Maximum Power Point Tracking Control of Wind Turbine Based on Neural Network Model Reference Adaptive Control

Abstract

To solve the problems of inaccurate wind speed, uncertainty, and interference in maximum power point tracking (MPPT), a novel MPPT control method according to neural network model reference adaptive control is put forward in this paper. First, under the premise of Betz’s theorem, the transmission chain model of a wind turbine is established, an effective wind velocity estimator is obtained by training the neural network through wind farm operation and maintenance data. Then, a composite controller composed of model reference adaptive controller (MRAC) and neural network controller is designed to make up for the uncertainties and disturbances in the system effectively. Finally, the update rate of the controller is adjusted according to Lyapunov stability theorem to ensure the asymptotic convergence of the variables in the system. MATLAB is used to simulate different wind speeds, and results show that compared with the traditional MRAC controller, the proposed method has better anti-interference ability and robustness, and can further enhance the wind energy utilization efficiency of the wind machine.