A Magnetic Flux Compensated Series Active Power Filter using Deadbeat Control based on Repetitive Predictor Theory

This paper presents a magnetic flux compensated series hybrid active power filter using deadbeat control based on repetitive predictor theory. Firstly, the primary winding of series transformer in the SHAPF would present leakage impedance to fundamental current and present magnetizing impedance to harmonic current which force the harmonic current to flow into passive power filter. In order to achieve the digital control of current signal tracking, this paper derives the deadbeat control based on repetitive predictor theory, which is especially suitable for the realization of digital control. The repetitive prediction strategy is used to predict the next beat current, which can effectively improve the prediction tracking accuracy at steady state and the response speed at transient state. The model of the system was built on MATLAB/Simulink which verified the effectiveness of the proposed control method.