|本期目录/Table of Contents|

[1]孟晓静,王玮丹,李为君,等.双参数驱动的高层宿舍火源信息反演预测研究*[J].中国安全生产科学技术,2025,21(5):157-164.[doi:10.11731/j.issn.1673-193x.2025.05.020]
　MENG Xiaojing,WANG Weidan,LI Weijun,et al.Research on inversion prediction of fire source information in high-rise dormitory driven by double parameters[J].Journal of Safety Science and Technology,2025,21(5):157-164.[doi:10.11731/j.issn.1673-193x.2025.05.020]
点击复制

双参数驱动的高层宿舍火源信息反演预测研究*

《中国安全生产科学技术》[ISSN:1673-193X/CN:11-5335/TB]

卷:
21
期数:
2025年5期
页码:
157-164
栏目:
职业安全卫生管理与技术
出版日期:
2025-05-30

文章信息/Info

Title:
Research on inversion prediction of fire source information in high-rise dormitory driven by double parameters
文章编号:
1673-193X(2025)-05-0157-08
作者:
孟晓静; 王玮丹; 李为君; 陈磊
（1.西安建筑科技大学 资源工程学院，陕西 西安 710055；
2.中建丝路建设投资有限公司，陕西 西安 710065）
Author(s):
MENG Xiaojing;  WANG Weidan;  LI Weijun;  CHEN Lei
(1.School of Resources Engineering,Xi’an University of Architecture and Technology,Xi’an Shaanxi 710055,China;
2.China Construction Silk Road Construction Investment Co.,Ltd.,Xi’an Shaanxi 710065,China)
关键词:
高层宿舍; 火灾源强; 反演预测; 深度学习; TCN; LSTM
Keywords:
high-rise dormitory;  fire source intensity;  inversion prediction;  deep learning;  TCN;  LSTM
分类号:
X932；TP183
DOI:
10.11731/j.issn.1673-193x.2025.05.020
文献标志码:
A
摘要:
为了提升高层宿舍火灾中火源位置与强度的反演预测精度，基于温度与CO体积分数双参数传感器探测数据，结合数值模拟与深度学习方法，逆向反演火源信息，利用火灾动力学软件FDS对某校高层宿舍楼进行火灾模拟仿真，得到温度与CO体积分数双参数传感器数据，建立火灾场景数据库，结合时间卷积网络（TCN）与长短期记忆网络（LSTM）算法优势，构建1种基于双参数的高层宿舍火灾位置与强度实时预测模型。研究结果表明：火源信息显著影响温度与CO体积分数变化，利用双参数数据能够准确反演预测火源信息。模型对火源位置、火源强度预测准确率均超过95%，火源位置与强度联合预测准确率可达90.50%；当传感器损失率小于30%时，联合预测准确率仍能达82.1%以上。研究结果可为高层宿舍火源信息预测提供参考。
Abstract:
In order to improve the inversion prediction accuracy of fire source position and intensity in high-rise dormitory fires,based on the detection data collected from dual-parameter sensors measuring the temperature and CO volume fraction,the reverse inversion of fire source information was conducted combining with the numerical simulation and deep learning methods.The fire simulation was carried out on the high-rise dormitory of an university by using the fire dynamics software FDS,then the dual-parameter sensors data of temperature and CO volume fraction were collected to establish a fire scenario database.Taking advantages of temporal convolutional networks (TCN) and long short-term memory (LSTM) algorithms,a real-time prediction model for the fire source position and intensity in high-rise dormitory based on dual-parameter data was constructed.The results show that the fire source information significantly affect the change of temperature and CO volume fraction,and the dual-parameter data can be used to accurately invert and predict the fire source information.The prediction accuracies of the model on both fire source position and intensity exceed 95%,and the combined prediction accuracy of fire source position and intensity can reach 90.50%.When the sensor loss rate is less than 30%,the combined prediction accuracy can still reach more than 82.1%.The research results can provide a reference for predicting the fire source information in high-rise dormitories.

参考文献/References:

