Research On Fire Detection Algorithm Based On Data Fusion Technology

With the enlargement of scale of high and large buildings, their fire fatalness is more and more dangerous. It is more and more difficult to save life and property in the building. So it has more and more request about the performance of the fire detection system performance. The fire parameter gathered by the detectors is unable to know in advance, the non- constitutive signal. The traditional survey methods merely carry on the judgment and the recognition through gathering sole fire characteristic parameter information. So it is disturbed by the environment inevitably. The question about the system's high mistakenly reporting rate is prominent. In recent years, the accuracy of fire alarm becomes better, while the sensibility and reliability of detectors are improved. But it can't match the request of the automation of the fire detection system. It can decrease mistakenly reporting only by describing fire inherent characteristic completely and exactly. Thus fire detection algorithm based on multi-criteria is the main research direction in the present fire surveying field.Through researching the fire mechanism and the present fire surveying methods, and combined with fire detection system's own characteristic, this article proposes fire detection algorithm based on the data fusion technology. It is to module fire detection algorithm, and to layer the information fusion. Expand the direct criteria such as characteristic parameter to the indirect criteria such as the fire hazard degree and fire disaster damage degree. The performance of decision-making of fire detection system is enhanced by fusing scene auxiliary information.Main contents and conclusion of this article:It divides fire detection algorithm into three layers: data layer, feature layer, decision-making layer.On data layer, the detector first handles some characteristic parameter such as smoke, temperature, CO, etc. Through the partial decision-making, achieve the distributed processing of the fire detection algorithm; enhance the overall performance of the system. When some parameters appear exceptionally, the same bunching parameter will submit for Feature layer to carry on the characteristic recognition.On feature layer, while NN has learning and association ability, it uses BP Algorithm Based on L-M to recognition open flame and smoldering fire and achieves their probability. It offers the direct criteria of fire for decision-making layer. On decision-making layer,it introduces the fire hazard degree and fire disaster damage degree as the interface of algorithm for each kind of scene auxiliary information. Thus the fire detection system can automatically make the reasonable policy-making output according to different region environment situation in the buildings. Decision-making layer uses the fuzzy inference algorithm to inosculate the direct criteria- open flame and smoldering fire probability,the duration,the indirect criteria - region fire hazard degree and the fire disaster damage degree. According to the rules formulated by experts' experience, finally makes the decision-making output of the fire detection system.The result of emulator experiments indicates that the method can improve the performance of decision-making of fire detection system.