Holmberg S, Li Y
Year:
1997
Bibliographic info:
18th AIVC Conference "Ventilation and Cooling", Athens, Greece, 23-24 September 1997

Concentrations of indoor air contaminants are normally calculated by assuming that they fullyfollow airflow paths in a room. This assumption is also used to predict the local residence timeof contaminants in a room, which may further be used to characterise the ventilation effectiveness.In this paper, a different methodology has been adopted, in which indoor airborne particles do notalways follow the main airstream induced by the ventilation system. Dispersion of particles ispredicted by a drift-flux model. The model assumes that the settling velocity of each particle issufficiently small when compared to the high inflow velocity and the volumetric concentration ofparticles is also very low. This assumption has justified the use of a drift-flux multi-phase modelrather than a fully coupled multi-fluid model. Additionally, the effect of the particles on turbulencecan be neglected. In the drift-flux model, a settling term is added to the concentration equation,and the body force term in the momentum equation is treated using the principle of a Bousinnesqapproximation, similar to that in a thermal-buoyancy-driven flow.The model has been previously validated by comparing numerically calculated results with thosemeasured in an aerosol chamber. In the examples in this paper, particles with diameters rangingfrom 0.5 micrometer to 5 micrometer are supplied into a room through a ventilation and air-conditioning register.Local particle concentrations are calculated for different particle size groups. It is shown that inthe air-conditioning situations considered, the thermal-buoyancy-introduced flows provide anadditional mixing mechanism for particle dispersion in the room. The developed model can be auseful tool for minimising particle concentrations in the air by evaluating and selecting anoptimum air distribution and air-conditioning system.