Fire Safety Science Digital Archive

IAFSS Symposiums

IAFSS Symposiums All Symposiums Symposium 1 Symposium 2 Symposium 3 Symposium 4 Symposium 5 Symposium 6 Symposium 7 Symposium 8 Symposium 9 Symposium 10 Symposium 11 Fire Research Notes AOFST Symposiums
Approach To Detection Of Fires In Their Very Early Stage By Odor Sensors And Neural Net

Okayama, Y., 1991. Approach To Detection Of Fires In Their Very Early Stage By Odor Sensors And Neural Net. Fire Safety Science 3: 955-964. doi:10.3801/IAFSS.FSS.3-955


ABSTRACT

Two odor sensors having SnO2 film layers of different thickness were used for detection of different odors and their outputs were plotted on X and Y axes respectively. There was a clear difference between the traces of their outputs drawn as they were exposed to odors from smoldering materials and from organic solvents used everyday. So, these outputs of the sensors were given to a neural net with which a system capable of discriminating different odors was developed. This system was able to distinguish odors from smoldering materials among others with high accuracy.


Keyword(s):

fire detection, smoldering fire, odor sensor, tin dioxide, neural net, odor recognition


View Article

Member's Page | Join IAFSS | Author's Site

Copyright © International Association for Fire Safety Science