Thermal analysis of non-Newtonian tetra-hybrid nanofluid with lubrication layer effects under mhd and radiation using artificial-intelligence and response surface methodology

Effective heat regulation and decreased frictional losses are essential in sophisticated lubrication systems used in both micro- and macro-scale engineering applications. This study investigates the stagnation-point flow of a non-Newtonian Jeffrey tetra-hybrid nanofluid, comprising Al₂O₃, Cu, TiO₂, and SiO₂ nanoparticles suspended in sodium alginate, with an emphasis on improving heat transfer and minimising wall shear stress within the framework of magnetohydrodynamics, Joule heating, thermal radiation, and slip phenomena. To solve the governing partial differential equations (PDEs) numerically, MATLAB bvp4c is used to convert them into a coupled system of ordinary differential equations using appropriate similarity variables. To enhance prediction and optimisation, a Scaled Conjugate Gradient-based artificial neural network (ANN) and Response Surface Methodology (RSM) are further used. The findings indicate that augmenting the slip parameter and nanoparticle volume percentage diminishes the velocity profile, but a greater Deborah number and magnetic parameter amplify it. As the radiation parameter and Eckert number increase, the temperature field also increases. The ANN model has robust predictive ability, shown by correlation coefficients over 0.99 and a minimal mean squared error. RSM research shows that the Deborah number and radiation parameter have the most effects on skin friction and heat transmission, respectively. The ideal circumstances provide a minimum skin-friction coefficient of 0.0919 at ( ϕ 4 = 0.02 , β = 1.5 , α = 2.0 ) and a maximum Nusselt number of 15.6426 at ϕ 4 = 0.02 , R d = 0.6 , E c = 0.1 ). By lowering shear resistance and increasing heat dissipation, our results demonstrate that tetra-hybrid sodium alginate nanolubricants may improve lubrication performance.