Discrete-Time Neural Sliding Mode Indirect Power Control for a DFIG in Presence of Grid Disturbances

This paper proposes a Discrete-time Neural Sliding Mode Indirect Power Control (N-SMIPC) controller for a Doubly Fed Induction Generator (DFIG) connected to the grid. Under unbalanced grid voltage, the DFIG control strategies should be modified, becoming very complex. In addition, the DFIG under short-circuit conditions alters its behavior depending to the short-circuit type, and calculation of the equivalent circuit model is needed to redesign the controller. By using a Recurrent High Order Neural Network (RHONN) identifier, trained by an Extended Kalman Filter (EKF), an adequate model for the controlled system is obtained, which helps to propose a controller able to reject this disturbance, ensuring stability, and improving the Low Voltage Ride-Through (LVRT) capacity of the DFIG. Based on such identification, the proposed controller is used to track a desired Direct Current (DC) voltage desired value at the DC link, to keep constant the grid power factor controlled by the Grid Side Converter (GSC), and to control independently the rotor currents defined from the required stator powers, controlled by the Rotor Side Converter (RSC), under unbalanced grid conditions and short-circuits. The proposed controller is simulated by using SimPower toolbox of Matlab. The proposed controller effectiveness is confirmed even in presence of non-ideal grid conditions.