دراسة مقارنة بين متحكمي الضبابي والعصبوني لضبط زوايا ميل شفرات التوربين الريحي

Abstract

Wind turbines face significant challenges due to rapid and unpredictable changes in wind speed, which adversely affect the stability of the electrical grid and turbine efficiency. This research aims to design and evaluate intelligent control systems for the pitch angles of fixed-speed wind turbine blades, based on squirrel cage induction generators, to enhance the dynamic stability of the electrical grid.

Two types of control systems were designed using MATLAB/Simulink. The systems rely on the difference between the measured power and the reference power as the input signal to the controller, while the output represents the required blade pitch angle.

The first controller is a Fuzzy Logic Controller (FLC), which is characterized by its ability to handle imprecise or ambiguous data, such as variations in wind speed. It was designed with two inputs: error and rate of change of error.

The second controller is based on a Neural Network (NN), which is distinguished by its self-learning ability. The network was generated and trained using the nnstart tool based on the results from the fuzzy controller.

Simulation results conducted during a sudden wind disturbance demonstrated the superiority of the neural network in improving the system's dynamic stability. The performance indicator P_smooth decreased from 9.6 (p.u) with the fuzzy controller to 8 (p.u) with the neural network controller. This indicates that output power fluctuations were further reduced, enhancing the dynamic stability and improving the electrical system's ability to return to a stable state after being subjected to a wind disturbance.