Investigations on Dynamic Performance of Hybrid Shunt Active Power Filter Strategy for Power Quality Enhancement Using PI Controller and Neural Learning Algorithm for Three Phase Three Wire Distribution System

Abstract

this paper investigates the dynamic behaviour of the hybrid shunt active power filter strategy for mitigating the voltage harmonics and the current imperfections for three phase three wire distribution system. The unit vector template generation control technique is designed as a DC link voltage controller for the hybrid shunt active power filter strategy. The PI controller is designed to estimate the rated source current magnitude from the error between the actual and reference DC link voltage. The transient period and peak overshoot of DC link voltage using PI controller is observed to be high in initial and load change conditions. This paper explores the application of artificial intelligence on reducing the transient period and peak overshoot of DC link voltage. In this paper, a neural learning algorithm is proposed for the DC link voltage of the shunt active power filter strategy. To validate the efficiency and reliability of the proposed system is designed and tested in the MATLAB / Simulink environment.