Experimental Comparison of Conventional and Intelligent MPPT methods for PV Systems operating under Partial Shading

Abstract

This work introduces an experimental comparison of four Maximum Power Point Tracking (MPPT) algorithms for photovoltaic (PV) systems: Incremental Conductance (INC), Perturb and Observe (P&O), Artificial Neural Network (ANN) and Fuzzy Logic Controller (FLC).The study assesses each method based on its response time, tracking efficiency, and oscillation behavior under both constant irradiation and partial shading conditions. Experiments were conducted on two identical PV test benches with control implemented via MATLAB/Simulink and dSPACE hardware. Results show that although the P&O method is easy to be implemented, it exhibits from significant fluctuations near the Maximum Power Point (MPP). INC achieves similar efficiency but with improved voltage stability. FLC eliminates oscillations and enhances efficiency, while ANN demonstrates superior robustness and highest efficiency under partial shading by maintaining a stable duty cycle. The findings highlight the advantages of intelligent MPPT algorithms, particularly FLC and ANN, in maximizing PV energy yield and operational stability in variable environmental conditions.