Experimental Validation of Intelligent MPPT Control for Photovoltaic Energy Chain

This paper presents a comparative study of two Maximum Power Point Tracking (MPPT) algorithm techniques for a Photovoltaic (PV) system, which includes a PV generator, a DC-DC boost converter, and a resistive load. The study compares the performance of Artificial Neural Networks (ANN) and Perturb and Observe (P&O) algorithms in extracting maximum power under both stable and variable climatic conditions. To this end, simulation tests are performed using MATLAB Simulink, with a focus on energy efficiency and response time in different scenarios. The findings are validated through a hardware setup using the LAUNCHPAD-XL 28F379D and C2000 embedded coder. The results demonstrate that the ANN-based MPPT technique outperforms the traditional P&O method, particularly under rapidly changing environmental conditions, highlighting its superior efficiency in PV systems. Additionally, the ANN algorithm has been shown to exhibit enhanced adaptability to variable irradiance and temperature, thereby ensuring more stable and consistent power output across a broad spectrum of operating conditions.