侧向多开口地铁列车和隧道温度特征参数预测

doi:10.16265/j.cnki.issn1003-3033.2026.01.1196

中国安全科学学报 ›› 2026, Vol. 36 ›› Issue (1): 130-137.doi: 10.16265/j.cnki.issn1003-3033.2026.01.1196

侧向多开口地铁列车和隧道温度特征参数预测

吴振坤¹^,²(), 彭敏²^,^**(), 朱国庆², 刘璐², 秦东子³

¹ 无锡学院应急管理学院,江苏无锡 214105
² 中国矿业大学安全工程学院,江苏徐州 221116
³ 中国科学技术大学火灾科学国家重点实验室,安徽合肥 230026

收稿日期:2025-09-05 修回日期:2025-11-22 出版日期:2026-01-28
通信作者:
** 彭敏(1995—),女,安徽宿州人,博士,副教授,主要从事城市轨道火灾安全研究。E-mail:pengmin@cumt.edu.cn。
作者简介:
吴振坤 (1990—),男,甘肃张掖人,博士,讲师,主要从事轨道交通火灾防控及监测预警等方面的研究。E-mail:TB21120027B0@cumt.edu.cn。
朱国庆, 教授。
基金资助:
国家重点研发计划项目(2023YFC3009900); 国家自然科学基金(52574289); 国家自然科学基金(52204254); 江苏省自然科学基金(BK20221124); 公共火灾防治技术四川省高等学校重点实验室开放基金(SC_KLPFPCT2024Z02)

Study on prediction of temperature characteristic parameters for subway train with multiple lateral openings and tnnnels

WU Zhenkun¹^,²(), PENG Min²^,^**(), ZHU Guoqing², LIU Lu², QIN Dongzi³

¹ School of Emergency Management, Wuxi University, Wuxi Jiangsu 214105, China
² School of Safety Engineering, China University of Mining and Technology, Xuzhou Jiangsu 221116, China
³ State Key Laboratory of Fire Science, University of Science and Technology of China, Hefei Anhui 230026, China

Received:2025-09-05 Revised:2025-11-22 Published:2026-01-28

摘要/Abstract

摘要：

为解决现有地铁列车和隧道火灾预测方法大多依赖于物理模型和经验公式而导致预测精度不足的问题,从人工智能角度出发,基于遗传算法(GA)优化反向传神经网络(BPNN),构建GA-BPNN网络模型;利用GA对BPNN的权重和阈值进行全局寻优;训练与预测车厢与隧道顶棚温度分布,智能反演火灾温度场。结果表明:GA-BPNN 模型对车厢温度预测的平均绝对误差(MAE)为8.17,均方根误差(RMSE)为9.76,决定系数R²为0.99;隧道温度预测的MAE为3.95,RMSE为5.63,R²为0.98。通过对比发现,GA-BPNN模型在准确性和泛化能力上都优于传统BPNN模型。

关键词: 侧向多开口, 地铁列车, 温度预测, 特征参数, 反向传播神经网络(BPNN), 遗传算法(GA)

Abstract:

Existing methods for predicting temperature from subway carriage and tunnel fires mainly rely on physical models and empirical methods that are valid only under narrowly-defined environmental conditions. To solve this issue, this study adopts an artificial-intelligence-based approach. A GA-BPNN network model is constructed by optimizing BPNN using a GA. The GA is employed to global optimize BPNN's weights and thresholds, after which the model is trained to predict the temperature distribution of both the subway carriage and the tunnel, thereby achieving intelligent inversion of the fire temperature field. The results show that, for subway carriage temperature prediction, GA-BPNN model yields a mean absolute error (MAE) of 8.17, a root mean square error (RMSE) of 9.76, and a coefficient of determination (R²) of 0.99. For tunnel temperature prediction, MAE is 3.95, RMSE is 5.63, and R² reaches 0.98. By comparing the results with those of the traditional BPNN, it is found that the GA-BPNN model outperforms the conventional BPNN in both prediction accuracy and generalization capability.

