CFD simulations of near-field pollutant dispersion with different plume buoyancies. (March 2018)

• Record Type: Journal Article
• Authors: Tominaga, Yoshihide; Stathopoulos, Ted
• Is Part Of: Building and environment. Volume 131(2018)
• Page Start: 128
• Page End: 139
• Publication Date: 2018-03
• Subjects: CFD -- Plume buoyancy -- Near-field dispersion -- Cubical building; Buildings -- Environmental engineering -- Periodicals; Building -- Research -- Periodicals; Constructions -- Technique de l'environnement -- Périodiques; Electronic journals; 696
• DOI: 10.1016/j.buildenv.2018.01.008
• Languages: English
• ISSNs: 0360-1323

Abstract

Abstract: This study performs computational fluid dynamics simulations for flow and dispersion fields around an isolated cubic building model with tracer gases being exhausted from an exit behind the building. The tracer gases have three different buoyancies according to the difference in density with ambient air and, therefore, behave as neutral, light, and heavy gases. The performance of steady Reynolds-averaged Navier–Stokes (RANS) simulations with the Boussinesq approximation is examined herein by comparing the simulation results with the experimental results for different plume buoyancies. The steady RANS computations can generally reproduce the impact of plume buoyancy on the mean concentration in the experimental results even if the model performance for heavy gases is better than that for light gases and worse than that for neutral gases. This tendency is closely related to the prediction accuracy of the mean velocity and turbulent kinetic energy behind the building, which is restricted by the steady RANS simulations. The study also confirmed that the buoyancy modeling in the ε equation shows a negligible influence on the results. Graphical abstract: Image 1 Highlights: Steady RANS simulations for dispersion with plume buoyancies were performed. The model performance was examined by comparing the results with experimental results. CFD generally reproduced the impact of plume buoyancies on the mean concentration. The model performance was affected by the prediction accuracy of the flow field.