Machine learning analysis of tangent hyperbolic nanofluid with radiation and Arrhenius activation energy over falling cone under gravity

Abstract

This study is a machine learning investigation of the advance level nanofluidic coolant through a cone in a two-dimensional transitory boundary layer. The model accounts for both radiation absorption and the Arrhenius activation energy. Synthetic datasets from governing mathematical model are used in Artificial Intelligence (AI) based Levenberg Marquardt Back Propagation algorithm (LM-BP). Multiple scenarios of Tangent Hyperbolic Nanofluidic (THNF) coolant are framed with variation of influencing characteristics like Magnetic field M, power law index n, permeability k, Radiation absorption Q, Prandtl ratio Pr, Brownian motion Nb, Lewis number Le and Chemical reaction parameter γ. Convergence parameters of AI-based feed routing Neural Network computing is presented through graphs and numerical tables. Results indicate that flow slows when the Lorentz force and surface permeability grow, but it gets stronger when thermal absorption and momentum to thermal diffusivity ratio Pr increase. Meanwhile, the temperature increases when thermal absorption rises and drops when thermal to mass diffusivity ratio Le increases so that temperature falls for greater chemical reaction influence.