Burgos, España
Madrid, España
There are many control challenges in wind turbines: controlling the generator speed, blade angle adjustment (pitch control), and the rotation of the entire wind turbine (yaw control). In this work a neuro-control strategy is proposed to control the pitch angle of the wind turbine. The control architecture is based on an RBF neural network and an on-line learning algorithm. The neural network is not pre-trained but it learns from the system response (power output) in an unsupervised way. Simulation results on a small wind turbine show how the controller is able to stabilize the power output around the rated value for different wind ranges. The controller has been compared with a PID regulator with encouraging results.
© 2001-2024 Fundación Dialnet · Todos los derechos reservados