Machine learning analysis for two phase flow of manganese zinc ferrite and nickel zinc ferrite in dusty hybrid nanofluid applications of melting heat

Abstract

This study employs a neural network based on the back-propagated Levenberg-Marquardt algorithm to explore the impacts of melting heat on gyrotactic bacteria in the two-phase flow of a MnZnFe2O4-NiZnFe2O4/Water-based dusty hybrid nanofluid across a sheet, taking Soret and Dufour importance into consideration. The efficient Bvp4c Method is then used to numerically resolve the system of equations. This model has numerous uses in industrial and energy systems where two-phase flows, melting heat transmission, and coupled heat-mass diffusion are crucial. The machine learning analysis of two-phase flow of dusty hybrid nanofluid with melting heat and Soret-Dufour impacts can be used in metal and polymer melting processes, thermal energy storage systems, solidification and casting industries, and particulate-laden heat exchangers. The model improves thermal efficiency, better controls melting and mass transfer rates, and optimises performance in advanced manufacturing and high-temperature industrial applications by capturing the interaction between fluid, nanoparticles, and dust particles. The concentration and thermal fields of the dusty hybrid nanofluid rise with increasing Soret and Dufour numbers.