Implementation of an Optimal Control based on ANN and LADRC for Two-Stage Single-Phase Photovoltaic Grid-Connected Inverter

Abstract

In this paper, a two-stage single-phase grid-connected photovoltaic (PV) configuration, is presented. To extract the maximum power from the photovoltaic (PV) system, inject a clean and stable power into the grid and improve the power quality at the point of common coupling (PCC), artificial neural network (ANN) based algorithms, are employed. A boost converter is employed to track the Maximum Power Point (MPP) of the PV system and an ANN trained with the Perturb and Observe (P&O) algorithm is used. The inverter is controlled using an Adaptive Linear Neuron (ADALINE) technique which is reinforced by Linear Active Disturbance Rejection Control (LADRC) strategy to guarantee stable and uninterruptible power into the grid as well as power quality improvement. The performance of the two-stage single-phase configuration and their control strategies are tested using MATLAB/Simulink under variation of load, solar conditions change, and nonlinear load.