Aerothermal sensitivity and heat transfer analysis of hybrid nanofluid flow for solar-powered aircraft thermal management under realistic solar irradiance

Solar energy is a major renewable heat source used in photovoltaic systems, solar thermal collectors, water purification and related technologies. In recent years, the coupling of solar radiation with nanoparticle-based working fluids has gained attention as a practical route to enhance heat transfer for aeronautical and energy-limited platforms. Solar-powered aircraft operate under strict energy and weight constraints; therefore, thermal losses and irreversible entropy generation can reduce endurance and system reliability. In wing-integrated thermal management and PTSC-based concepts, the working fluid must deliver high heat-transfer performance while limiting dissipative and radiative losses. Motivated by these requirements, the present study investigates a second-grade hybrid nanofluid (HNFs) model to quantify heat-transfer enhancement, entropy generation and parameter sensitivity for flow in a Parabolic Trough Solar Collector (PTSC) embedded within the wing of a solar-powered aircraft. The primary focus of this study is to investigate the kerosene oil-based hybrid nanofluid flow over a stretching sheet within solar wings. Heat transfer efficiency is analyzed under the influence of solar radiation, magnetic fields, heat sources, viscous dissipation and Ohmic heating. Non-Newtonian HNFs flow phenomena are enhanced by radiant radiation and external heat sources. By applying the non-similar transformations, the governing equations are simplified into nonlinear dimensionless ODEs and then tackled numerically using the bvp4c function. Furthermore, this study aims to improve thermal performance and minimize energy in solar aircraft via the analysis of response quantities across a stretched sheet with the use of artificial neural networks (ANNs) and response surface methodology (RSM). Sensitivity analysis of key parameters on thermal transport and entropy generation is also analyzed. The outcome reveals that R2 and Adj (R2) for heat transfer rates of HNFs are 96.77% and 97.57%, respectively, as well as for entropy generation, the values are 97.96% and 96.99%, which shows a close correlation and high model accuracy for this model. The highest sensitivity values for thermal transport are recorded as 0.44859 for the radiation parameter, −0.06973 for the magnetic number and −0.30105 for the heat generation parameter. Furthermore, an ANN was utilized to predict thermal transfer rates and entropy. An analysis was conducted to examine the probabilistic instance distribution using error histograms, continuous curve modeling at each progressive epoch and a review of adaptive control parameters in Artificial Intelligence (AI)-based neural network computations. The HNFs provides the highest heat-transfer enhancement, where the Eckert number increases Nux by 20.94% (NFs) and 63.63% (HNFs) compared with the base fluid at Ec=0.4[jls-end-space/], with HNFs remaining significantly higher than the base fluid. However, this improvement is accompanied by higher wall shear, as the skin-friction magnitude increases by about 56.94–60.40% for the hybrid nanofluid (and ≈22.60–22.64% for NFs) relative to the base fluid across the tested conditions.