Key words: multiple lateral openings, subway train, temperature prediction, characteristic parameters, back-propagation neural network(BPNN), genetic algorithm(GA)

中图分类号:

X951交通运输安全

吴振坤, 彭敏, 朱国庆, 刘璐, 秦东子. 侧向多开口地铁列车和隧道温度特征参数预测[J]. 中国安全科学学报, 2026, 36(1): 130-137.

WU Zhenkun, PENG Min, ZHU Guoqing, LIU Lu, QIN Dongzi. Study on prediction of temperature characteristic parameters for subway train with multiple lateral openings and tnnnels[J]. China Safety Science Journal, 2026, 36(1): 130-137.

图/表 15

图1

表1

表2

表3

图2

图3

图4

图5

图6

图7

图8

图9

图10

图11

图12

参考文献 21

[1]	汤青慧, 刘硕, 张瑾. 基于WSR-可拓云理论的地铁车站火灾安全韧性评估[J]. 中国安全科学学报, 2025, 35(1): 154-162. doi: 10.16265/j.cnki.issn1003-3033.2025.01.1377
	TANG Qinghui, LIU Shuo, ZHANG Jin. Resilience assessment of fire safety in subway stations based on WSR-Extension cloud theory[J]. China Safety Science Journal, 2025, 35(1):154-162. doi: 10.16265/j.cnki.issn1003-3033.2025.01.1377
[2]	彭敏, 潘翔骁, 何坤, 等. 侧向多开口轨道列车火灾烟气输运特性[J]. 中国安全科学学报, 2024, 34(5): 129-138. doi: 10.16265/j.cnki.issn1003-3033.2024.05.0758
	PENG Min, PAN Xiangxiao, HE Kun, etal. Study on smoke transport characteristics of fire inside rail train with multiple lateral openings[J]. China Safety Science Journal, 2024, 34(5):129-138. doi: 10.16265/j.cnki.issn1003-3033.2024.05.0758
[3]	雷斌, 杜依潼, 寻天祥, 等. 地铁换乘站火灾排烟及疏散路径研究[J]. 中国安全生产科学技术, 2025, 21(6): 140-148.
	LEI Bin, DU Yitong, XUN Tianxiang, et al. Research on fire smoke exhaust and evacuation paths in subway transfer station[J]. Journal of Safety Science and Technology, 2025, 21(6): 140-148.
[4]	NG Y W, CHOW W K, CHENG Chihonn, et al. Scale modeling study on flame colour in a ventilation-limited train car pool fire[J]. Tunnelling and Underground Space Technology, 2019, 85: 375-391. doi: 10.1016/j.tust.2018.12.026
[5]	姜学鹏, 肖倪琪, 路世昌. 不同车门开闭状态下地铁车厢火灾特性研究[J]. 消防科学与技术, 2021, 40(1): 76-81.
	JIANG Xuepeng, XIAO Niqi, LU Shichang. Fire characteristics of metro cabin under different doors opening and closing conditions[J]. Fire Science and Technology, 2021, 40(1): 76-81.
[6]	SHI Congling, ZHONG Maohua, CHEN Changkun, et al. Metro train carriage combustion behaviors:full-scale experiment study[J]. Tunnelling and Underground Space Technology, 2020, 104: DOI:10.1016/j.tust.2020.103544.
[7]	彭敏. 侧向多开口地铁列车车厢火灾燃烧特性研究[D]. 合肥: 中国科学技术大学, 2021.
	PENG Min. Study on combustion characteristics of fires inside subway train with multiple lateral openings[D]. Hefei: University of Science and Technology of China, 2021.
[8]	INGASON H. Model scale railcar fire tests[J]. Fire Safety Journal, 2007, 42(4): 271-282. doi: 10.1016/j.firesaf.2006.11.004
[9]	LÖNNERMARK A, INGASON H, LI Yingzhen et al. Fire development in a 1/3 train carriage mock-up[J]. Fire Safety Journal, 2017, 91: 432-440. doi: 10.1016/j.firesaf.2017.03.016
