Cognitive neural modeling of MHD maxwell nanoliquid flow with variable thermosolutal properties in darcy–forchheimer porous Media

Artificial neural networks are invaluable for addressing complex and nonlinear mathematical problems due to their exceptional effectiveness in handling such challenges. Artificial Neural Networks provide a distinctive computational framework that proves highly useful in various systems such as lubrication behavior, materials and chemical engineering, life sciences and surface interactions with computational fluid dynamics. The main objective of this study is to explore the potential of Levenberg-Marquardt method for evaluating fluid flow and heat transmission in viscoelastic nano-lubricant flow over a stretchable surface using back-propagation artificial neural networks. The implications of thermal and solutal fluxes on temperature and concentration gradients are addressed. The Maxwell nanomaterial fluid flow over a tribological surface is described using nonlinear partial differential equations. After transforming the partial differential equations into ordinary differential equations, the bvp4c approach is employed for numerical solutions. The Levenberg-Marquardt algorithm is implemented on reference datasets using back-propagation artificial neural networks, allocating 70 % training, 15 % testing and 15 % validation. The approach is verified through mean squared error, error histograms and in-depth regression investigation. Flow and thermosolutal characteristics have been explored visually and statistically demonstrated. The fluid's temperature and concentration rise with thermal conductivity and mass diffusivity parameters, while velocity declines with Deborah number, Forchheimer inertia coefficient and mixed convection parameters. A minimal absolute error nearing zero demonstrates consistency, highlighting the effectiveness of the proposed approach. The present study demonstrates the potential of AI-driven models in improving tribological surface systems, allowing for precise predictions of hydrodynamic and thermal behavior in nano-lubricated coatings surface subjected to exacerbated physical domains. These outcomes offer insights for optimizing the layout as well as performance of future coatings interfaces in high-efficiency applications in engineering systems.