Enhancing Grid-Connected Photovoltaic System Performance Using Artificial Intelligence

Abstract

This thesis focuses on performance enhancement of a grid-connected photovoltaic (PV) system using fuzzy logic technique. Firstly, a general presentation of the PV system, including its principle operating, key components, and grid integration has been presented. Then, Perturb and Observe (P&O) algorithm is used to extract the Maximum Power Point (MPP), where its limitations are discussed. In addition, a fuzzy logic-based intelligent algorithm is proposed to control and manage the studied PV system energy for both non-MPPT and MPPT modes. Simulation results demonstrated that its performance is improved in terms of control, it can deliver the energy demanded by the grid manager without exceeding its maximum power. Finally, the robustness of the proposed strategy is successfully tested under various operating conditions.