The computational and neural network analysis of the Williamson hybrid nanofluid for solar energy applications

Purpose The purpose of this study is to explore the efficiency in the solar energy system using the TiO2 and Ag hybrid nanofluid. In this work, the focus has been on solar energy efficiency. The theoretical, computational and artificial neural network (ANN) combination has been used to determine the heat transfer efficiency in terms of solar radiation and its utilization in the solar system. Design/methodology/approach The thin layer of the TiO2–Ag hybrid nanofluid has been used to consider the 3D formation of the sheet. The finite element method (FEM) is used to solve the transformed nonlinear governing equations numerically. ANN and Levenberg–Marquardt neural network models are used to verify the numerical results further. Findings Solar radiation increases the thermal energy of the fluid, leading to higher temperature profiles and improving energy efficiency. It has been found that the higher the nanoparticle volume fraction, up to 5%, the higher the heat transfer rate and the hybrid nanofluid has superior performance over other fluids. The statistical comparisons demonstrate that the proposed solution methodology for complex radiative non-Newtonian hybrid nanofluid flows is reliable due to an excellent match between numerical and intelligent predictions. Originality/value To the best of the authors’ knowledge, a first analysis of fluid flow over a 3D sheet under Williamson fluid conditions has been conducted. The TiO2 and Ag hybrid nanofluid based on ethylene glycol has been used for the 3D model, which is in the class of the Williamson fluid model. The FEM and ANN combined create a novel approach.