Lagrangian particle modeling in the indoor environment: A comparison of RANS and LES turbulence methods (RP-1512)

Abstract

Indoor particulate contaminants can be generated in many ways, commonly from human activities, infiltration of HVAC systems, or resuspension from indoor surfaces. Most of these sources are transient and generate nonuniform particle distribution in the space. This study used experimental and numerical methods to investigate the dispersion of three different particle sizes (0.7, 2.5, and 7 μm) emitted from typical source positions. A test room and simplified thermal manikins were employed to mimic a realistic indoor environment, and experimental data were compared with particle modeling using the Lagrangian method coupled with Reynolds averaged Navier-Stokes (RANS) and large eddy simulation (LES) computational fluid dynamics (CFD) turbulence models. Particle dispersion was studied for two ventilation patterns: buoyancy-driven ventilation and well-mixed ventilation. The results provided a comparison of Lagrangian-RANS particle modeling, Lagrangian-LES particle modeling, and experimental data considering nonuniform temporal and spatial particle concentrations. Experimental and modeling results were evaluated with three different metrics: peak normalized concentration at various locations, peak-concentration occurrence time, and mean exposure defined as the averaged concentration in the occupant's breathing zone. The results show that Lagrangian-LES more accurately predicts concentration fluctuation during particle emission. Considering long-term exposure, however, both methods show similar results.