This study numerically investigates the performance of an air-cooled photovoltaic thermal (PVT) collector integrated with NACA 8412 airfoils arranged at various positions in both flow (y) and transverse (x) directions, an application not previously explored in the literature. Using the finite element method, the effects of geometric parameters and a range of operational conditions, including airflow characteristics and environmental inputs, were evaluated. Artificial neural network (ANN) models were also developed to predict outlet air temperature, cell temperature, and pressure drop. Favorable energy and exergy performance was achieved with airfoil spacing of 50–70 mm (y-direction) and 35 mm (x-direction). To limit fan power consumption, the inlet air velocity should not exceed 5 m/s. The collector achieved efficiencies of 8.71–9.23 % (electrical), 50.34–70.85 % (thermal), 73.30–93.77 % (primary energy saving), and 10.23–10.69 % (exergy). The electricity production cost ranged from 0.158 to 0.668 $/kWh, and the exergoeconomic parameter varied between 4.10 and 17.55 kWh/$. Annual CO₂ mitigation reached up to 0.713 t (energy-based) and 0.275 t (exergy-based), depending on regional solar conditions. Two ANN model groups, one for single-output and one for multi-output predictions, were developed, both providing accurate and reliable results. This study offers a novel approach for enhancing and predicting PVT collector performance with airfoils.

This paper proposes and simulates a desiccant air cooling system integrated with a vapor compression cooling unit and a heat recovery unit for an office building in Çanakkale, Turkey, during the summer season. The required electrical energy for equipment of the proposed system is supplied by an Solid Oxide Fuel Cells (SOFC) unit using human waste as fuel. Moreover, some of the waste heat generated by the SOFC is used to regenerate the desiccant wheel. The simulation also includes the effects of three different refrigerants for the vapor compression cooling unit. Among the refrigerants, the highest electrical COP was obtained for the system using R1234ze(Z), which is 3.14% and 2.40% higher than the systems using R717 and R1233zd(E), respectively. Additionally, the system using R1234ze(Z) achieved electrical savings of 9.97% and 9.23% compared to the other systems. These electrical savings resulted in fuel savings of 1.19% and 0.90% for R1234ze(Z) compared to R717 and R1233zd(E), respectively. During the summer season, the electricity production from the existing SOFC unit met 82.00% of the total electricity consumption of the desiccant hybrid cooling system. Furthermore, a difference of 3984.56 kWh in primary energy consumption was identified between the desiccant hybrid cooling systems operating with the SOFC and without the SOFC during the summer season.

In this study, a desiccant-based hybrid cooling system supported by a vapor compression system and a heat recovery unit (rotary heat wheel) was analyzed from energetic and exergetic perspectives during the daily working hours of an office building in Istanbul to meet the desired comfort conditions. We focus on the impact of different refrigerants; namely R32, R1234yf, R290, R134a, R600a, R245fa, and R717, on hybrid rotary desiccant-vapor compression systems. While the highest electricity consumption was obtained in the system using R1234yf, the lowest electricity consumption was achieved with R717. However, with the effectiveness of the desiccant wheel, the best results were obtained for R1234yf with those pertinent to R717 at the other extreme. Considering the total electricity consumption of the system, the highest energetic and exergetic performance parameters were achieved with the use of R717 as the refrigerant. Compared to R1234yf, the daily average energetic performance parameters obtained with R717 increased by 22.3 % for COP _r, 21.8 % for COP _el, and 4.7 % for COP _th. Similarly, compared to R1234yf, the daily average exergetic performance parameters in R717 presented increases of 13.6 % for COP _x,el, 7.5 % for COP _x,th, and 8.1 % for η _x.