A parallel computational study of power-law non-Newtonian nanofluid in a C-shaped enclosure by multiple-relaxation-time lattice Boltzmann simulation

Abstract

ABSTRACT This study aims to numerically examine the heat transfer and entropy generation of non-Newtonian power-law nanofluids in a C-shaped cavity filled with porous media considering the Brinkman-Forchheimer extended Darcy model (BFEDM). The graphics processing unit (GPU) based multiple-relaxation-time lattice Boltzmann method (MRT-LBM) was used through computing unified device architecture (CUDA) C/C++ platform. It is a known fact that the fluid viscosity varies with the shear rate and nanoparticle sizes and the thermal conductivity also varies with the fluid temperature. However, the extension of such phenomena needs to be further investigated to understand any specific threshold in heat transfer application. To leverage such characteristics, it is permissible to add certain input parameters. As a result, the power-law indices ( n ), the Darcy number ( Da ), the volume fraction of nanoparticles ( ϕ ), the Rayleigh number ( Ra ), and the fixed porosity ( ε = 0.4 ) were the five key parameters considered to analyze the numerical results. The GPU-based numerical code was validated with the available benchmark results, and good agreements were obtained. The results showed that the average rate of heat transfer in terms of the average Nusselt number ( Nu ‾ ) increased due to an increase in Ra due to the enhanced buoyancy force and a reduction in the .