Library Subscription: Guest

ISSN Online: 2379-1748

7th Thermal and Fluids Engineering Conference (TFEC)
SJR: 0.152 SNIP: 0.14 CiteScore™:: 0.5

Indexed in

Clarivate CPCI (Proceedings) Scopus
May, 15-18, 2022 , Las Vegas, NV, USA

A CFD SOLVER BASED ON THE LATTICE BOLTZMANN METHOD TO SOLVE THREE-DIMENSIONAL THERMALLY DRIVEN FLOWS AND COUPLED MOLECULAR GAS RADIATION: COMPARISON AND VALIDATION AGAINST A BENCHMARK SOLUTION

Get access (open in a dialog) pages 409-418
DOI: 10.1615/TFEC2022.emt.041362

Abstract

Characterizations of heat exchanges and flow structures of thermally driven convection in simple configurations such as enclosed cavities have been intensively carried out both experimentally and numerically for a wide range of Rayleigh numbers. In most cases, the problem is reduced to a thermal conduction-advection flow induced by temperature differences at the boundaries. However, some recent works have emphasized the significant role of radiative transfer in the thermal stratification and stability of the flow. The present work introduces a new solver built from a CFD code based on the lattice Boltzmann method under the Boussinesq approximation coupled with a radiative transfer model solving the Radiative Transfer Equation in participating molecular gases. Simulations are performed in a differentially heated threedimensional cubical cavity filled with a air/H2O/CO2 mixture. Stationary solutions of the resulting laminar flow are obtained and the overall effects of gas radiation on the flow characteristics are obtained. Results are compared to a benchmark solution. Temperature, velocity and radiative heat source profiles in the cavity fit particularly well in all cases, along with convective and radiative heat fluxes at the walls. The present numerical model is shown to be relevant to solve thermal natural convection problems at low Rayleigh numbers with limited computational costs. Results are discussed with the aim of highlighting some modelling features which should be handled with care in order to perform accurate simulations.