Machine learning approach using Levenberg Marquardt artificial neural networks for magnetized ternary hybrid nanofluid across a permeable moving wedge

Résumé fourni par la source

The study used a mass-based ternary hybrid nanofluid model using the Levenberg Marquardt artificial network (THNF-LMANN) to analyze the magnetohydrodynamic (MHD) flow across a moving wedge. The Levenberg Marquardt artificial network (LMANN) is a novel technique in machine learning that has convergence stability via histogram representation, training, testing, and validation of the LMANN model using acquired data. Convective boundary conditions, heat radiation, along with wall porosity are all included in the study. The aggregate masses of volumetric concentration of the first, second, and third nanoparticles are considered rather than the individual masses of the nanoparticles (Titania, silver, and graphene) and the base fluid. The novelty of the present study is to investigate the effect of radiation and magnetic in THNF using LMANN. To reduce the number of variables and reduce the key THNF equations in terms of dimensionless ordinary differential equations, a self-similarity approach is employed. The statistical information for THNF-LMANN has been generated using the Runge–Kutta–Fehlberg technique (RK4). It is shown how important the new parameters affect temperature and velocity curves. Higher wedge angles are thought to directly affect the temperature and velocity distributions in the boundary layer, which reduces the depth of the velocity boundary layer and raises the velocity gradient at the wedge’s surface. It is scrutinized that the viscosity of the nanofluid increases with the concentration of nanomaterials. Furthermore, a decrease in the thickness of the boundary layer might result from an increase in wall suction. Interestingly, the hydrodynamic and thermal boundary layers both are increasing function of thermal radiation and Biot number. The present study is compared with the published work and good agreement is found.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.