Ann. occup. Hyg., Vol. 46, No. 8, pp. 701-712, 2002
© 2002 British Occupational Hygiene Society
Published by Oxford University Press
An Investigation of Air Inlet Velocity in Simulating the Dispersion of Indoor Contaminants via Computational Fluid Dynamics
1 Department of Environmental Health Sciences, HESC, Room 311, Norman J. Arnold School of Public Health; 2 Department of Mechanical Engineering, University of South Carolina, Columbia, SC 29208, USA
Received 17 December 2001; in final form 22 July 2002
Computational fluid dynamics (CFD) is potentially a valuable tool for simulating the dispersion of air contaminants in workrooms. However, CFD-estimated airflow and contaminant concentration patterns have not always shown good agreement with experimental results. Thus, understanding the factors affecting the accuracy of such simulations is critical for their successful application in occupational hygiene. The purposes of this study were to validate CFD approaches for simulating the dispersion of gases and vapors in an enclosed space at two air flow rates and to demonstrate the impact of one important determinant of simulation accuracy. The concentration of a tracer gas, isobutylene, was measured at 117 points in a rectangular chamber [1 (L) x 0.3 (H) x 0.7 m (W)] using a photoionization analyzer. Chamber air flow rates were scaled using geometric and kinematic similarity criteria to represent a full-sized room at two Reynolds numbers (Re = 5 x 102 and 5 x 103). Also, CFD simulations were conducted to estimate tracer gas concentrations throughout the chamber. The simulation results for two treatments of air inlet velocity (profiled inlet velocity measured in traverses across the air inlet and the assumption that air velocity is uniform across the inlet) were compared with experimental observations. The CFD-simulated 3-dimensional distribution of tracer gas concentration using the profiled inlet velocity showed better agreement qualitatively and quantitatively with measured chamber concentration, while the concentration estimated using the uniform inlet velocity showed poor agreement for both comparisons. For estimating room air contaminant concentrations when inlet velocities can be determined, this study suggests that using the inlet velocity distribution to define inlet boundary conditions for CFD simulations can provide more reliable estimates. When the inlet velocity distribution is not known, for instance for prospective design of dilution ventilation systems, the trials of several velocity profiles with different source, air inlet and air outlet locations may be useful for determining the most efficient workroom layout.
Keywords: air inlet velocity; boundary condition; computational fluid dynamics; workroom exposure; validation; ventilation