| Summary: | Smoke control systems limit the spread of smoke in buildings, in the event of a fire. They have only been provided as engineered systems for the past forty years. There is little published data available on their performance in real fire situations. The effectiveness of a smoke control system is dependent on the accuracy of the modelling of the fire and smoke environment, and on the performance of the installed system. This thesis examines how these two aspects of smoke control systems may vary. An experimental program was developed to examine smoke movement from a compartment and within an atrium. Comparison of the results was made to a set of design calculations. Velocity measurements were made using particle imaging velocimetry (PIV). This unobtrusive technique has provided additional information on smoke flow movement by providing a visual image of the flow, which is capable of measurement to generate velocity fields. The cause of faults during the life-cycle of smoke control systems and the effect on their performance have also been examined. A questionnaire was developed to determine engineers' experiences with smoke control systems. In addition to completing the questionnaire, some respondents also provided information for case studies. From this, and information from published sources, a taxonomy of smoke control system failure has been developed. This may form the basis for the future study of the reliability of smoke control systems.
|
|---|
