A Breakthrough in Penta-Hybrid Nanofluid Flow Modeling for Heat Transfer Enhancement in a Spatially Dependent Magnetic Field: Machine Learning Approach

Abstract

A versatile penta-hybrid nanofluid has been successfully developed by combining silver (Ag), single-walled carbon nanotubes (SWCNTs), titanium dioxide (TiO₂), copper (Cu), and iron oxide (Fe₃O₄) nanoparticles with a base fluid. This nanofluid is utilized in a range of advanced applications, including coatings, sensors, energy storage, water purification, enhanced heat transfer, biomedical uses, and lubricants. The synergistic properties of these nanoparticles significantly enhance the performance of the base fluid, offering substantial benefits across various industries. Therefore, this study delves into the influence of localized magnetic fields, augmented by machine learning, on vortex dynamics under the light of penta-hybrid nanofluid flow, confined in a horizontal cavity with a 2:1 aspect ratio. The Stream-Vorticity formulation tackles the dimensionless governing partial differential equation. Single-phase model has been employed to model the nanofluid. An Alternating Direction Implicit (ADI) technique has been employed to address the governing equations. The research highlights a significant increase in the Nusselt number (Nu) with intensified magnetic fields. Additionally, introducing more nanoparticles enhances Nu with varied effects for different particles. Silver (Ag) and Copper (Cu) exhibit the highest increase in Nu (53%), indicating robust thermal-fluid coupling, while Titanium Dioxide (TiO₂) shows lower increases (37%), implying weaker coupling in the flow. These findings hold relevance for diverse applications, including transportation, energy, medical technology, materials science, and fundamental physics.