Evaluation of measured and computer test case results from Annex 20, Subtask 1.

The ability to accurately predict air movement and temperature distribution in spaces offers the potential for design engineers to evaluate and optimise room air distribution systems at an early stage, leading to improved thermal comfort and ventilation effectiveness. The computer models which are used for detailed analyses are based on computational fluid dynamics [1,2] and employ sophisticated numerical algorithms to satisfy the basic laws of physics. The programs are such that they are more complex and more difficult to use than those with which design engineers may be more familiar. Specialised skills are required to get the best from the codes, and, as with most new techniques, a greater confidence is needed before their use can be expected to become more routine. It is the latter point concerning confidence in use which is addressed by IEA Annex 20, Subtask 1. IEA Annex 20 is a task-sharing project on "Air flow patterns within buildings". The objective is to evaluate the performance of single- and multi-zone air and contaminant flow simulation techniques and to establish their viability as design tools. In subtask 1 of the Annex, which deals with single-zone spaces, laboratory experiments in similar test rooms and computer simulations have been carried out at a number of sites in Europe and North America. The data comprises information on air flow patterns and on point-by-point values of mean velocity, turbulent velocity, temperature and contaminant concentration throughout a space. This paper outlines an initial evaluation of this data and highlights some of the features which the comparisons of measured and computed room air distribution have yielded. Work is continuing in completing the evaluation for tests and data not reported or discussed here. Besides giving a unified perspective on data from different sites to quantify the general degree of agreement, the evaluation exercise also has the potential for: - establishing benchmarks for the validation and evaluation of computer codes for room air movement; - highlighting advantages/limitations of the methods used; - assessing overall confidence level in computer simulations; - indicating accuracy and repeatability of measurements and simulations; - guiding research on simplified models of air movement and identifying problem areas where attention should be focused.