Experimental and numerical modeling of photovoltaic modules temperature under varying ambient conditions

Abstract

In this work, comprehensive three-dimensional computational fluid dynamics simulation, of fluid flow and heat transfer phenomena around a free-standing polycrystalline silicon photovoltaic module is carried out. The objective is to provide accurate calculation of module’s temperature as a key parameter to estimate its power output. Therefore, experiments were conducted at the university of El Oued, south-east Algeria, to collect the necessary dataset for simulations. Considering different heat transfer mechanisms, modeling absorbed solar energy within the cells, and after mesh refinement study and model validation, simulations were performed and different parameters have been investigated. Results show that more accurate module temperature (Tback) estimation can be achieved based on numerical simulations. It was also found that numerical simulation overcome other models from literature and provides better results, achieving an R2 of 0.995 and a mean absolute error (MAE) of 0.822. Results also indicate that, solar radiation (G), ambient temperature (Ta) and wind speed (Ws) tend to have the major impact on Tback, an increase of 100W/m2 in G can produce an increase of 4°C in Tback at low wind speeds, and about 2.4°C for relatively higher Ws. Ta also tends to yield linear increase in Tback, expecting 5.8°C rise, for 6°C increase in Ta at 700W/m2 and 1m/s of solar radiation and wind speed, respectively. Additionally, a regression-based model was proposed for engineering applications, providing accurate results with an R2 of 0.989, a MAE of 1.009, which is 10% more accurate than the best model from literature.