为了能够在火灾中对不同类型舱室实时判断舱室危险等级，本文提出一种能够实时判断火灾中不同舱室危险等级的评估模型。首先利用Pyrosim软件设计了一个单层多舱室平台，分别定性和定量地分析火灾烟气蔓延的规律，根据温度、$ {\mathrm{C}\mathrm{O}}_{2} $浓度、CO浓度、$ {\mathrm{O}}_{2} $浓度和烟尘密度对人的影响将火灾划分为4个等级，然后利用一个6层的BP神经网络对舱室火灾危险等级进行实时划分。测试结果表明，本文的模型分类准确度达到了98%，为判断火灾蔓延的路径和灭火救援提供了依据。
In order to be able to judge the cabin hazard level in real time in different types of cabins in the fire, this paper proposes an evaluation model that can judge the hazard levels of different cabins in the fire in real time. Firstly, a single-layer multi-chamber platform was designed by using Pyrosim software to analyze the spreading law of fire smoke qualitatively and quantitatively. Four levels are divided in line with the temperature, $ {\mathrm{C}\mathrm{O}}_{2} $ concentration, $ \mathrm{C}\mathrm{O} $ concentration, $ {\mathrm{O}}_{2} $ concentration and soot density which have significant influence on people. Then, taking advantage of 6-layer BP neural network to classify the hazard levels of different cabins in real time. The test results show that the classification accuracy of this model reaches 96%, which provides basis and foundation for evaluating the path of the fire and rescue work.
2020,42(10): 72-77 收稿日期：2019-10-09
DOI：10.3404/j.issn.1672-7649.2020.10.015
分类号：U665；X932；TP183
基金项目：产业前瞻与共性关键技术——竞争项目(BE2018109)
作者简介：徐昊(1995-)，男，硕士研究生，主要从事邮轮火灾研究
参考文献：
[1] Til Baalisampang, Rouzbeh Abbassi, Vikram Garaniya, Faisal Khan, Mohammad Dadashzadeh, Review and analysis of fire and explosion accidents in maritime transportation, Ocean Engineering, Volume 158, 2018, Pages 350−366.
[2] 陈国庆, 陆守香. 船舶火灾安全工程研究现状[J]. 消防技术与产品信息, 2004(8): 21-24
CHEN Guo-qing, LU Shou-xiang. Research status of ship fire safety engineering[J]. Fire Technique and Products Information, 2004(8): 21-24
[3] 黄衍顺, 陈梦华, 马焱. 船舶机舱火灾风险评估[J]. 中国修船, 2009, 22(5): 15-18
HUANG Yan-shun, CHEN Meng-hua, Ma Yan, et al. Fire risk evaluation of the engine room[J]. China Ship Repair, 2009, 22(5): 15-18
[4] 姚绪梁, 蔡晶. 一种船舶舱室火灾连锁报警等级评估方法[J]. 舰船科学技术, 2013, 35(8): 6-10
YAO Xu-liang, CAI Jing. One grade evaluation method of ship cabin fire chain alarm[J]. Ship Science and Technology, 2013, 35(8): 6-10
[5] WEI Ya-yun, ZHANG Jing-yan, WANG Jia, Research on building fire risk fast assessment method based on fuzzy comprehensive evaluation and SVM[J]. Procedia Engineering, Volume 211, 2018, Pages 1141−1150.
[6] 沈惠忠. 浅析船舶火灾特点及处置对策[J]. 水上消防, 2018(6): 36-40
SHEN Hui-zhong. Analysis on the characteristics of ship fire and its countermeasures[J]. Marine Fire, 2018(6): 36-40
[7] 王位. 面对浓烟如何躲避[J]. 安全生产与监督, 2015(2): 55-55
[8] 徐志胜, 常玉锋, 白国强, 等. 高层建筑火灾防排烟的研究[J]. 西部探矿工程, 2003, 15(12): 179-181
[9] 王友博. 火灾烟气致死原因分析[J]. 消防科学与技术, 2003(S1): 109-110
WANG You-bo. Causes of death from fire smoke[J]. Fire Science and Technology, 2003(S1): 109-110
[10] 上虞消防. 火灾中的三大杀手[EB/OL]. https://kknews.cc/news/6nppyav.html, 2017-02-09.
[11] 厦门湖里消防. 火灾死亡原因最高的是因为窒息!窒息!窒息![EB/OL]. http://www.sohu.com/a/274782164_479406, 2018-11-12.
[12] RUMELHART, D E. HINTION, G E. WILLIAMS, R J. Learning representations by back-propagating errors. Nature. 8 October 1986, 323 (6088): 533−536.
[13] KRIZHEVSKY. A, SUTSKEVER. I, HINTON. G. ImageNet classification with deep convolutional neural networks. In NIPS, 2012.
