Robust Predictive Power-Winding Current Control of Grid-Connected Brushless Doubly-Fed Induction Generator Based on an Adaptive Ultralocal Model

Control-winding (CW) current control of a brushless doubly-fed induction generator (BDFIG) generally employs a dual-loop structure comprising an outer power loop and an inner CW current loop. This scheme is complex, depends on parameters and models, even necessitates flux observers and PI controllers. Thus, this study presents a theoretical derivation and simulation validation of the deadbeat direct power-winding (PW) current control. However, the results demonstrated that it cannot achieve stable closed-loop performance. To solve these problems, this article proposes an innovative robust predictive PW current control (RP-PWCC) based on an adaptive ultralocal model. The proposed RP-PWCC method directly calculates the PW current reference value from the complex power reference value, eliminating the PI controllers in CW current control methods and improving the dynamic process. To address the inability of traditional ultralocal models to directly control the PW current, the proposed RP-PWCC takes the CW current into account. All parameters of the improved ultralocal model can be obtained adaptively without design or tuning. The simulation and experimental results indicate that the proposed RP-PWCC shows better steady-state performance and quicker dynamic response than CW current control methods, and exhibits strong parameter robustness and computational efficiency.