Sensitivity exploration of micropolar fluid through porous permeable walls using ANN

Abstract

This paper investigates the heat transfer process in a two-dimensional flow of a micropolar fluid through porous permeable walls under a constant heat flux condition. To simplify the problem, a similarity transformation is applied, converting into a set of nonlinear boundary value problems. The Runge–Kutta–Fehlberg (RKF) method is then used to derive the series solutions for key variables such as microrotation, velocity, concentration distributions, and temperature. The study also explores the influence of several important parameters, including Reynolds number, magnetic and radiation parameter, Prandtl number, and Peclet numbers for mass and heat transfer. Through the use of response surface methodology (RSM), sensitivity analysis, and artificial neural networks (ANN), the research demonstrates that the system is well defined and behaves predictably. The study includes a range of graphical representations, such as residual contour and surface plots, which help to visualize and analyze the key components of the flow field. Additionally, tables are provided to present the results more clearly, allowing for a deeper understanding of how various parameters influence the system’s behavior.