Model-Free Predictive Current Control of Doubly Fed Induction Generator

When the machine parameters of doubly fed induction generator (DFIG) are inaccurate, the control performance will be weakened with the traditional control method. In this paper, a new model-free predictive current control (MFPCC) is proposed to solve this problem. This method is mainly to replace the mathematical model of the machine with an ultra-local model. To further improve the control performance, the value of the disturbance is observed through an extended state observer (ESO). Regardless of the accuracy of the machine parameters, the proposed method has good steady-state and dynamic performance because of the fact that no machine parameters are included in the final control expression. The effectiveness of the proposed MFPCC scheme based on an ESO is confirmed by the simulation and experimental results compared to direct power control based on space vector modulation (DPC-SVM).