ANN-based two hidden layers computational procedure for analysis of heat transport dynamics in polymer-based trihybrid Carreau nanofluid flow over needle geometry

Abstract

Significance

Nanofluids over a continuously moving thin needle play a crucial role in thermal transport processes in various situations. This geometry facilitates the heat transfer mechanism, which could be crucial in many real-world applications such as cooling of electronic devices, heat exchangers and advanced manufacturing techniques.

Purpose

A novel investigation of polymer-based trihybrid Carreau nanofluid flow subjected to thermal radiation and magnetohydrodynamic consequences (MHD) over a continuously moving thin needle has been made in this research attempt. Velocity of fluid is scrutinized through magnetic aspect and transport of heat is inspected through thermal radiation and heat sink source. In addition, implementing advance ANN-based computational procedures such as multi layers neural networks (MLNNs) provide valuable aid in unmatched capability for capturing the high complexity of heat transfer in fluid flow problems. Their advantages in handling nonlinearities and modeling high-dimensional data through integrating physical laws make them far superior to simpler machine learning and other traditional techniques, despite requiring greater data and computational resources.

Methodology

The physical model is originally formed with the help of partial differential equations (PDEs), that are formulated with pre-defined assumption of fluid flow mechanism. These governing system is transformed into ordinary differential equations (ODEs) via appropriate similarity transformations. Numerical computation of ODEs is made by a well-known bvp4c scheme and then an advanced artificial neural network (ANN) computational framework is integrated to train the resulting dataset, which is based on scaled conjugate gradient neural network (SCG-NN) to facilitate predictions regarding advanced solutions.

Findings

The velocity profile of a trihybrid nanofluid decreases with an increasing values of Weissenberg number and magnetic parameter but in case of numeric growth in Carreau index parameter, the magnitude of velocity is increasing due to shear-thinning behavior. On the other hand, temperature profile of a polymer-based trihybrid nanofluids decreased with augmented values of radiation parameter and heat generation parameter due to the enhanced radiative heat transfer and the specific thermal properties of the nanofluid as well as generated amount of heat respectively.