Comparative Analysis between Proportional-Integral and Artificial Neural Network Control of a Grid-Connected PV System

Abstract

This article presents a comprehensive analysis of the modeling and control techniques applied to a photovoltaic (PV) system that is connected to a three-phase grid. To successfully integrate the PV system with the electrical grid, an innovative and reliable controller has been designed and put into practice. The utilization of an artificial neural network (ANN) enables the system to optimize power extraction from the PV panels, benefiting from the ANN's resilience and swift response to varying conditions. Moreover, a robust proportional-integral (PI) control strategy is introduced to govern the grid-side operations. This strategy of action focuses on managing the injection of both active and reactive electricity into the grid as well as controlling the voltage of the DC bus. A series of detailed simulations were carried out evaluating the efficiency of the suggested control strategy in the MATLAB/SIMULINK. The results obtained from these simulations share insightful information on the effectiveness and efficiency of the control system in ensuring optimal operation and power management of the PV system within the grid-connected setup.

Keywords

• photovoltaic (pv)
• artificial neural network
• maximum power point tracking (mppt)
• proportional-integral (pi)
• perturb & observe(p&o)