Improving The Low-Voltage Ride-Through Capacity of a DFIG Using Neural Inverse Optimal Field Oriented Control

In this paper a Neural Inverse Optimal Control technique is developed for a Doubly Fed Induction Generator based variable speed wind turbine connected to an infinite bus through a transmission line. The proposed control scheme allows to track the desired values of the stator active and reactive power by means of the DFIG rotor currents control, and to improve the Low-Voltage Ride-Through capacity of the Doubly Fed Induction Generator. The proposed control scheme is based on a recurrent high order neural network identifier trained on-line with an extended Kalmen filter. This identifier is used to approximate the rotor currents under different grid scenarios, which helps to eliminate the effects of parameter variations and/or grid disturbances. The proposed control scheme is simulated by using SimPower toolbox of Matlab. The effectiveness of the proposed control scheme is validated under normal and abnormal grid conditions, and in presence of parameter variations. In addition the performance of the proposed controller is compared with the decoupled proportional-integral controller.