Supervised machine learning computing paradigm to measure melting and dissipative effects in entropy induced Darcy–Forchheimer flow with ternary-hybrid nanofluids

Abstract

This study investigates the change in thermophysical characteristics of Darcy–Forchheimer flow of tri-hybrid nanofluid (THNF) over stretchable porous skin. Machine learning (ML) technique is employed to compare thermal exchange of water-based hybrid nanofluid (HNF), composed of aluminum oxide (Al2O3) and silicon dioxide (SiO2) with THNF by introducing copper oxide (CuO) nanoparticles in HNF. Flow rate profile, dissipation effect, and entropy generation are analyzed with novel, time, and cost-effective evaluation technique. By using similarity parameters, the governing system of partial differential equations (PDEs) is transformed to a system of ordinary differential equations (ODEs). Finite difference method is used for dataset generation by the aid of Python environment, then neural fitting computation technique is utilized for optimized results. Turbulence and randomness are incorporated in the present nonlinear mathematical model which is tackled by artificial intelligence (AI)-based Levenberg Marquardt Neural-Network Algorithms (LMNA), to produce optimized outputs. Velocity, temperature, and entropy profile are evaluated against various influencing factors, such as Forchheimer value, porosity, melting parameter, radiation, Prandtl, Eckert, and Brinkmann coefficient. Numerical and AI-generated outputs and errors graphs with complete neural network training scheme, are portrayed to draw the effective concluding inferences. Enhanced results are beneficial for industrial and engineering production units, for efficient thermal management.