CFD prediction of contaminant distribution in indoor car park.

The assessment of human exposure to airborne contaminant is an important issue in building design. The physiological significance of such exposure and technical means to minimise such risks have long been known in literatures. (1, 2, 3) In recent years, computational works have increasingly been seen used as design assessment tools as an alternative to site measurement and wind tunnel tests. Questions arise that most of such computation studies were restricted to single spaces due to high computing requirements and theoretical complexities ignoring the physical reality that the spaces were not isolated, it is encased by walls, occupants and furniture which were participating in the thermal environment. To address these issues two schools of thoughts have emerged. One was to enhance the CFD codes with the fabric and long-wave radiation models to enable it to model both airflow and thermal conduction simultaneously by prescribing boundary conditions external to the building. (4, 5, 6) The other was to extend the conduction based building thermal simulator to include the CFD codes so as to enable the two models interacting on a time-step basis, exchanging information at their model boundaries. (7, 8) Theoretically the two thoughts are sound with which the former was already be seen in rocket and turbine engine designs. While in CFD simulation of . buildings they only succeeded in limited cases of academic studies. The main reason for this is that both the two approaches have been unable to handle the differences in time constants between the walls and the air volumes, which very often differ hundreds of times of order of magnitude. In the case of a realistic building, any attempt to simulate the two systems on a time-step basis would be rendered unnecessary. In this paper an integrated approach is employed to take into account of the large differences in time constants between walls and air in the buildings, the CFD code is refined to account for the turbulent, buoyancy driven flow and the particle movement of contaminant.