Thermo-solutal convective flow of nanofluid with Marangoni convection: An artificial neural network study considering thermophoresis and thermal radiation effects

Abstract

A neural network backpropagation approach combined with the AI-integrated Levenberg-Marquardt algorithm provides a comprehensive analysis of the thermal radiation on thermos-solutal Marangoni convective flow of nanofluid with thermophoresis influenced by Lorentz force effects. In order to explain mass and heat transmission and fluid flow, non-linear, coupled PDE (partial differential equations) are transformed into ODE (ordinary differential equations) with similarity scaling. The Bvp4c method is then used to resolve these equations numerically. The suggested model has important uses in many industrial and technical processes where mass transfer and heat are important factors. It is used in sophisticated material processing, cooling systems, and microfluidic devices where accurate thermal control is crucial. The research holds special significance for cooling mechanisms based on nanotechnology, including nuclear reactor safety and semiconductor chip cooling. Furthermore, the model's incorporation of Marangoni convection, thermophoresis, and thermal radiation effects renders it applicable to solar energy harvesting, biomedical engineering, and aerospace technology, all of which depend on effective heat dissipation and fluid stability. The concentration profile decreases as increase the thermophoretic parameter rise.