This paper describes the application of numerical models to predict the ventilation rate and internal air movement patterns for a naturally ventilated industrial building and compares the results with measured data. Two modelling techniques have been employed. Firstly, a zonal network model (HTBVent), using leakage area data derived from fan pressurisation measurements, was used to predict the time varying ventilation rate in response to variations in wind velocity and internal-external air temperature difference.
Zonal models are a promising way to predict air movement, in a room with respect to comfort conditions and gradient of temperature, because they require extremely low computer time and may be therefore rather easily included in multizone air movement models. The main objective of this paper is to study the ability of the zonal models to predict the thermal behaviour of air in case of natural convection coupled with a radiator. First, we present simplified two zone and five zone models.
A "HESCO"-type diffuser was selected as an example for the validation exercise in the IEA Annex 20 project (Air flow pattern within buildings). It consists of 84 small round nozzles that are arranged in four rows in an area of 0.71 m x 0.17 m. With the same effective area, the diffuser is simulated by 1, 12, and 84 simple rectangular slots and by the momentum method. In the momentum method, the supply air momentum is set to be that of the 84 small round nozzles. The simulation of the diffuser is incorporated in the airflow computation in a room.
The International Energy Agency (IEA) task-sharing project "Air Flow Patterns within Buildings" was initiated in May 1988 for a duration of 3,5 years. Twelve nations contribute work and expertise and "share the task" specified in the project's objectives. This project and the AIVC belong to the same Implementing Agreement: The Energy Conservation in Buildings and Community Systems Program. As "Attachments" to the Implementing Agreement, they are called Annexes.
Air is the main transport medium for contaminants in buildings. Minimizing source strengths has first priority, second is to control air flow rates, supply and exhaust, and directions between zones in buildings. Computer simulation models forventilation and pollutant spread in buildings have been proven to give useful predictions. Large measurement campaigns for optimizing ventilation and pollutant problems are complex and expensive. They are often jammed by too many vague parameters influencing the result. The computer models are an alternative and form a supplement to measurements.
Increasing design standards within the building industry mean that some form of pre-construction testing of the building envelope is required. Expensive and time consuming field tests are becoming more impractical whereas the cost-effectiveness a