Information Processing Of Fire Detecting Based On Fuzzy Neural Network

This paper mainly concentrates on the intelligent algorithm applied to the information processing of fire detecting. The research tasks include using fuzzy neural network to process fire signals and decrease the rate of wrong alarm of fire detecting system. A distributed fire alarm system implemented by an embedded industrial control computer PC/104 and multi-sensors is introduced. The algorithm mentioned above is applied to the process of fire signals.The performance, applicability and the cause of wrong alarm of fire detectors were compared comprehensively based on the fire detector classification. The key problem of improving the accuracy of fire detection is to reduce the rale of wrong alarm. The feasibility about how to detect complexly and how to deal with the detection results intelligently is discussed.Neural networks technology and fuzzy control technology are available in processing information but each has defect. Fuzzy neural network has been comprised connected the two technology's advantage. Input signals are first fuzzied and then processed on a neural network. Which is a forward BP network with three layers. BP algorithm is used to train network parameters. Thereafter, fire is judged by means of fuzzy logic. The simulation result shows that this network can detect different standard test fire accurately and have strong anti-interference capability.At the end of the paper, after reviewing the recent developments of intelligent buildings, this paper presents a fuzzy neural network fire detecting system consisting of an embedded industrial control computer PC/104 andmulti-sensors, analyzing the structure and characteristics of PC/104. How to realize the fuzzy neural network, increase the speed and reliability is also discussed.