correlation

Reviews the literature on wind tunnel modelling of natural ventilation. Lists advantages of using wind tunnels. Discusses strategies for utilizing natural ventilation. Describes the mean windspeed coefficient method and the wind discharge coefficient method of estimating natural ventilation for design of buildings in hot climates, and gives their advantages and disadvantages. Gives criteria for constructing wind tunnel models.

Natural and forced ventilation are directly and indirectly influenced by the pressure distribution around a building. Results of full-scale pressure measurements on a typical Swedish timber house are presented. The rate of air infiltration has been calculated by employing the values obtained from full-scale pressure distribution, air leakage characteristics and temperature differences. The results are compared with the actual ventilation obtained from tracer gas measurements.

As more component leakage test data become available, all available data has been merged using the orifice flow equation for a standard pressure differential of 50 Pa existing across the component. All component leakage areas are added to pro

The problems associated with the measurement of ventilation rates and air movement patterns in large single cell buildings which were unoccupied and unheated were investigated using the tracer gases nitrous oxide and sulphur hexafluoride. 

Describes research work in Finland concerning air infiltration and ventilation in buildings from 1979. Types of ventilation system in finnish houses, flats and other buildings are discussed. Reports on a model developed to predict the correlation between various factors and air infiltration. Gives a summary of pressure test measurements carried out in a few hundred small houses, and presents proposals for recommended airtightness levels in new buildings. Describes warm air heating systems, heat recovery systems in flats, and maintenance problems with ventilation systems.

Studies the predictive accuracies of infiltration-pressurization correlation models by subjecting a group of identical homes in Freehold NJ to both pressurization and tracer gas measurements of infiltration. Compares infiltration and pressurization test results. Uses four air infiltration models to predict the infiltration rates in the houses, and compares these predictions with the measured rates. Makes predictions using several different values of the inputs required for the models, and studies the effects of varying these inputs.

A knowledge of the pressure fluctuations on buildings exposed to strong winds is important for wind loading calculations. Presents the statistical quantities of such fluctuations in terms of rms values and power spectra for models resembling the Aylesbury experimental building of BRE, and compares this with full-scale results. Suggests that, provided the properties of the longitudinal velocity component are suitably simulated, then agreement between full-scale and model results in terms of rms values and power spectra can be achieved.

States that the calculation of building energy flow is very complex, and so validation is a vital element in the development of any model. Describes an IEA R and D project to compare 23 computer programs (from 8 different countries) both in terms of consistency between programs, and in thei relative accuracy in modelling the behaviour of a real building (the Avonbank office block in Bristol). Summarizes the major conclusions developing out of the project and discusses the most important aspects which need to be considered in the development of a reliable computer program.

Investigates the accuracy of the TWOZONE computer auditing program (used in analyzing the energy performances of residences) when used with home-owner supplied data. Compares actual energy consumption data collected from houses in New Mexico with estimates obtained from the TWOZONE model using input data from occupant questionnaires. Finds a high degree of variability between predicted and observed values, and concludes that much of the error introduced into the prediction is attributable to the qualitative nature of the data supplied by the respondents.

Looks at the requirements for computer model validation, especially in regard to predicting energy usage in buildings. Discusses the IEA project for comparing and validating several computer programs in this context. Describes the Glasgow commercial building monitoring project, which includes detailed measurement of temperature and air flow rates to provide data for model validation. States what type of data is needed for validation.