The present investigation explores the implications of thermal radiation on the MHD flow and heat transfer features of a hybrid nanofluid configured by a rotating disk. In the framework, aluminum Al 2 O 3 and copper Cu nanoparticles are taken into account to form a hybrid nanofluid with fundamental fluid water (60%) and ethylene glycol (40%). To compute the governing set of partial differential equations, a near-surface framework is developed to validate the flow problem. The appropriate similarity transformation is applied to convert the set of partial differential equations into an ordinary differential equation. The obtained series of ordinary differential equations is solved numerically by using the bvp4c built-in function in the computational software MATLAB. The impact of significant physical parameters, like as Prandtl number, Reynolds number and nanoparticle volume fraction, on the skin friction and Nusselt number against the flow and thermal distribution. To facilitate the numerical results, an artificial neural network model is established to evaluate the flow and thermal properties of the flow. The minimal prediction errors and dependability of the model are confirmed by the ANN findings, which are remarkably consistent with the numerical data. Heat transfer performance improved dramatically with an increase of 18–22% in the local Nusselt number when the nanoparticle volume fraction was raised from 0 to 2%. A 10–14% rise in the skin friction coefficient, which indicates stronger shear effects along the rotating surface, followed this improvement. The thermal radiation parameter had the opposite effect; larger values reduced the heat transfer rate by around 7–11%. In contrast, higher Reynolds numbers increased both radial and tangential velocity components by around 15%, indicating a more energetic flow field. The ANN model correctly matched the numerical trends, with a mean absolute error of less than 0.5% and a coefficient of determination greater than 0.999, demonstrating its reliability and predictability.
Manzoor et al. (Mon,) studied this question.