结合当前舰船火灾探测报警系统的现状和对于火灾早期探测的需求，提出一种将主动吸气式空气采样感烟探测器和阴燃类电气火灾燃烧产物特征量CO及传统温度信号通过模糊神经网络和遗传算法进行融合判定的系统设计方法，详细描述系统的构成，通过数据采集、模糊推理、火灾判定的软件实现过程。同时通过仿真计算表明该系统对于提高火灾探测报警的关键性能指标“响应时间”和“误报率”的有效性。该系统设计方法为电气类阴燃火的早期探测提供有效的技术解决途径，同时提高了系统抗误报警能力。
Combined with the current status of ship fire detection and alarm system and the demand for early fire detection, this paper presents a method to fuse the active aspirating air sampling smoke detector, the characteristic quantity CO of smoldering electrical fire products and traditional temperature signal through fuzzy neural network and genetic algorithm, The article describes the system structure in detail, through data acquisition, fuzzy reasoning, fire descision software implementation process. At the same time, the simulation results show that the system is effective to improve the detection and alarm performance. So as to achieve the goal of early fire detection and alarm, and improve the anti-false alarm ability of the system.
2021,43(12): 161-165 收稿日期：2021-01-06
DOI：10.3404/j.issn.1672-7649.2021.12.029
分类号：U664.88
作者简介：郑珊珊(1979-),女,高级工程师,主要从事舰船消防、火灾自动报警系统设计与研究
参考文献：
[1] 吴龙标, 袁宏永, 火灾探测与控制工程[M]. 北京: 中国科学技术大学出版社, 1999.
[2] WANG X H, XIAO J M, BAO M Z. A ship alarm system based on fuzzy neural network[C]//Proceedings of the 3rd World Congress on Intelligent Control and Automation, Hefei, China, 2000.1734-1736.
[3] 付永丽, 董爱华. 模糊神经网络在火灾探测系统中的应用研究[J]. 电气技术, 2008(2): 54–57
[4] 李国勇, 神经模糊控制理论及应用[M]. 北京: 电子工业出版社, 2009.
[5] 王娜, 徐凤荣, 刘海龙. 火灾探测的模糊神经网络数据融合算法[J]. 控制工程, 2007, 14(3): 44–49
[6] 刘雪飞, 贾勤. 基于模糊神经网络的中庭火灾探测仿真研究[J]. 计算机仿真, 2012, 29(8): 159–162, 259
[7] 王锡淮, 肖健梅, 鲍敏中. 模糊神经网络和遗传算法结合的船舶火灾探测[J]. 仪器仪表学报, 2001, 22(3): 312–314
[8] GOTTUK D T, PEATROSS M J, ROBY R J, et al. Advanced fire detection using multi-signature alarmalgorithms[J]. Fire Safety Journal, 2002, 37(4): 381–394
