Artificial neural network-based robust tracking control for doubly fed induction generator used in wind energy conversion systems

This paper deals with a variable speed device to produce electrical energy on the power network, based on a doubly fed induction generator (DFIG). This machine is intended to equip nacelles of wind turbines. First, a mathematical model of the machine written in an appropriate d-q reference frame is established to investigate simulations. In order to control the power flowing between the stator of the DFIG and the power network, a control law is synthesized by using two types of controllers: Proportional-Integral (PI) controller and Neural Networks (NN) based controller. Their respective performances are compared by simulation in terms of power reference tracking and robustness against machine parameters variations.