Thermodynamic case assessment of the micropolar fluid using neural network fitting tool and quasi-linearization technique: An asymmetric channel flow application

Abstract

Asymmetric channel flows incorporating micropolar fluids are applicable in designing lubrication systems for bearings, gears, and seals, especially in industries like automotive and aerospace engineering. This research presents a comprehensive thermodynamic assessment of the micropolar fluid flow in an asymmetric channel using a combination of the neural network fitting tool (NNFT) and the quasi-linearization (QL) technique. The thermodynamic properties, mass transfer characteristics, and flow dynamics of the micro-structured fluid are the main focus of this study. The system of governing equations is transformed into a set of ordinary differential equations that are solved iteratively with the help of QL method. The source parameters of the problem are the Peclet number for heat diffusion, microinertia density, Peclet number for mass diffusion, chemical reaction parameter, spin-gradient viscosity parameter, Reynolds number, vortex viscosity, porosity parameter, and Eckert number. A 10 % increase in the Peclet number $P{e}_h$ lead to an increase of 10–25 % in heat transfer rate. In the same way, a 10 % increase in Peclet number $P{e}_m$ for mass diffusion changed heat transfer by 1–10 %. The change depends on how strongly mass diffusion affects thermal transport. The NNFT yields accurate predictions of the thermodynamic performance of micropolar fluid flow within an asymmetric channel having permeable walls. The surface drag force is reduced on both channel walls due to the higher values of micropolar material parameters.