Optimization and classification of thermal transport on a convective surface with non-uniformly shaped ternary hybrid nanofluid flows

Abstract

Research background: This research makes it possible to classify heat transport precisely in sophisticated cooling systems, which improves performance in energy, electronics, and aerospace applications. Integrating Support Vector Machine Learning classification and ternary hybrid nanofluids offers innovative solutions for efficient heat dissipation in complex three-dimensional geometries. Research problem /issue: The current model investigates the effects of radiation, Casson, magnetic field, and non-Fourier heat flux on the behaviour of seawater-based hybrid nanofluid fluid flow dynamics and heat transfer using ternary hybrid nanoparticles, such as aluminium alloy (spherical shape), singular-walled carbon nanotubes (cylindrical shape), and multi-walled carbon nanotubes (platelet shape) in the convective surface. Method: Governing partial differential equations are transformed into ordinary differential equations by self-similar transformation; solutions for the nonlinear system are obtained via a boundary value problem solver using numerical techniques Boundary Value Problem 5th-order Collocation method in matrix laboratory software. Results and conclusion: Support vector machine learning was able to categorize heat transfer performance according to essential contributing elements for the suction and injection scenarios; the effect of non-dimensional flow parameters on flow profiles is shown graphically and tabularly, as the radiation parameter R increases, ternary hybrid nanofluid continuously exhibits a heat transfer rate that is roughly 46–47 % higher than hybrid nanofluid flow, industrial applications of the present study are Paper manufacture, plastics, ceramics, treatments, the automotive sector, cosmetics, and medicines, just a few of the many technical uses for these nanoparticles.