• Partially shaded photovoltaic panels. • Online data and real-time processing. • A Simulink digital model. • Experimental validation protocols. • Optimization of energy production. This article proposes an innovative experimental analysis with online control through a secure internet connection on the characterization process of photovoltaic solar panels at the University of West Attica in Greece with the objectives of evaluating their performance, optimizing the production system, detecting defects and especially improving the design. The methodological approach develops a numerical model in the Matlab Simulink environment, followed by experimental validation of the 3D dynamic behavior study for four 20 W photovoltaic panels, each comprising 36 monochromatic cells in series, connected respectively in parallel and individually for normal operation; then under partial shading conditions for different irradiances, temperatures, and maximum incidence angles for the parallel, series, and individual configurations. The results obtained show a root mean square error (RMSE) of 0.1346 and 0.6175 for the simple and parallel architecture models, respectively, under normal operating conditions. With a maximum current of 4.2 A and a voltage of 21.80 V under partial shading conditions for the SP (series–parallel) architecture, this experimental result of the 3D dynamic behavior positions the SP architecture as the one that best withstands the impact of partial shading. This allows for a better understanding of the physical mechanisms that determine the behavior of photovoltaic systems, their efficiency, and their production costs in terms of performance and reliability; and offers manufacturers the possibility of producing cells that are more resistant to high temperatures and have higher efficiency in the future. This work thus effectively contributes to the sustainable development of new innovative technologies in the process of controlling and validating the performance of photovoltaic systems, with the creation of online databases that can be used by industrial manufacturers.
Luc Vivien Assiene Mouodo (Sun,) studied this question.