Skip to main content

A real-time forecast of tunnel fire based on numerical database and artificial intelligence


The extreme temperature induced by fire and hot toxic smokes in tunnels threaten the trapped personnel and firefighters. To alleviate the potential casualties, fast while reasonable decisions should be made for rescuing, based on the timely prediction of fire development in tunnels. This paper targets to achieve a real-time prediction (within 1 s) of the spatial-temporal temperature distribution inside the numerical tunnel model by using artificial intelligence (AI) methods. A CFD database of 100 simulated tunnel fire scenarios under various fire location, fire size, and ventilation condition is established. The proposed AI model combines a Long Short-term Memory (LSTM) model and a Transpose Convolution Neural Network (TCNN). The real-time ceiling temperature profile and thousands of temperature-field images are used as the training input and output. Results show that the predicted temperature field 60 s in advance achieves a high accuracy of around 97%. Also, the AI model can quickly identify the critical temperature field for safe evacuation (i.e., a critical event) and guide emergency responses and firefighting activities. This study demonstrates the promising prospects of AI-based fire forecasts and smart firefighting in tunnel spaces.

This is a preview of subscription content, access via your institution.



artificial intelligence


artificial neural networks


computational fluid dynamics


convolutional neural network


critical ventilation velocity


fire dynamics simulator


heat release rate (MW)


long short-term memory


machine learning


recurrent neural network


support vector machine


transpose convolutional neural network


  • Adamowski J, Karapataki C (2010). Comparison of multivariate regression and artificial neural networks for peak urban water-demand forecasting: Evaluation of different ANN learning algorithms. Journal of Hydrologic Engineering, 15: 729–743.

    Article  Google Scholar 

  • Akhloufi MA, Booto Tokime R, Elassady H (2018). Wildland fires detection and segmentation using deep learning. In: Proceedings of SPIE 10649, Pattern Recognition and Tracking XXIX, 06490B.

  • Aksoy S, Haralick RM (2001). Feature normalization and likelihood-based similarity measures for image retrieval. Pattern Recognition Letters, 22: 563–582.

    Article  MATH  Google Scholar 

  • Alzubaidi L, Al-Shamma O, Fadhel MA, et al. (2020). Classification of red blood cells in sickle cell anemia using deep convolutional neural network. In: Abraham A, Cherukuri A, Melin P, Gandhi N (eds), Intelligent Systems Design and Applications (ISDA 2018). Advances in Intelligent Systems and Computing, vol. 940. Cham, Switzerland: Springer.

    Google Scholar 

  • Arora R, Basu A, Mianjy P, et al. (2018). Understanding deep neural networks with rectified linear units. In: Proceedings of the 6th International Conference on Learning Representations (ICLR 2018).

  • Babrauskas V, Peacock RD (1992). Heat release rate: The single most important variable in fire hazard. Fire Safety Journal, 18: 255–272.

    Article  Google Scholar 

  • Beard AN (2009). Fire safety in tunnels. Fire Safety Journal, 44: 276–278.

    Article  Google Scholar 

  • Beard A, Carvel R (2012). The Handbook of Tunnel Fire Safety, 2nd edn. London: ICE Publishing.

    Google Scholar 

  • Bengio Y, Simard P, Frasconi P (1994). Learning long-term dependencies with gradient descent is difficult. IEEE Transactions on Neural Networks, 5: 157–166.

    Article  Google Scholar 

  • Cao Y, Yang F, Tang Q, Lu X (2019). An attention enhanced bidirectional LSTM for early forest fire smoke recognition. IEEE Access, 7: 154732–154742.

    Article  Google Scholar 

  • Carvel RO, Beard AN, Jowitt PW (2005). Fire spread between vehicles in tunnels: Effects of tunnel size, longitudinal ventilation and vehicle spacing. Fire Technology, 41: 271–304.

    Article  Google Scholar 

  • Carvel R (2019). A review of tunnel fire research from Edinburgh. Fire Safety Journal, 105: 300–306.

    Article  Google Scholar 

  • Casey N (2020). Fire incident data for Australian road tunnels. Fire Safety Journal, 111: 102909.

    Article  Google Scholar 

  • Chehreh Chelgani S, Shahbazi B, Hadavandi E (2018). Support vector regression modeling of coal flotation based on variable importance measurements by mutual information method. Measurement, 114: 102–108.

    Article  Google Scholar 

  • Chen MY (2011). Predicting corporate financial distress based on integration of decision tree classification and logistic regression. Expert Systems with Applications, 38: 11261–11272.

    Article  Google Scholar 

  • Cohen PR, Greenberg ML, Hart DM, et al. (1989). Trial by fire: Understanding the design requirements for agents in complex environments. AI Magazine, 10(3): 32–48.

    Google Scholar 

  • Danziger NH, Kennedy WD (1982). Longitudinal ventilation analysis for the Glenwood Canyon tunnels. In: Proceedings of the 4th International Symposium on the Aerodynamics and Ventilation of Vehicle Tunnels.

