[Objective] In case of catching fire, a train moving in a tunnel section might lose power and come to an emergency halt while approaching the next station for emergency rescue. Predicting the distribution of smoke and temperature in tunnels under such fire scenarios is difficult because of the influence of moving trains. This difficulty in prediction might seriously threaten the safety of passengers and impede metro evacuation management. [Methods] This study considers influencing factors, including train speed, fire source location, and fire heat release rate, to solve this problem, and 75 different fire scenarios are designed and simulated. Train movement in the simulated scenarios is realized using the equivalent piston wind method. The simulation results of the smoke and temperature distributions collected using sensors near the tunnel ceiling are used to construct a dataset for deep learning. Accordingly, a deep learning model comprising long short-term memory networks, a convolution (Conv) module, and a deconvolution (DeConv) module is then proposed for rapid prediction of temperature distribution in tunnels under moving train fire conditions. The train speed, train braking time, and temperature time-series information from the sensors together are fed as inputs to the model. [Results] The results indicated that: (1) Under various train movement states, the model was able to predict the temperature distribution of the lateral evacuation platform in the tunnel 30 s in advance using the current sensor data, with a mean absolute error (MAE) of only 2.2 ℃ and a mean absolute percentage error (MAPE) of 4%, indicating high accuracy. (2) In a stark contrast with the week-long time taken to obtain temperature distribution in a fire dynamics simulator (FDS), this deep learning model could make prediction within only 0.08 s, hence representing a computational efficiency improvement of four orders of magnitude versus the computational fluid dynamics method. (3) Validation with fire scenarios in none of the training, validation, and test datasets resulted in model MAE and MAPE values of 3.1 ℃ and 5%, respectively, indicating a strong generalization ability. (4) Considering the possibility of sensor failure within tunnels, this study investigated the influence of simulated sensor failures on the model's prediction accuracy by varying sensor spacing. The model continued to exhibit an effective predictive ability even at a sensor spacing of 4.00 m. At 8.00 m sensor spacing, the model's errors were larger albeit at very few time frames (with a maximum MAE lower than 10.0 ℃ and a maximum MAPE lower than 15%). However, for other sensor spacing cases, the model's MAE and MAPE were less than 5.0 ℃and 10%, respectively. Hence, it could be concluded that the model has strong robustness. [Conclusions] This study constructs a comprehensive dataset for a tunnel with moving train fire conditions using an FDS and leverages advanced deep neural networks to completely exploit the extensive information within the dataset, ultimately resulting in a high-precision, robust model for rapid prediction of temperature distribution in tunnels under moving train fire conditions. These advancements are highly important for effective emergency management and response planning in tunnels under these challenging conditions.
HUANG Ran
,
ZHU Shiyou
,
HE Mengchen
,
LI Ruoyu
,
GE Xinru
,
WANG Qiao
,
CHEN Juan
,
LO Jacqueline T. Y.
,
MA Jian
. Rapid prediction method for tunnel temperature distributions during fires on moving trains[J]. Journal of Tsinghua University(Science and Technology), 2025
, 65(3)
: 479
-494
.
DOI: 10.16511/j.cnki.qhdxxb.2025.26.013
[1] 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. ZHONG M H, LIU C, SHI C L. Progress of full-scale experimental study on metro fire [J]. China Safety Science Journal, 2019, 29(10): 51-63. (in Chinese)
[2] 袁勇, 邱俊男. 地铁火灾的原因与统计分析[J]. 城市轨道交通研究, 2014, 17(7): 26-31, 61. YUAN Y, QIU J N. Statistical analysis of metro fire cases [J]. Urban Mass Transit, 2014, 17(7): 26-31, 61. (in Chinese)
[3] 中华人民共和国住房和城乡建设部. 地铁设计规范: GB 50157—2013[S]. 北京: 中国建筑工业出版社, 2014. Ministry of Housing and Urban-Rural Development of the People's Republic of China. Code for design of metro: GB 50157—2013[S]. Beijing: China Architecture & Building Press, 2014. (in Chinese)
[4] 钟茂华, 刘畅. 地铁隧道火灾防控全尺寸实验研究进展[J]. 中国科学基金, 2021, 35(6): 878-884. ZHONG M H, LIU C. Progress of full-scale experimental study for fire prevention and control in metro tunnel [J]. Bulletin of National Natural Science Foundation of China, 2021, 35(6): 878-884. (in Chinese)
[5] MCGRATTAN K, MILES S. Modeling fires using computational fluid dynamics (CFD) [M]//HURLEY M J, GOTTUK D, HALL J R, et al. SFPE handbook of fire protection engineering. 5th ed. New York: Springer, 2016: 1034-1065.
[6] YANG Y X, LIU C, LONG Z, et al. Full-scale experimental study on fire under vehicle operations in a sloped tunnel [J]. International Journal of Thermal Sciences, 2020, 158: 106524.
