Energy and Statistical Assessment of Solar Air Conditioning System by Using PV/T Module and Nanofluids

Renewable energy technologies are increasingly used to ensure environmental sustainability and mitigate fuel cost escalation. Among them, photovoltaic thermal (PV/T) collectors stand out for their ability to generate both heat and electricity simultaneously. In this study, a PV/T system was modeled in TRNSYS to provide winter space heating for a hotel building located in Cairo. Four nanofluid types (MWCNT, Fe₃O₄, CuO, and Al₂O₃) were investigated at concentrations of 1–5%. The MWCNT-based nanofluid exhibited the best performance, particularly at concentrations of 5%, where the cell temperature was significantly reduced compared to water, resulting in improved cooling, electrical output of 1.27 kW, and thermal efficiency of 80.75%. A shallow neural network with small data set (50 points) was used to predict the relation between two important outputs in solar field with nanofluids which are electric power and electric efficiency to the main input parameters (solar irradiance (W/m2), ambient temperature (oC), density (kg/m3), and specific heat (J/kg.K)). Levenberg-Marquardt back propagation algorithm was used in the two simulation cases.