Fast prediction for building fire spread based on machine learning

doi:10.16265/j.cnki.issn1003-3033.2023.11.2219

Abstract

Abstract:

In order to quickly rehearse or reproduce the fire scene, the key influencing factors of building fire were studied by using fire dynamics software(FDS) and machine learning technology, so as to adjust the fire rescue strategy in real-time, and finally provide a space composition scheme conducive to fire engineering for architectural design. In this paper, a single-room fire smoke overflow was taken as a case, and 5 kinds of algorithm models were used to perform machine learning training and efficiency evaluation on 11 spatial composition parameters and fire conditions, a total of 7 776 sets of fire simulation result data. The experimental results show that the machine learning algorithm is suitable for parameter learning and evaluation prediction of discrete types such as building space. It can intuitively give the weight of each parameter, mine the key information in the fire dynamics system, and realize the visualization of fire data. The Random Forests(RF) algorithm has the highest prediction efficiency, and its best prediction accuracy can reach 91.82%.

Key words: building fires, rapid fire prediction, machine learning, weight analysis, spatial composition

GUO Zhen, JIA Xiaoyan, LI Fumin, HU Yan, YAN Qiuyan. Fast prediction for building fire spread based on machine learning[J]. China Safety Science Journal, 2023, 33(11): 117-125.

