The purpose of this study is to develop a unified numerical and intelligent predictive framework for analyzing unsteady mixed convection of a Carreau fluid over a rotating cone in the presence of magnetic, viscous dissipation and Soret–Dufour cross-diffusion effects. The work seeks to clarify the influence of these coupled mechanisms on momentum, heat and mass transfer characteristics and to establish an artificial neural network (ANN)-based surrogate model for rapid and accurate prediction of thermal performance metrics relevant to rotating machinery, thermal management devices and other engineering systems involving non-Newtonian double-diffusive transport.
An unsteady MHD mixed-convection model for Carreau fluid flow over a rotating cone is developed by incorporating viscous dissipation and Soret–Dufour cross-diffusion effects. Under boundary-layer assumptions, the governing partial differential equations are reduced to coupled nonlinear ordinary differential equations using similarity transformations. These equations are solved numerically with MATLAB’s BVP4C solver under suitable boundary conditions, and the numerical results are validated against published benchmark data. The generated high-fidelity data set is subsequently used to construct a feed-forward ANN trained by the Levenberg–Marquardt algorithm to obtain fast and reliable predictions of the principal heat- and mass-transfer characteristics.
This study reveals that heat and mass transfer in Carreau fluid flow over a rotating cone are highly sensitive to magnetic, elastic, thermal and cross-diffusion effects. Magnetic forcing intensifies the velocity field, while the Dufour effect raises the temperature distribution within the boundary layer. Higher Prandtl, Schmidt and Soret numbers strengthen the surface heat and mass transfer rates, whereas larger Weissenberg and Eckert numbers weaken the Nusselt number. Variations in buoyancy-related parameters produce notable rotational flow modulation. The ANN surrogate shows excellent agreement with the numerical solutions, with a regression coefficient exceeding 0.999 and MSE on the order of 10–6.
The study’s value lies in presenting a unified analysis of unsteady mixed convection of a Carreau fluid over a rotating cone under the combined effects of magnetic forcing, viscous dissipation and Soret–Dufour double diffusion. While these mechanisms have often been studied separately, their coupled influence in this rotating-cone configuration is addressed here within a single framework. The work further adds value by integrating high-fidelity numerical solutions with an ANN surrogate for rapid prediction of key transport quantities, offering both physical insight and a practical predictive tool for thermal design and optimization in rotating engineering systems.
