Air Sense smoke detectors

The ClassiFire system recognizes this situation and automatically increases the sensitivity of the detector under these circumstances, automatically compensating for the reduction in performance caused by the filter. By monitoring the changes in the histogram over a long period of time, it is possible to infer that as the mean output of the detector signal falls and the variance of the distribution reduces, the filter is becoming impassable to smoke, and the detector is increasingly seeing ultra-filtered air. This can be used to trigger a filter renew fault.

This has been confirmed in a report produced by the University of Duisburg, Germany dated October 7th 2002. the sensitivity of the fire detector changes in dependence on filter loading In such a condition, there may be very little effect on the air-flow through the filter or the air pressure across it, but clearly it will reduce the sensitivity of the smoke detection system placed downstream of this filter. This ultra-filtering shows on the histogram of the AirSense detector as a reduced Mean output and a reduction in variance/standard deviation.

AirSense smoke detectors continuously monitor the environment in order to set the most appropriate alarm level(s) through the Perceptive Artificial Intelligence system. This system uses well understood statistical models to build and compare a continually updating histogram of the protected environment as a method of recording the distribution of smoke density readings, and this histogram tracks changes in the environment and influences detector sensitivity as a result of the analysis. As a filter becomes progressively more clogged, it will progressively impede the passage of ever smaller particles, until it is able to block virtually all particles, including smoke but will allow molecules of air to pass almost unimpeded.