ANN‑based analysis of oscillatory Jeffrey hybrid nanofluid with nonlinear thermal radiation and Cattaneo–Christov flux model

This study consists of investigating the oscillatory magnetohydrodynamics heat and mass transfer of a Jeffrey hybrid nanofluid composed of Ti 6 Al 4 V and ZnO nanoparticles dispersed in blood over a permeable, elastically oscillating surface embedded in a saturated porous medium. The model incorporates nonlinear thermal radiation and Cattaneo–Christov heat and mass fluxes to capture finite-speed transport, together with thermophoresis and a homogeneous first-order chemical reaction. After attaining dimensionless form of the governing equations, the homotopy analysis method (HAM) is implemented. In order to achieve the accuracy of scheme, the convergence region has been specified. An artificial neural network (ANN) surrogate is then trained on the HAM database to predict the skin-friction coefficient, Nusselt number, and Sherwood number from the governing dimensionless groups, achieving high predictive accuracy (high R², low MSE) and enabling rapid parametric sweeps. The analysis indicates that stronger magnetic fields damp velocity via Lorentz forces; increasing Biot and radiation parameters enhance surface heat transfer; thermal relaxation reduces temperature owing to finite-speed effects; and thermophoresis and chemical reaction lower near-wall concentration. The combined HAM–ANN workflow offers a transparent, efficient route to fast, accurate performance maps for hybrid blood-based nanofluids in oscillatory thermal systems.