ANF-MAP Based Multipurpose Control Approach for Single-Stage Grid-Connected PV System

A control technique based on an adaptive notch filter-modified affine projection (ANF-MAP) algorithm is implemented for a single-stage grid-connected photovoltaic (GCPV) system designed for residential applications. The residential loads are mainly fed by photovoltaic (PV), and absence of PV is compensated by the grid. However, the dominance of non-linear load affects the grid power quality. To address these issues, MAP generates a sinusoidal reference current by estimating active part of the weight component of the load current. Furthermore, the performance of MAP during nonideal grid voltage conditions is improved by utilizing an ANF in the MAP to estimate the sinusoidal synchronizing unit templates. Moreover, the proposed control approach effectively handles PV intermittency, load variation, unbalanced loading, and grid voltage sag conditions. Additionally, the performance of ANF-MAP is tested with affine projection, least mean square, and least mean forth algorithms in dynamic condition with ideal grid condition in the MATLAB/Simulink platform and experimentally with phase lock loop, MAP and second-order generalized integrator frequency lock loop techniques in nonideal grid condition. Finally, the system responses with the proposed control technique are validated and justified by IEEE 519 and 1547 standard by developing the laboratory prototype.