Performance Analysis Of PV Interfaced Neural Network Based Hybrid Active Power Filter

摘要

This paper presents a comparative analysis of neural network controlled PV interfaced hybrid active power filter designed for harmonic compensation for nonlinear load.The neural network has been chosen for reference current generation because of its fast adaptiveness, simple calculation and high accuracy to eliminate harmonics.This paper shows a novel approach to interface PV array to hybrid active power filter to keep the capacitor voltage stable. To obtain efficient output from PV Array Maximum power point tracking (MPPT) is employed in it. MPPT is able to extract maximum possible power from PV Array of change in atmospheric condition. Simulation and analysis of hybrid active power filter and PV Array is done under nonlinear load, sudden change in load and unbalanced load conditions. The detailed simulation results have been presented to validate the proposed methodology.