［1］吴书强,邵必林,边根庆,等.基于多态模糊贝叶斯网络的高校宿舍火灾风险评价研究［J］.灾害学,2023,38(2):31-36. WU Shuqiang,SHAO Bilin,BIAN Genqing,et al.Fire risk assessment of university dormitory based on polymorphic fuzzy bayesian network［J］.Journal of Catastrophology,2023,38(2):31-36.
［2］曾彦夫,李逸舟,黄鑫炎.应用深度学习模型预测复杂平面房间内的火灾温度场［J］.消防科学与技术,2024,43(1):51-55，64. ZENG Yanfu,LI Yizhou,HUANG Xinyan.Deep-learning prediction on fire-induced temperature field in complex room layouts［J］.Fire Science and Technology,2024,43(1):51-55，64.
［3］李伟,胡淋翔,杨满江,等.基于人工神经网络的烟气及温度实时预测模型［J］.中国安全生产科学技术,2023,19(3):5-12. LI Wei,HU Linxiang,YANG Manjiang,et al.Real-time prediction model of smoke and temperature based on artificial neural network［J］.Journal of Safety Science and Technology,2023,19(3):5-12.
［4］李正周,方朝阳,顾园山,等.基于无线多传感器信息融合的火灾检测系统［J］.数据采集与处理,2014,29(5):694-698. LI Zhengzhou,FANG Chaoyang,GU Yuanshan,et al.Fire detection system based on wireless multi-sensor information fusion［J］.Journal of Data Acquisition and Processing,2014,29(5):694-698.
［5］刘全义,朱博,胡林,等.基于贝叶斯网络的双参数火灾探测系统［J］.南京工业大学学报(自然科学版),2022,44(6):684-690. LIU Quanyi,ZHU Bo,HU Lin,et al.Two-parameters fire detection system based on Bayesian network［J］.Journal of Nanjing University of Technology (Natural Science Edition),2022,44(6):684-690.
［6］刘全义,吴孟洋,艾洪舟,等.基于多特征参数的GA-WOA-BP火灾概率预测模型研究［J］.消防科学与技术,2024,43(6):820-825. LIU Quanyi,WU Mengyang,AI Hongzhou,et al.Research on GA-WOA-BP fire probability prediction model based on multi-feature parameters［J］.Fire Science and Technology,2024,43(6):820-825.
［7］杨云浩,张国维,朱国庆,等.基于机器学习的火源热释放速率预测方法［J］.清华大学学报(自然科学版),2024,64(5):922-932. YANG Yunhao,ZHANG Guowei,ZHU Guoqing,et al.Machine learning based prediction method for the heat release rate of a fire source［J］.Journal of Tsinghua University(Science and Technology),2024,64(5):922-932.
［8］刘阔,刘文元,曾会彬,等.基于多传感器信息融合的火灾探测和自动化预警系统设计［J］.自动化与仪表,2024,39(5):89-93. LIU Kuo,LIU Wenyuan,ZENG Huibin,et al.Design of fire detection and automated early warning system based on multi sensor information fusion［J］.Automation & Instrumentation,2024,39(5):89-93.
［9］QU N,LI Z Z,LI X X,et al.Multi-parameter fire detection method based on feature depth extraction and stacking ensemble learning model［J］.Fire Safety Journal,2022,128:103541.
［10］金晨,蔡光强,陈果,等.基于无线传感网和数据融合技术的火灾报警系统设计［J］.仪表技术与传感器,2016(6):66-68. JIN Chen,CAI Guangqiang,CHEN Guo,et al.Design of fire alarm system based on wireless sensor networks and sensor fusion technology［J］.Instrument Technique and Sensor,2016(6):66-68.
［11］郭超,闫治国.基于多源数据深度学习的地下空间智慧防灾系统框架研究［J］.现代隧道技术,2020,57(增刊1):13-23. GUO Chao,YAN Zhiguo.Research on the framework of tunnel intelligent disaster prevention system based on multi-source deep learning［J］.Modern Tunnelling Technology,2020,57(Supplement 1):13-23.
［12］YU M M,YUAN H Y,LIU G,et al.Building fire alarm model based on fire source inversion according to smoke arrival time intervals［J］.Journal of Building Engineering,2023,73:106650.
［13］BU F J,GHARAJEH M S.Intelligent and vision-based fire detection systems:a survey［J］.Image and Vision Computing,2019,91:103803.
［14］LI L J,YE J J,WANG C Y,et al.A fire source localization algorithm based on temperature and smoke sensor data fusion［J］.Fire Technology,2023,59:663-690
［15］ZHANG X N,WU X Q,HUANG X Y.Smart real-time forecast of transient tunnel fires by a dual-agent deep learning model［J］.Tunnelling and Underground Space Technology,2022,129:104631.
［16］LI Y,SU Y M,ZENG X Y,et al.Research on multi-sensor fusion indoor fire perception algorithm based on improved TCN［J］.Sensors,2022,22(12):4550.
［17］SUN B,LI Y,ZHANG Y,et al.Multi-source heterogeneous data fusion prediction technique for the utility tunnel fire detection［J］.Reliability Engineering & System Safety,2024,248:110154.
［18］WEI K Y.Research on fire alarm network for high-rise buildings based on wireless sensor networks［J］.Science and Technology of Engineering,Chemistry and Environmental Protections,2024,1(7):1102-1105.
［19］马卓,李亚飞,卢洋,等.用于早期火灾探测的红外一氧化碳传感系统［J］.光子学报,2022,51(5):328-339. MA Zhuo,LI Yafei,LU Yang,et al.Infrared carbon monoxide sensor system for early fire detection［J］.Acta Photonica Sinica,2022,51(5):328-339.
［20］中华人民共和国住房和城乡建设部.火灾自动报警系统设计规范:GB 50116—2013 ［S］.北京:中国计划出版社,2014.
［21］中华人民共和国住房和城乡建设部.消防设施通用规范:GB 55036—2022 ［S］.北京:中国计划出版社,2022.

相似文献/References:

备注/Memo

备注/Memo:
收稿日期： 2024-11-04
* 基金项目： 中建丝路科技研发课题（20240286）
作者简介： 孟晓静，博士，教授，主要研究方向为城市公共安全及防灾减灾。
更新日期/Last Update: 2025-05-26