Robust Neural Control of Wind Turbine Based Doubly Fed Induction Generator and NPC Three Level Inverter

Abstract

This paper presents dynamic modeling and control of Doubly Fed Induction Generator (DFIG) based on wind turbine systems, where the stator of DFIG is directly connected to the grid and the rotor was fed by a three level PWM NPC inverter. The active and reactive power control of the DFIG is based on the feedback technique by vector control method by using a classical regulator of Proportional-Integral (PI) type which allows us, in association with the looping of powers, to obtain an efficient and robust system. This approach is a very attractive solution for devices using DFIG as wind energy conversion systems; because, it is a simple, practical implementation, commonly applied in the wind turbine industry and it presents very acceptable performance, However, this control approach has certain limitations and has several causes, vector command with NPC three-level inverter pulse width modulation (PWM) is used to control the reactive power and active power of the generator. Then, use the neural network design to replace the traditional proportional-integral (PI) controller. Finally, the Matlab/Simulink software is used for simulation to prove the effectiveness of the command strategy.