Numerical analysis of MHD Jeffrey hybrid nanofluid flow over a solar curved sheet using ANN model

Abstract The conversion of solar radiation into thermal energy has gained increasing attention due to the growing demand for renewable sources of heat and electricity. Nanofluids, owing to their enhanced heat transfer capabilities, play a significant role in improving the efficiency of solar thermal systems. In this study, the flow of silicone oil containing diamond and silicon dioxide nanoparticles over a curved extended permeable sheet is investigated in the presence of Darcy–Forchheimer porous medium, thermal radiation, and Lorentz force. The non-Newtonian behavior of the working fluid is modeled using the Jeffrey fluid model. The governing flow equations are transformed into ordinary differential equations (ODEs) and solved numerically using the MATLAB bvp4c solver. Furthermore, an intelligent computational approach based on the Levenberg–Marquardt algorithm combined with a multilayer perceptron (MLP) feed-forward backpropagation artificial neural network is employed. The effects of key parameters, including the Deborah number, injection/suction parameter, permeability parameter, Forchheimer number, Hartmann number, curvature parameter, Prandtl number, Eckert number, and heat generation/absorption parameter, are analyzed in terms of pressure, velocity, temperature, and heat transfer rate. The results indicate that porous resistance and magnetic effects significantly influence boundary layer formation and heat extraction efficiency, showing that optimally adjusted hybrid nanofluids can greatly enhance thermal transfer in porous solar collectors and curved absorber surfaces, thus providing valuable insights for the design of advanced solar thermal systems.