Model-Free Predictive Current Control of a DFIG Using an Ultra-Local Model for Grid Synchronization and Power Regulation

Traditional model predictive control has difficulty in achieving satisfactory control performance when parameter identification is inaccurate or when a parameter changes. To solve this problem, this article proposed a new model-free predictive current control (MFPCC) scheme for a doubly fed induction generator (DFIG), which combines an ultra-local model with predictive current control (PCC). This method replaces the mathematical model of a DFIG with an ultra-local model and has strong parameter robustness. In addition, this method is suitable for the synchronization and power regulation processes of a DFIG, and it is also easy to apply to a DFIG with uncertain parameters. The proposed method is compared to traditional model-based deadbeat control with space vector modulation (SVM), and the presented experimental results confirm its superiority and effectiveness.