This study investigates the performance of diode-based photovoltaic (PV) modules models by analyzing their effectiveness in predicting the electric behviour under conventional solar irradiation and low-concentration photovoltaic (LCPV) conditions. The parameters of one-diode (1-DM), two-diode (2-DM), three-diode (3-DM) and four-diode models (4-DM) are first extracted using the particle swarm optimization technique (PSO) and validated through a comparative analysis with experimental measurements carried out on a PV module (ISOFOTON 106 W-12 V) in real-world temperature and irradiation conditions of 27.2°C and 755 W/m², respectively. The findings reveal that the 4-DM exhibits the minimum deviation from experimental data in predicting key performance metrics such as short-circuit current (Isc), open-circuit voltage (Voc), and maximum output power (Pm). However, this increased accuracy comes at the cost of higher computational complexity in optimizing the 4-DM’s parameters. The studies carried out under several low-concentration photovoltaic conditions show clearly the limitation of the 1-DM in terms of predicted (Pm), efficiency, and fill factor (FF). Indeed, the gaps in the obtained values of efficiency and FF with respect to the 4-DM increase with the concentration ratio and reach 0.74% and 0.04, respectively, at 3 suns. The performances obtained with the 2-DM and 3-DM remain stable and close to those of the 4-DM with constant gaps in the obtained values of efficiency and FF, remaining close to 0.1% and 0.01, respectively, regardless of the concentration ratio. The insights gained from this work underscore the significance of selecting an appropriate PV model for LCPV systems, balancing accuracy and computational efficiency.
Kechiche et al. (Tue,) studied this question.