A fire detection system based on intelligent data fusion technology

In this paper, multi-sensor data fusion technology is used in fire detection, which can better solve problems existing in traditional fire detection, such as lower-grade intelligence, high mis-warning rate, delay-warning etc. According to the fire signal\´s inherent character, a 3-layers data fusion structure is used. In this fire detecting system, the fire experiential characteristic and the fire data-fitting characteristic of fire signal data are fused by the fuzzy inference system to get the last fire probability. After introducing the "feedback" of cybernetics, a new algorithm that extracts the fire data-fitting characteristic is presented. The algorithm may improve the accuracy and quickness of early warning with a tendency feedback signal. Finally simulation experiments are done for typical flaming fire, typical smoldering fire and the fire under typical disturbance signals in kitchen environment. And satisfactory results are obtained.