- Author
- Milke, J. A.
- Title
- Discriminating Fire Detection With Multiple Sensors and Neural Networks.
- Coporate
- Maryland Univ., College Park
- Sponsor
- National Institute of Standards and Technology, Gaithersburg, MD
- Book or Conf
- Fire Suppression and Detection Research Application Symposium. Research and Practice: Bridging the Gap. Proceedings. National Fire Protection Research Foundation. February 12-14, 1997, Orlando, FL, 12-26 p., 1997
- Keywords
- fire suppression | fire detection | fire research | fire safety | sensors | neural networks | smoke detection | evacuation | egress | detection time | small scale fire tests | large scale fire tests | carbon dioxide | carbon monoxide
- Identifiers
- single source experiments; classification of test sources; maximum rate of rise of CO₂ concentration; combinations of fire and non-fire sources; average CO/CO₂ ratio for fire and nuisance source combinations
- Abstract
- A primary objective of residential smoke detection is to increase the time available for evacuation. In a recent experimental study, only 200 to 300 seconds of available egress time was provided by a smoke detector located in the room of origin with a smoldering fire. In come cases the smoke detector in an adjacent room actuated after untenable conditions had developed in the room of origin.