| In the taditional way, when controlling the variable generation system, it needs to build a valid system model.Bacause the dynamics is uncertainity, and electric power model is complexity.It's not easy to make sure the model, so the brainpower control technology is used to wind generation system, which is based on the Fuzzy and Neural Network (FNN). In this paper, what is to work is to build a Fuzzy and Neural Network control system. It adjusts the rotate speed of synchronous generator and changes the pitch angle of the wind turbine, so that the wind turbine can extract maximum power from the wind energy, and at the same time to make sure to provide a receivable electric power quality.First, it need to analyze the variable wind generation and build the model, afterward, designing the controllers. The system has two controllers, one is Fuzzy excitation controller, which works by adjusting the torque to control the rotate speed, and when the wind speed is lower than rating speed, for tracking the wind speed to make the wind turbine extract a maximum power. Another is FNN, which works by adusting blade pitch when the wind speed is lower than rating value. To make sure the power retain rating value is its job. In the paper, it's first applied the genetic algorithm which is an important innovate point in the FNN controller.Final work is to imitate the model by the NN toolbox which is embodyed in the MATLAB/Simulink. The result is indicated that the system has a good robust, it can be used to realtime control. |
