Electro-osmosis flow of Oldroyd-B nanofluid: ANN-based prediction of heat and mass transfer characteristics in microfluidic systems

• Investigated electro-osmotic effects on Oldroyd-B nanofluids to enhance fluid motion in micro/nano-scale engineering and biomedical applications. • Combined electrokinetic forces, viscoelasticity, and nanoparticle behavior to model heat and mass transfer in complex non-Newtonian flows. • Solved nonlinear boundary value problems using bvp4c and validated numerical results via neural network-based predictive analysis techniques. • It demonstrated that thermal and concentration profiles significantly depend on thermophoresis, Brownian motion, and activation energy parameters. • Highlighted magnetic field, heat source/sink, and Deborah number impacts on flow, offering insights for efficient thermal energy system design. The research on electro-osmosis in Oldroyd-B nanofluids aims to improve how fluids move in micro/nano-scale devices. The purpose is to explore the impact of electric fields on non-Newtonian nanofluid flow behaviour. This attempt requires understanding viscoelastic properties, nanoparticle dynamics, and electrokinetic effects for efficient energy and biomedical applications. The governing equations, which include electro-osmosis and activation energy effects, are solved numerically using the bvp4c method. A neural network-based approach is utilized for validation and predictive analysis of the flow and thermal profiles under varying physical parameters. Graphical and tabular results are provided for the distributions of skin friction coefficient, Nusselt number, and Sherwood number along the stretched surface. These results demonstrate substantial changes in temperature and velocity profiles attributable to magnetic field intensity, heat source/sink parameters, and nanoparticle characteristics. This research’s findings are quite like those of previous studies. The velocity profile exhibits an increase as the Deborah number (β2) and the electro-osmosis parameter (me) are elevated. The thermal profile exhibits an upward trend as the thermophoresis parameter (Nt) and space-dependent parameter (A*) increase. The thermal transfer efficiency was also significantly improved. The concentration profile is noted to rise due to elevated values of the activation energy parameter, whereas a contrasting behaviour is displayed when the Brownian motion parameter (Nb) is heightened. This phenomenon highlights the impacts of enhanced nanoparticle dispersion and mass transfer. The model that uses neural networks accurately predicts how fluids will flow, making it a valuable tool for studying complex fluid dynamics problems. This study provides important information on improving thermal energy systems, advancing extrusion processes, and enhancing material processing, which are all essential for effective HMT in industrial engineering.