[10]	PENG Min, SHI Long, HE Kun, et al. Experimental study on fire plume characteristics in a subway carriage with doors[J]. Fire Technology, 2020, 56(2): 401-423. doi: 10.1007/s10694-019-00882-0
[11]	PENG Min, MING Yuyang, WU Zhenkun, et al. Experimental study on fire characteristics of sidewall fires in a long-narrow compartment with multiple lateral openings[J]. Case Studies in Thermal Engineering, 2023, 51: DOI:10.1016/j.csite.2023.103515.
[12]	钟茂华, 程辉航, 陈俊沣, 等. 地铁车站火灾全尺寸实验设计与工程实践[J]. 实验技术与管理, 2022, 39(9): 1-8.
	ZHONG Maohua, CHENG Huihang, CHEN Junfeng, et al. Full-scale experimental design and engineering practice of metro station fire[J]. Experimental Technology and Management, 2022, 39(9): 1-8.
[13]	XI Yanhong, ZHOU Zhihan, MAO Jun, et al. Maximum ceiling temperature and longitudinal distribution in a corridor-like enclosure with opening[J]. Tunnelling and Underground Space Technology, 2023, 134:DOI:10.1016/j.tust.2023.104994.
[14]	XI Yanhong, ZHOU Zhihan, LIAN Huanqiang, et al. Temperature variation inside a corridor-like enclosure under limited ventilation[J]. Tunnelling and Underground Space Technology, 2022, 126: DOI:10.1016/j.tust.2022.104539.
[15]	YANG Juntao, YANG Yun, WANG Helan, et al. Effect of the open ways of screen doors on fire smoke in a subway platform[J]. Procedia Engineering, 2011, 11: 416-423. doi: 10.1016/j.proeng.2011.04.677
[16]	HUANG Dongmei, WANG Panpan, LI Wei, et al. Modeling, response and BPNN-PID control of symmetric multistable galloping energy harvesters[J]. Chaos, Solitons & Fractals, 2025, 199: DOI:10.1016/j.chaos.2025.116736.
[17]	BEYLER C L, HIRSCHLER M M. The SFPE handbook of fire protection engineering[M]. London: Springer New York, 2016: 102-134.
[18]	吕淑然, 田江雪, 党鑫宇. 基于改进灰狼优化BP网络的城中村火灾预测[J]. 中国安全科学学报, 2025, 35(8):196-204. doi: 10.16265/j.cnki.issn1003-3033.2025.08.1426
	LYU Shuran, TIAN Jiangxue, DANG Xinyu. Fire prediction in urban villages based on improved grey wolf optimization BP network[J]. China Safety Science Journal, 2025, 35(8):196-204. doi: 10.16265/j.cnki.issn1003-3033.2025.08.1426
[19]	黄欣然. 时变路网下接驳轨道交通的社区公交路径规划研究[D]. 大连: 大连交通大学, 2025.6.
	HUANG Xinran. Research on route planning of community bus connecting rail transit under time-dependent road network[D]. Dalian: Dalian Jiaotong University, 2025.6.
[20]	赵思琦. 突发事件下城市轨道交通网络韧性修复分析[D]. 大连: 大连交通大学, 2025.6.
	ZHAO Siqi. Analysis of urban rail transit network resilience repair under emergency events[D]. Dalian: Dalian Jiaotong University, 2025.6.
[21]	白慧. 基于神经网络的隧道火灾温度能见度预测研究[D]. 成都: 西南交通大学, 2022.6.
	BAI Hui. Research on temperature and visibility prediction of tunnel fire based on neural network[D]. Chengdu: Southwest Jiaotong University, 2025.6.

