Data-driven computational framework for convective energy storage of Sisko trihybrid nanomaterials with cross-diffusion processes impinging over a radially stretchable surface

The axisymmetric flow of an incompressible non-Newtonian fluid over a stretchable sheet has important industrial and technological uses in modern era. The current study explores the axisymmetric magnetohydrodynamic (MHD) axisymmetric boundary layer flow (BLF), heat, and mass transfer of a Sisko trihybrid nanofluid (NF) over a radially stretching sheet, extending the foundational BLF by incorporating several novel physical features. This research is motivated by the broad applications and the increasing need for predictive intelligent modelling of NF systems under multi-physics conditions. Three different nanomaterials Aluminum oxide (Al 2 O 3 ), Titanium (TiO 2 ), and Silica dioxide (SiO 2 ) are dispersed in ethylene glycol (C 2 H 2 O 2 ) which serves as a base fluid to create the trihybrid NF. Thermal radiation, Joule heating, viscous dissipation, cross-diffusion processes, first-order chemical reactions, velocity slip, and convective heat and mass transfer boundary conditions are all taken into consideration. The governing BLF equations are generated under the BLF estimations. The Runge-Kutta-Fehlberg (RKF-45) approach is used to solve the corresponding nonlinear system numerically. An artificial neural network (ANN) model is trained using the Levenberg-Marquardt backpropagation (LMB) algorithm to address the computational overhead of recurrent computational solutions across large parameter spaces. Important results show that the Dufour number, thermal radiation, and thermal Biot number all increases the thickness of the thermal BL. An increasing the Sisko fluid parameter improves the velocity profile while dampening temperature and concentration distributions. The magnetic parameter detract the flow field. The surface drag, heat and mass transfer improved with fluid parameter and power index.