Levenberg–Marquardt artificial neural network approach to analyze the effect of surface roughness,horizontal magnetic field,and nanoparticle shape factor on ternary hybrid nanofluid flow

Abstract

Investigating ternary nanofluid flow and heat transport across a rough revolving disk has important applications in modern engineering systems requiring efficient thermal management. The roughness of the rotating disk mimics real-world surfaces, enhancing turbulence and thus heat transport efficacy. In view of this, the current study focuses on heat transmission and ternary hybrid nanoliquid flow over a rough rotating disk in the presence of a porous medium and a horizontal magnetic field. Additionally, the influence of nanoparticle shape factor, variable thermal conductivity, and quadratic thermal radiation on the fluid motion is scrutinized in the study. The governing partial differential equations (PDEs) are transformed to ordinary differential equations (ODEs) with the use of similarity transformations. The Runge–Kutta–Fehlberg fourth–fifth order (RKF-45) method is used to numerically solve the resulting ODEs. Additionally, the Levenberg–Marquardt artificial neural network (LMANN) is used to evaluate fluid flow and heat transfer properties. Also, the values of the RKF-45 technique are compared with the results of the LMANN technique. Graphical representations are used to show a thorough inspection of the physical behavior of various parameters on the energy and velocity profiles. The enhancement in the shape factor enhances the temperature profile. An increment in the variable thermal conductivity and radiation parameters increases the energy profile. The greater porosity and magnetic parameters decline the velocity profiles.