Measurement and prediction of indoor air flow in a model room

In the interest of designing an efficient and acceptable indoor air environment in modem buildings, it is important to resolve the relationship between geometric room parameters and the air flow patterns produced by mechanical ventilation systems. Toward this end, we compare results from relatively simple three-dimensional numerical simulations (CFD) with laser Doppler anemometry (LDA) and particle image velocimetry (PIV) experimental measurements of indoor air flows in a one-tenth sub-scale model room. Lammar, k-epsilon turbulence, and RNG k-epsilon turbulence numerical models are used and evaluated with respect to their performance in simulating the flow in the model room, and results of the numerical simulations and velocimetry measurements show how obstructions can greatly influence the air flow and contaminant transport in a room. It is important, therefore, that obstructions be considered in ventilation design. Simulations predict the measured trends in a model room very well, with relative errors not much larger than 20%. In this study, the RNG model most accurately predicts the flow in a partitioned room, capturing the gross effects of a large flow obstruction. (C) 2002 Elsevier Science B.V. All rights reserved.