现代电子技术2016，Vol.39Issue(18)：26-29,34,5.DOI:10.16652/j.issn.1004-373x.2016.18.007
基于神经网络的大型火情自动识别系统设计与研发
Design and development of large-scale fire automatic recognition system based on neural network
摘要
Abstract
The fire recognition capacity and positioning ability of large⁃scale fire automatic identification system designed in the past were influenced by its line layout and fire recognition hardware. Therefore,a new large⁃scale fire automatic recognition system based on neural network was designed. The system utilizes sensors to induce the abnormal signal existing in the environ⁃ment,and calls its detector to identify whether there is the large⁃scale fire hazard in the abnormal signal. Its controller performs amplification,analog⁃to⁃digital conversion and cyclic redundancy check of hidden trouble signal identified by the detector. The results are output to the alarm device,and regarded as the alarm basis to send to users. The system uses the arrangement mode with two bus lines,and merges the functions of the cable and the signal wire into a line,which can decrease the difficulty of cir⁃cuit layout and save the cost of system maintenance. The BP neural network model and the probabilistic neural network model are designed for the system to deal with large⁃scale fire signal. The experimental results shows that the designed system has high ability of fire recognition and fire positioning.关键词
神经网络/火情自动识别系统/火情定位/信号放大Key words
neural network/fire automatic identification system/fire positioning/signal amplification分类
信息技术与安全科学引用本文复制引用
鄢莉..基于神经网络的大型火情自动识别系统设计与研发[J].现代电子技术,2016,39(18):26-29,34,5.基金项目
教育部信息技术自然基金 （）