序号	热释放速率/MW
1	2.5
2	5.0
3	7.5
4	10

序号	长×宽×高/(m×m×m)	网格总数
1	0.1×0.1×0.1	10 368 000
2	0.2×0.2×0.2	1 260 000
3	0.4×0.4×0.4	162 000
4	0.5×0.5×0.5	82 944

迭代次数	RMSE	R²
250	5.403 0	0.971 7
500	5.185 4	0.973 3
1 000	5.617 4	0.970 0
1 500	6.008 9	0.970 1
2 000	6.779 7	0.968 7
3 000	7.283 7	0.964 5

侧向多开口地铁列车和隧道温度特征参数预测

Study on prediction of temperature characteristic parameters for subway train with multiple lateral openings and tnnnels

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

图/表 15

参考文献 21

相关文章 15

编辑推荐

Metrics

本文评价

[1]	陈铁华, 刘瑞康, 李红霞. 基于SD-BPNN法的矿工安全生产责任压力演化建模与仿真[J]. 中国安全科学学报, 2025, 35(9): 70-77.
[2]	汪伟, 崔欣超, 祁云, 李绪萍, 王璜瑞, 齐庆杰. 改进SSA优化BPNN的煤体瓦斯渗透率预测模型[J]. 中国安全科学学报, 2025, 35(2): 137-143.
[3]	郝景开, 李红艳, 张峰, 张翀, 毛立波, 刘大为. 基于MCDM-BPNN的城市内涝风险评价及调蓄池选址[J]. 中国安全科学学报, 2024, 34(8): 214-221.
[4]	彭敏, 潘翔骁, 何坤, 李忠辉, 朱国庆. 侧向多开口轨道列车火灾烟气输运特性[J]. 中国安全科学学报, 2024, 34(5): 129-138.
[5]	周怡, 彭国文, 黄召, 阳鹏飞, 刘丹丹, 陈小丽. 基于SBAS-InSAR和BPNN的铀尾矿坝形变智能监测与预测[J]. 中国安全科学学报, 2024, 34(4): 145-152.
[6]	井庆贺, 张启良, 王增仁, 刘东星, 谢苗, 孟庆爽. 带式输送机中间段托辊故障检测方法研究[J]. 中国安全科学学报, 2023, 33(S2): 41-48.
[7]	陈伟, 田仪帅, 曾卫华, 郭道远, 赵卓雅. 强降雨导致的建设工程安全事故人因分析HC-GC模型[J]. 中国安全科学学报, 2023, 33(8): 15-23.
[8]	汪伟, 梁然, 祁云, 贾宝山, 武泽伟. 基于PSO-BPNN的煤自燃危险性预测模型[J]. 中国安全科学学报, 2023, 33(7): 127-132.
[9]	谢露强, 徐靖, 王海燕. 基于RS-RBFNN的邮轮建造物资物流集配风险预警[J]. 中国安全科学学报, 2023, 33(6): 114-121.
[10]	郄彦辉, 郭涛, 周凌志, 王昱. 基于SVM的含缺陷20钢弯管爆破压力预测[J]. 中国安全科学学报, 2023, 33(2): 89-95.
[11]	余强, 黄逸橙, 沙策. 台风特征参数对灾害链概率计算的影响分析*[J]. 中国安全科学学报, 2022, 32(8): 168-175.
[12]	朱江洪, 王睿, 李延来. 基于Fine-Kinney与TODIM的地铁列车车门系统风险评估[J]. 中国安全科学学报, 2019, 29(10): 167-173.
[13]	柴超君, 凌永生, 岳琪, 贾文宝. 维数约简用于BPNN的核事故源项估算方法[J]. 中国安全科学学报, 2018, 28(9): 122-127.
[14]	谢尊贤, 郭琰, 蒲涛. 基于BIM-GA-BN的养老社区施工安全评价模型[J]. 中国安全科学学报, 2018, 28(9): 19-25.
[15]	白晓平, 蒲涛. 数据驱动的施工安全评价模型研究与应用[J]. 中国安全科学学报, 2017, 27(9): 134-139.