Application of GA based fuzzy neural network predictive control in Active Power Filter

In accordance with the technique features of Active Power Filter control in harmonics suppression, e.g., time lag, non-linearity and frequent disturbance in work field, a fuzzy neural network predictive controller based on genetic algorithm is designed. By combining fuzzy control, neural network and predictive control, it can enhance self-studying, tracking and anti-interference capabilities of the algorithm, and the neural network can compensate with the limitation of conventional predictive control that based on linear model. With this algorithm the compensating current is controlled, and the simulation experimental curves show that the proposed method is more effective and feasible than PI control or digit adaptive control.