[7] CHEN L F, ZHOU S X, LAN Y J, et al. Experimental study on fire characteristics of the bifurcated tunnel under multi-directional ventilation [J]. International Journal of Thermal Sciences, 2023, 185: 108072.
[8] HUANG P, YE S L, XIE J X, et al. Study on the maximum and longitudinal distribution of ceiling gas temperature in a naturally ventilated tunnel: The effect of longitudinal fire location [J]. International Journal of Thermal Sciences, 2023, 185: 108037.
[9] 宋卫国, 马剑, SATOH K, 等. 森林火险与气象因素的多元相关性及其分析[J]. 中国工程科学, 2006, 8(2): 61-66. SONG W G, MA J, SATOH K, et al. ANN analysis of multi-correlation between forest fire risk and weather parameters [J]. Engineering Sciences, 2006, 8(2): 61-66. (in Chinese)
[10] HODGES J L, LATTIMER B Y. Wildland fire spread modeling using convolutional neural networks [J]. Fire Technology, 2019, 55(6): 2115-2142.
[11] ZHAI C J, ZHANG S Y, CAO Z L, et al. Learning-based prediction of wildfire spread with real-time rate of spread measurement [J]. Combustion and Flame, 2020, 215: 333-341.
[12] NGUYEN N T, DANG B T N, PHAM X C, et al. Spatial pattern assessment of tropical forest fire danger at Thuan Chau area (Vietnam) using GIS-based advanced machine learning algorithms: A comparative study [J]. Ecological Informatics, 2018, 46: 74-85.
[13] LI Z Z, HUANG Y F, LI X X, et al. Wildland fire burned areas prediction using long short-term memory neural network with attention mechanism [J]. Fire Technology, 2021, 57(6): 1-23.
[14] XUE C J. The road tunnel fire detection of multi-parameters based on BP neural network [C]//Proceedings of the 20102nd International Asia Conference on Informatics in Control, Automation and Robotics (CAR 2010). Wuhan, China: IEEE, 2010: 246-249.
[15] WU X Q, PARK Y, LI A, et al. Smart detection of fire source in tunnel based on the numerical database and artificial intelligence [J]. Fire Technology, 2021, 57(2): 657-682.
[16] LI M J, JIANG Y, WU Z, et al. Real-time prediction of smoke spread affected by multiple factors in subway tunnel using CAERES-DNN model [J]. Fire Technology, 2021, 57(4): 2025-2059.
[17] 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.
[18] GUO C, GUO Q H, ZHANG T, et al. Study on real-time heat release rate inversion for dynamic reconstruction and visualization of tunnel fire scenarios [J]. Tunnelling and Underground Space Technology, 2022, 122: 104333.
[19] 中华人民共和国铁道部. 铁路隧道运营通风设计规范: TB 10068—2010[S]. 北京: 中国铁道出版社, 2010. Ministry of Railways of the People's Republic of China. Code for design on operating ventilation of railway tunnel: TB 10068—2010[S]. Beijing: China Railway Publishing House, 2010. (in Chinese)
[20] 张子洋. 特长隧道内高速列车移动火灾烟气蔓延规律研究[D]. 成都: 四川师范大学, 2021. ZHANG Z Y. Research on smoke spreading regularity of high- speed train moving fire in extra-long tunnel [D]. Chengdu: Sichuan Normal University, 2021. (in Chinese)
[21] BABRAUSKAS V, PEACOCK R D. Heat release rate: The single most important variable in fire hazard [J]. Fire Safety Journal, 1992, 18(3): 255-272.
[22] ZHANG S G, YAO Y Z, ZHU K, et al. Prediction of smoke back-layering length under different longitudinal ventilations in the subway tunnel with metro train [J]. Tunnelling and Underground Space Technology, 2016, 53: 13-21.
[23] 中华人民共和国住房和城乡建设部, 国家市场监督管理总局. 地铁设计防火标准: GB 51298—2018[S]. 北京: 中国计划出版社, 2018. Ministry of Housing and Urban-Rural Development of the People's Republic of China, State Administration for Market Regulation. Standard for fire protection design of metro: GB 51298—2018[S]. Beijing: China Planning Press, 2018. (in Chinese)
[24] MCGRATTAN K, MCDERMOTT R, HOSTIKKA S, et al. Fire dynamics simulator(version 5): User's guide [EB/OL].(2010-10-29) [2024-06-07]. http://fire.fsv.cvut.cz/ifer/2014-Training_school/FDS_5_User_Guide.pdf.
[25] LI H H, ZHU W J, TANG M, et al. Burning characteristic and ceiling temperature of moving fires in a tunnel: A comparative study [J]. Tunnelling and Underground Space Technology, 2024, 145: 105571.
[26] HODGES J L, LATTIMER B Y, LUXBACHER K D. Compartment fire predictions using transpose convolutional neural networks [J]. Fire Safety Journal, 2019, 108: 102854.
