为提升隧道火灾全物理场信息的预测速度以及解决隧道火灾关键控制参数的反向预测问题,该文构建了深度学习模型用于隧道火灾中全物理场和关键控制参数间的快速双向预测,使用了大型数值数据库训练所构建的模型,评估了模型对数据的学习能力及其预测能力。结果表明:经过 100 个训练周期后,使用所提出的双向预测模型和数据集取得了良好的训练收敛效果,物理场和关键控制参数在训练集上均达到结果快速重现。模型训练完成后,隧道火灾的平均温度场和 6 项隧道火灾的关键控制参数得到了基本的预测,预测结果同时涵盖了隧道的几何信息和物理信息。该研究结果可为隧道火灾演化规律的快速预测提供参考。
[Objective] Tunnel fires pose a serious threat to life and property. The prediction of tunnel fires could reduce the risk and loss from thermal disasters. Computational fluid dynamics (CFD) provides a strong tool for quantitatively analyzing tunnel fires. However, CFD calculations are time-consuming, and reverse prediction from physical fields to key control parameters using the governing equation is impossible. To improve the prediction efficiency of tunnel fire information and solve the reverse prediction problem of key control parameters in tunnel fires, this paper proposes a deep learning model for fast bidirectional prediction between the entire physical fields and key control parameters of tunnel fires.[Methods] In this study, a deep learning model based on an encoder and a decoder is constructed, in which the encoder is used to construct the mapping from the physical fields to the key control parameters, and the decoder is used to construct the mapping from the key control parameters to the physical fields. In the model training process, the input of the encoder and the output of the decoder are required to be as close as possible, and the output of the encoder and the input of the decoder are also required to be as close as possible. The mathematical differences between them are therefore defined as the loss function. In this way, the encoder and the decoder form a cyclic structure. Data processing approaches are proposed so that all physical fields have a unified format and all key control parameters have the same distribution. [Results] The proposed model is trained using a large high-resolution numerical database with different cases under various key control parameters. The data learning ability and prediction capacity of the deep learning model are evaluated. With the increase of the training epoch, the calculated temperature field and key control parameters increasingly agree with the true temperature field and key control parameters. After 100 training epochs, the loss function almost converges, and the proposed bidirectional prediction model with the constructed dataset achieves good training convergence. In addition, the physical fields and key control parameters can be reproduced on the training set. After the completion of model training, the prediction performance of the deep learning model is tested. The average temperature field of tunnel fires and the six key parameters of tunnel fires are accurately predicted, and the predictions encompass the geometric and physical information of the tunnel. [Conclusions] Overall, this article proposes a deep learning network model based on the characteristics of tunnel fires for predicting various physical fields and key control parameters of tunnel fires. This study can be applied for the rapid acquisition of the full physical fields of tunnel fires, which helps design ventilation systems in tunnels and risk evaluation. In addition, another application is to retrieve the key control parameters of tunnel fires, which helps to quickly obtain the key control parameters according to the recorded infrared temperature field in the postinvestigation of tunnel fires. The above application scenarios can provide theoretical bases and new ideas for the prevention and control of tunnel fires.
[1] CARVEL R. A review of tunnel fire research from Edinburgh[J]. Fire Safety Journal, 2019, 105:300-306.
[2] CASEY N. Fire incident data for Australian road tunnels[J]. Fire Safety Journal, 2020, 111:102909.
[3] 田向亮,钟茂华,刘畅,等.不同阻塞条件下的隧道火灾全尺寸试验研究[J].煤炭科学技术, 2021, 49(2):93-101. TIAN X L, ZHONG M H, LIU C, et al. Experimental study on safety full-scale of tunnel fire disaster under different blocking conditions[J]. Coal Science and Technology, 2021, 49(2):93-101.(in Chinese)
[4] 卢平,丛北华,廖光煊,等.纵向通风水平隧道火灾烟气流动特性研究[J].中国工程科学, 2004, 6(10):59-64. LU P, CONG B H, LIAO G X, et al. Study of fire smoke flow characteristics of horizontal tunnel using longitudinal ventilation[J]. Strategic Study of CAE, 2004, 6(10):59-64.(in Chinese)
[5] 陈长坤,王玮玉,康恒,等.不同火源面积下隧道火灾温度场试验与数值模拟分析[J].中国公路学报, 2018, 31(6):235-243. CHEN C K, WANG W Y, KANG H, et al. Experimental and numerical simulation analysis of temperature field of tunnel fire with different fire source areas[J]. China Journal of Highway and Transport, 2018, 31(6):235-243.(in Chinese)
[6] 安伟光,广大庆,陈凡宝.油罐车行驶速度对隧道火灾温度及烟气蔓延的影响[J].消防科学与技术, 2023, 42(6):742-746. AN W G, GUANG D Q, CHEN F B. The effect of tank truck speed on the temperature distribution and smoke spread of tunnel fire[J]. Fire Science and Technology, 2023, 42(6):742-746.(in Chinese)
[7] HONG Y, KANG J H, FU C J. Rapid prediction of mine tunnel fire smoke movement with machine learning and supercomputing techniques[J]. Fire Safety Journal, 2022, 127:103492.
[8] 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.
[9] HU P, PENG X Y, TANG F. Prediction of maximum ceiling temperature of rectangular fire against wall in longitudinally ventilation tunnels:Experimental analysis and machine learning modeling[J]. Tunnelling and Underground Space Technology, 2023, 140:105275.
[10] MCGRATTAN K B, BAUM H R, REHM R G, et al. Fire dynamics simulator:Technical reference guide:NISTIR 6467[R]. Gaithersburg, USA:National Institute of Standards and Technology, 2000.
[11] HURLEY M J, GOTTUK D, HALL J R, et al. SFPE handbook of fire protection engineering[M]. 5th ed. New York, USA:Springer, 2016.
[12] MCGRATTAN K B, MCDERMOTT R, WEINSCHENK C, et al. Fire dynamics simulator technical reference guide volume 1:Mathematical model[R]. Gaithersburg, USA:National Institute of Standards and Technology, 2013.
[13] AGHDAM H H, HERAVI E J. Guide to convolutional[JP1] neural networks[M]. New York, USA:Springer, 2017:51.
[14] ZHANG K, SUN M, HAN T X, et al. Residual networks of residual networks:Multilevel residual networks[J]. IEEE Transactions on Circuits and Systems for Video Technology, 2018, 28(6):1303-1314.
[15] YAROTSKY D. Error bounds for approximations with deep ReLU networks[J]. Neural Networks, 2017, 94:103-114.
[16] YAZAN E, TALU M F. Comparison of the stochastic gradient descent based optimization techniques[C]//Proceedings of 2017 International Artificial Intelligence and Data Processing Symposium. Malatya, Turkey, 2017:1-5.
[17] SARANYA C, MANIKANDAN G. A study on normalization techniques for privacy preserving data mining[J]. International Journal of Engineering and Technology, 2013, 5(3):2701-2704.
[18] SVSSTRUNK S, BUCKLEY R, SWEN S. Standard RGB color spaces[C]//Proceedings of the 7th Color Imaging Conference. Scottsdale, USA, 1999:127-134.
[19] HOCHREITER S. The vanishing gradient problem during learning recurrent neural nets and problem solutions[J]. International Journal of Uncertainty, Fuzziness and Knowledge-Based Systems, 1998, 6(2):107-116.
[20] SIRIGNANO W A. Driving mechanisms for combustion instability[J]. Combustion Science and Technology, 2015, 187(1-2):162-205.