Figures/Tables 13

Tab.1

Fig.1

Fig.2

Fig.3

Fig.4

Tab.2

Fig.5

Fig.6

Fig.7

Fig.8

Fig.9

Fig.10

Fig.11

References 34

[1]	Government. UK. Fire statistics data tables[EB/OL]. [2022-10-28]. https://www.gov.uk/government/statistical-data-sets/fire-statistics-data-tables.
[2]	钟茂华, 厉培德, 卢兆明, 等. 多层多室建筑火灾烟气运动的网络模拟[J]. 火灾科学, 2002, 11(2): 103-107,125.
	ZHONG Maohua, LI Peide, LU Zhaoming, et al. Network model simulation of smoke spread of multi-layer and multi-room building fire[J]. Fire Safety Science, 2002, 11(2): 103-107,125.
[3]	JOHANSSON N, SVENSSON S, VAN HEES P. A study of reproducibility of a full-scale mult-room compartment fire experiment[J]. Fire Technology, 2015, 51(3):645-665. doi: 10.1007/s10694-014-0408-3
[4]	MELCHER T, ZINKE R, TROTT M, et al. Experimental investigations on the repeatability of real scale fire tests[J]. Fire Safety Journal, 2016, 82:101-114. doi: 10.1016/j.firesaf.2016.03.009
[5]	钟茂华, 刘畅, 史聪灵. 地铁火灾全尺寸实验研究进展综述[J]. 中国安全科学学报, 2019, 29(10):51-63. doi: 10.16265/j.cnki.issn1003-3033.2019.10.009
	ZHONG Maohua, LIU Chang, SHI Congling. Progress of full-scale experimental study on metro fire[J]. China Safety Science Journal, 2019, 29(10):51-63. doi: 10.16265/j.cnki.issn1003-3033.2019.10.009
[6]	陈俊沣, 程辉航, 魏旋, 等. 隧道火灾全尺寸实验中温度测量误差[J]. 清华大学学报:自然科学版, 2022, 62(10):1618-1625.
	CHEN Junfeng, CHENG Huihang, WEI Xuan, et al. Temperature measurement errors in full-scale tunnel fire experiments[J]. Journal of Tsinghua University :Science and Technology, 2022, 62(10):1618-1625.
[7]	李国能, 周昊, 凌忠钱, 等. 标准火灾测试房间温度场和速度场的大涡模拟[J]. 浙江大学学报:工学版, 2006, 40(2):294-297.
	LI Guoneng, ZHOU Hao, LING Zhongqian, et al. Large eddy simulation of temperature field and velocity field in standard fire test room[J]. Journal of Zhejiang University :Engineering Science, 2006, 40(2):294-297.
[8]	ZHANG W, HAMER A, ANDREW K, et al. Turbulence statistics in a fire room model by large eddy simulation[J]. Fire Safety Journal, 2002, 37(8):721-752. doi: 10.1016/S0379-7112(02)00030-9
[9]	朱杰. 超高层建筑竖井结构内烟气运动规律及控制研究[D]. 合肥: 中国科学技术大学, 2008.
	ZHU Jie. Studies on the characteristic of smoke movement and smoke control in shaft of ultra-high-rise buildings[D]. Hefei: University of Science and Technology of China, 2008.
[10]	朱杰, 马金梅, 彭莉. 不同通风方式下地下环形受限空间火羽流卷吸特性研究[J]. 安全与环境工程, 2019, 26(2):145-155.
	ZHU Jie, MA Jinmei, PENG Li. Entrainment characteristics of fire plume in confined underground ring space under the different ventilation modes[J]. Safety and Environmental Engineering, 2019, 26(2):145-155.
[11]	朱杰, 罗笛书. 不同火源功率下地下环道火羽流运动机理研究[J]. 消防科学与技术, 2019, 38(9):1228-1231.
	ZHU Jie, LUO Dishu. Research on fire plume entrainment characteristics of underground ring road under different HRR[J]. Fire Science and Technology, 2019, 38(9):1228-1231.
[12]	朱杰, 杨天佑, 张立龙, 等. 单纯热烟浮力作用下竖井结构内火灾烟气温度变化规律研究[J]. 工程力学, 2011, 28(8):198-207.
	ZHU Jie, YANG Tianyou, LI Lilong, et al. Temperature change law of fire smoke in shaft structure under mere hot smoke buoyancy[J]. Engineering Mechanics, 2011, 28(8):198-207.
[13]	朱杰, 张子洋, 程友鹏. 不同火源位置对地下环形受限空间火羽流运动规律影响研究[J]. 安全与环境学报, 2019, 19(4):1166-1174.
	ZHU Jie, ZHANG Ziyang, CHENG Youpeng. Impact of the different fire sources on the movement and motion regularity of the fire plume in the confined limited underground circular space[J]. Journal of Safety and Environment, 2019, 19(4):1166-1174.
[14]	童艳, 施明恒, 何嘉鹏. 单室火灾烟气运动的双层区域模拟[J]. 东南大学学报:自然科学版, 2008, 38(3):449-453.
	TONG Yan, SHU Mingheng, HE Jiapeng. Two-layer zonal modeling of smoke movement in single fire room[J]. Journal of Southeast University:Natural Science Edition, 2008, 38(3):449-453.
[15]	褚燕燕, 张辉, 杨锐, 等. 基于火源和疏散信息的建筑火灾烟气控制策略模拟[J]. 清华大学学报:自然科学版, 2010, 50(8):1158-1162.
	CHU Yanyan, ZHANG Hui, YANG Yue, et al. Building fire smoke control strategies simulation using fire and evacuation information[J]. Journal of Tsinghua University:Science and Technology, 2010, 50(8):1158-1162.
[16]	马德仲, 丁文飞, 刘圣楠, 等. 基于贝叶斯网络的地下空间火灾风险评估方法研究[J]. 中国安全科学学报, 2013, 23(11):151-156.
	MA Dezhong, DING Wenfei, LIU Shengnan, et al. Risk assessment method for fire in underground space based on Bayesian network[J]. China Safety Science Journal, 2013, 23(11):151-156.
