An Improved Deadbeat Control Strategy with Command Current Fractional Order Rapid Repetitive Prediction Method for Active Power Filter

This paper presents improvements in conventional deadbeat control and command current prediction algorithms. To address the inherent one-cycle delay in traditional deadbeat control, an improved strategy is proposed that predicts the command voltage signal one control cycle in advance, enhancing steady-state tracking performance. Additionally, to tackle the issues of one-cycle dynamic delay and insufficient adaptability to frequency variations in traditional command current repetitive prediction methods, a fractional-order rapid repetitive prediction method for command current is proposed, integrating the concept of fast repetitive control. This improves the accuracy of the APF's compensation of harmonic currents. By using the conventional and improved deadbeat control strategies in the APF, the total harmonic distortion (THD) of the grid-side current was reduced to 2.43% and 2.02% respectively. This represents an approximately 16.7% improvement in compensation performance using the improved strategy. Through simulation and hardware-in-the-loop experiments, the effectiveness of the proposed control strategies is demonstrated.