Abstract

A process and system for flame detection includes a microprocessor-controlled detector with a first sensor for sensing temporal energy in a first optical frequency range, and a second sensor for sensing temporal energy in a second optical frequency range. The temporal energy sensed in the respective first and second optical frequency ranges are transformed into respective first and second spectra of frequency components. A compensated spectrum of frequency components is generated by performing a frequency bin subtraction of the first and second spectra of frequency components. The compensated spectrum of frequency components represents the energy emitted from the environment with energy emitted from false alarm sources. An average amplitude and centroid of the compensated spectrum of frequency components are obtained and used to determine if a monitored phenomenon represents an unwanted fire situation. The compensated spectrum of frequency components can be compared to reference compensated spectra of frequency components generated from known unwanted fire sources and known false alarm sources. This comparison can be facilitated by constructing a frequency space scatter plot from respective average amplitudes and centroids obtained from the reference compensated spectra. A fire detection boundary can be defined, which excludes substantially all of the false alarm sources. Inclusion of the unknown phenomenon within the fire detection boundary is indicative of an unwanted fire situation.

Other References

• Bjorklund, F.B. et al., "Fire Loss Reduction--Part I We've Only Scratched the Surface!," "Fire Loss Reduction, Part II--Technology Holds the Key?," reprint from AID magazine (a publication of the National Alarm Association of America), Mar. 1986
• "Optical Fire Sensors State-of-the-Art," technical brochure of Fire Sentry Corporation, 1989-9