Feedback-Feedforward PI-Type Iterative Learning Control Strategy for Hybrid Active Power Filter With Injection Circuit

In this paper, the configuration characteristic of Hybrid active power filter (APF) with injection circuit (IHAPF) is analyzed, as well as its current closed-loop control model is established. Because of the character of reperiod of current harmonics in steady-load power system, the iterative learning control algorithm based on the PI-type learning law is presented. The systemic robustness is enhanced by using a forgetting factor. In order to improve the dynamic performance of a control system, a feedforward based on the D-type learning law of referenced current error by fuzzy reasoning is proposed. The system of the IHAPF with the proposed control strategy has been applied in a steel plant in Guangxi, China. Simulation and industrial application results show that the IHAPF with the proposed control method is not only easy to calculate and implement but also very effective in improving the performance of the filter. Meanwhile, IHAPF shows great promise in reducing harmonics and improving power factor with a relatively low capacity of APF.