[17]	孙树立, 韩旭东, 陈璞. 建筑结构安全分析方法与评估技术综述[J]. 中国安全科学学报, 2022, 32(9):37-48. doi: 10.16265/j.cnki.issn1003-3033.2022.09.2621
	SUN Shuli, HAN Xudong, CHEN Pu. Review on safety analysis methods and evaluation techniques of building structures[J]. China Safety Science Journal, 2022, 32(9):37-48. doi: 10.16265/j.cnki.issn1003-3033.2022.09.2621
[18]	舒舜, 张羽杨, 陆梓萍, 等. 基于机器学习的热释放速率实时预测方法[J]. 火灾科学, 2022, 31(1):8-14.
	SHU Shun, ZHANG Yuyang, LU Ziping, et al. Real-time prediction of heat release rate based on machine learning[J]. Fire Safety Science, 2022, 31(1):8-14.
[19]	朱馨, 李建微, 郭伟, 等. 基于机器学习的森林火险预测模型[J]. 中国安全科学学报, 2022, 32(9):152-157. doi: 10.16265/j.cnki.issn1003-3033.2022.09.2729
	ZHU Xin, LI Jianwei, GUO Wei, et al. Forest fire risk prediction model based on machine learning[J]. China Safety Science Journal, 2022, 32(9):152-157. doi: 10.16265/j.cnki.issn1003-3033.2022.09.2729
[20]	刘全义, 朱博, 邓力, 等. 基于机器学习的双参数火灾探测方法[J]. 中国安全科学学报, 2022, 32(5):90-96. doi: 10.16265/j.cnki.issn1003-3033.2022.05.0874
	LIU Quanyi, ZHU BO, DENG Li, et al. Double parameters fire detection method based on machine learning[J]. China Safety Science Journal, 2022, 32(5):90-96. doi: 10.16265/j.cnki.issn1003-3033.2022.05.0874
[21]	GB/50368—2005, 住宅建筑规范[S].
	GB/50368-2005, Residential building code[S].
[22]	BABRAUSKAS V. Ignition of wood: a review of the state of the art[J]. Journal of Fire Protection Engineering, 2002, 12(3):163-189. doi: 10.1177/10423910260620482
[23]	李玉栋. 木材点燃温度的测定[J]. 火灾科学, 1992, 1(1):25-30.
	LI Yudong. Measurement of the ignition temperature of wood[J]. Fire Safety Science, 1992, 1(1):25-30.
[24]	GB/50084—2017, 自动喷水灭火系统设计规范[S].
	GB/50084-2017, Code for design of sprinkler systems[S].
[25]	GAO Y, CHOW W, WU M. Thermal performance of window glass panes in an enclosure fire[J]. Construction and Building Materials, 2013, 47: 530-546. doi: 10.1016/j.conbuildmat.2013.05.027
[26]	PURSER D A. Toxic product yields and hazard assessment for fully enclosed design fires[J]. Polymer International, 2000, 49(10): 1232-1255. doi: 10.1002/(ISSN)1097-0126
[27]	刘祥君, 张兴华, 席焕久, 等. 中国汉族乡村成年人的身高与体质量的地理性分布[J]. 人类学学报, 2018, 37(3):484-495.
	LIU Xiangjun, ZHANG Xinghua, XI Huanjiu, et al. Geographical distribution of stature and body mass of rural Chinese Han adults[J]. Acta Anthropologica Sinica, 2018, 37(3):484-495.
[28]	XU Q, CHEN L, HARRIES K A, et al. Combustion and charring properties of five common constructional wood species from cone calorimeter tests[J]. Construction and Building Materials, 2015, 96:416-427. doi: 10.1016/j.conbuildmat.2015.08.062
[29]	章涛林, 方廷勇, 卢平. 高层建筑火灾烟气迁移特性研究[J]. 安徽建筑工业学院学报:自然科学版, 2008, 16(5):58-64.
	ZHANG Taolin, FANG Tingyong, LU Ping. The research of smoke transfer characteristic in high-rise building fires[J]. Journal of Anhui Jianzhu University:Science and Technology, 2008, 16(5):58-64.
[30]	MANGALATHU S, JEON J S. Classification of failure mode and prediction of shear strength for reinforced concrete beam-column joints using machine learning techniques[J]. Engineering Structures, 2018, 160: 85-94. doi: 10.1016/j.engstruct.2018.01.008
[31]	KIANI J, CAMP C, PEZESHK S. On the application of machine learning techniques to derive seismic fragility curves[J]. Comput Struct, 2019, 218:108-122. doi: 10.1016/j.compstruc.2019.03.004
[32]	MANGALATHU S, JANG H, HWANG S H, et al. Data-driven machine-learning-based seismic failure mode identification of reinforced concrete shear walls[J]. Engineering Structures, 2020, 208: DOI:10.1016/j.engstruct.2020.110331.
[33]	MANGALATHU S, JEON J S. Machine learning-based failure mode recognition of circular reinforced concrete bridge columns: comparative study[J]. Journal of Structural Engineering, 2019, 145(10): DOI:10.1061/(ASCE)ST.1943-541X.0002402.
[34]	HUANG H, BURTON H V. Classification of in-plane failure modes for reinforced concrete frames with infills using machine learning[J]. Journal of Building Engineering, 2019, 25: DOI:10.1016/j.jobe.2019.100767.

序号		开间	进深	门所在位置	门垂壁高度	窗户相对于门所在位置	窗户高度	窗垂壁高度	火源位置 -x	火源位置 -y
1	空间参数	3 000 mm	3 600 mm	开间方向	400 mm	对向	1 500 mm	200 mm	1
1	数据表征	3 000	3 600	1	400	1	1 500	200	-100	120
2	空间参数	3 300 mm	4 200 mm	进深方向	600 mm	左向	1 800 mm	400 mm	5
2	数据表征	3 300	4 200	10	600	10	1 800	400	0	0