A Sensorless Adaptive Grid Side Control Approach for Doubly Fed Induction Generator (DFIG)

In this paper, a system identification based adaptive controller is proposed for a grid-connected doubly fed induction generator (DFIG). Recursive least squares (RLS) estimation based system identification provides the auto-regressive moving average exogenous (ARMAX) model of the system. A minimum variance control (MVC) architecture has been used that minimizes the variance of the system output from its reference setpoint utilizing the identified model of the system. The proposed architecture simplifies the Grid Side Converter (GSC) control loop of DFIG and overrides the issues in conventional techniques due to bad sensors and effects due to parameters variations. The stability properties of the proposed controller are discussed. Theoretical results are validated using MATLAB/Simulink and real-time simulation platform for 1.5 MW wind turbine system. It is observed that the proposed architecture provides better performance quantitatively discussed as in Table II.