Computational fluid dynamics challenge on indoor dispersion of pathogen-laden aerosols

Identifier:  imarina:9448908

Handle: https://hdl.handle.net/20.500.11797/imarina9448908

Authors:  Pallares, Jordi; Fabregat, Alexandre; Lavrinenko, Akim; Marques, Nelson; Santos, Bruno; Mosca, Gabriele; Vega, Pedro Obando; Ravnik, Jure; Vovk, Nejc; Fraga, Bruno; Monka, Aleksandra; Martinez, Manuel; Mestre-Curto, Naomi; de Souza, Francisco Jose; Fontes, Douglas; Juengling, Natalie; Niessner, Jennifer; Castilla, Robert; Garcia-Vilchez, Merce; Fletcher, David F; Inthavong, Kiao; Hribersek, Matjaz; Steinmann, Paul; Wedel, Jana; Duchaine, Florent; Sankurantripati, Shriram; Amari, Leo; Janiga, Gabor; Marchioli, Cristian; Cito, Salvatore

Abstract:
This paper presents and discusses the results of the "2024 International Computational Fluid Dynamics Challenge on the long-range indoor dispersion of pathogen-laden aerosols" aimed at assessing the ability of different computational codes and turbulence models to reproduce the dispersion of particles produced by a turbulent natural convection flow enclosed in a room sized cubical cavity. A total of 12 research groups from ten different countries have conducted 15 simulations of the same flow configuration by solving the Reynolds averaged Navier-Stokes (RANS) equations, the unsteady Reynolds averaged Navier-Stokes (URANS) equations or using scale adaptive simulations (SAS), large-eddy simulations (LES), or hybrid (URANS-LES) techniques. Results for the velocity field and the particle dispersion provided by the different simulations are compared extensively, including the reference results provided by a direct numerical simulation (DNS). In general, LES and hybrid methods reproduce the time-averaged flow field correctly, the spatial distribution of the turbulence kinetic energy, and the particle dispersion. The performance of SAS is similar to that of LES and hybrid methods while the predictions of the RANS and URANS simulations exhibit larger deviations with respect to DNS. In general, the particle dispersion is better reproduced by simulations that capture correctly the spatial distribution of the turbulence kinetic energy.