Control of Doubly Fed Induction Generator of Variable Speed Wind Turbine System using Neural Network

Abstract

This paper presents artificial intelligent controller based on Artificial Neural Network (ANN) for Doubly Fed Induction Generator (DFIG) wind turbine system. For the purpose of maintaining constant active power based on PI controller concepts. The ANN controller is based on rotor loop design for a variable speed wind turbine. An active power and reactive control law are created for controlling the rotor voltages of DFIG, and error signals for both the two powers act as the input to ANN Controller; this approach eliminates the use of current inner-loops and estimation of any flux components. For the purpose of comparison PI-based vector controller is developed. The simulation results based on dynamic performance indicate the superiority of ANN based rotor loop controller over the conventional PI controller.