Multivariate Methods for Fire Detection.
Multivariate Methods for Fire Detection.
(654 K)
Milke, J. A.; McAvoy, T. J.
NISTIR 6030; June 1997.
U.S./Japan Government Cooperative Program on Natural
Resources (UJNR). Fire Research and Safety. 13th Joint
Panel Meeting. Volume 2. March 13-20, 1996,
Gaithersburg, MD, Beall, K. A., Editor(s), 411-418 pp,
1997.
Sponsor:
National Institute of Standards and Technology,
Gaithersburg, MD
Available from:
National Technical Information Service
Order number: PB97-184204
Keywords:
fire safety; fire research; fire detection; fire
detectors; small scale fire tests; large scale fire
tests; experiments; light obscuration
Abstract:
Research is being conducted to describe the
characteristics of an improved fire detector which
promptly reacts to smoke while discriminating between
smoke and odors from fire and non-fire sources.
Discrimination is accomplished by comparing signature
response patterns from fire and environmental sources
collected in small- and large-scale tests. Smoke and
odors are produced in the tests from a variety of
conditions: flaming, pyrolyzing and heated samples, and
nuisance sources, such as aerosols, household products
and cooked food. Measurements include light
obscuration, temperature, mass loss, CO CO2, O2 and
oxidizable gas concentrations. The feasibility of an
elementary expert system to classify the source of the
signatures from small-scale experiments was demonstrated
in the first phase. In the second and third phases, an
expert system was developed by Multivariant statistical
methods to distinguish between fire and non-fire
sources. In addition, the presence of a fire could be
identified despite the interjection of signatures from
nuisance sources which could mask the fire signatures.
Building and Fire Research Laboratory
National Institute of Standards and Technology
Gaithersburg, MD 20899