  • Dexters A, Leisted RR, Van Coile R, et al. (2020). Testing for knowledge: Application of machine learning techniques for prediction of flashover in a 1/5 scale ISO 13784-1 enclosure. Fire and Materials,

  • Dindarloo SR, Siami-Irdemoosa E (2015). Maximum surface settlement based classification of shallow tunnels in soft ground. Tunnelling and Underground Space Technology, 49: 320–327.

    Article  Google Scholar 

  • Dubey V, Kumar P, Chauhan N (2019). Forest Fire Detection System Using IoT and Artificial Neural Network. Singapore: Springer.

    Book  Google Scholar 

  • Dumoulin V, Visin F (2016). A guide to convolution arithmetic for deep learning. arXiv preprint arXiv:160307285.

  • Forney GP (2010). Smokeview (Version 5). A Tool for Visualizing Fire Dynamics Simulation Data Volume I: User’s Guide. Gaithersburg, MD, USA: National Institute of Standards and Technology.

    Google Scholar 

  • Ghoreishi M, Bagchi A, Sultan M (2013). Review of the punching shear behavior of concrete flat slabs in ambient and elevated temperature. Journal of Structural Fire Engineering, 4: 259–280.

    Article  Google Scholar 

  • Glorot X, Bengio Y (2010). Understanding the difficulty of training deep feedforward neural networks. In: Proceedings of the 13th International Conference on Artificial Intelligence and Statistics Chia Laguna Resort, Sardinia, Italy.

  • Gong L, Jiang L, Li S, et al. (2016). Theoretical and experimental study on longitudinal smoke temperature distribution in tunnel fires. International Journal of Thermal Sciences, 102: 319–328.

    Article  Google Scholar 

  • Govil K, Welch ML, Ball JT, et al. (2020). Preliminary results from a wildfire detection system using deep learning on remote camera images. Remote Sensing, 12: 166.

    Article  Google Scholar 

  • Haack A (2002). Current safety issues in traffic tunnels. Tunnelling and Underground Space Technology, 17: 117–127.

    Article  Google Scholar 

  • Han L, Potter S, Beckett G, et al. (2010). FireGrid: An e-infrastructure for next-generation emergency response support. Journal of Parallel and Distributed Computing, 70: 1128–1141.

    Article  Google Scholar 

  • Hayou S, Doucet A, Rousseau J (2019). On the impact of the activation function on deep neural networks training. In: Proceedings of the 36th International Conference on Machine Learning (ICML 2019).

  • He K, Zhang X, Ren S, et al. (2016). Deep residual learning for image recognition. In: Proceedings of 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, NV, USA.

  • Hochreiter S, Schmidhuber J (1997). Long short-term memory. Neural Computation, 9: 1735–1780.

    Article  Google Scholar 

  • Hodges JL (2018). Predicting large domain multi-physics fire behavior using artificial neural networks. PhD Thesis, Virginia Polytechnic Institute and State University, USA.

    Google Scholar 

  • Hodges JL, Lattimer BY (2019). Wildland fire spread modeling using convolutional neural networks. Fire Technology, 55: 2115–2142.

    Article  Google Scholar 

  • Hodges JL, Lattimer BY, Luxbacher KD (2019). Compartment fire predictions using transpose convolutional neural networks. Fire Safety Journal, 108: 102854.

    Article  Google Scholar 

  • Hu LH, Tang F, Yang D, et al. (2010). Longitudinal distributions of CO concentration and difference with temperature field in a tunnel fire smoke flow. International Journal of Heat and Mass Transfer, 53: 2844–2855.

    Article  MATH  Google Scholar 

  • Hurley MJ, Gottuk D, Hall JR Jr, et al. (2016). SFPE Handbook of Fire Protection Engineering, 5th edn. New York: Springer.

    Book  Google Scholar 

  • Ingason H, Lönnermark A (2012). Heat release rates in tunnel fires: A summary. In: Beard A, Carvel R (eds), The Handbook of Tunnel Fire Safety, 2nd edn. London: ICE Publishing.

    Google Scholar 

  • Ingason H, Li YZ, Lönnermakr A (2015). Tunnel Fire Dynamics. New York: Springer.

    Book  Google Scholar 

  • Ji J, Wan H, Li K, et al. (2015). A numerical study on upstream maximum temperature in inclined urban road tunnel fires. International Journal of Heat and Mass Transfer, 88: 516–526.

    Article  Google Scholar 

  • Ji J, Tong Q, Wang L, et al. (2018). Application of the EnKF method for real-time forecasting of smoke movement during tunnel fires. Advances in Engineering Software, 115: 398–412.

    Article  Google Scholar 

  • Kavzoglu T, Mather PM (2003). The use of backpropagating artificial neural networks in land cover classification. International Journal of Remote Sensing, 24: 4907–4938.

    Article  Google Scholar 

  • Koekkoek EJW, Booltink H (1999). Neural network models to predict soil water retention. European Journal of Soil Science, 50: 489–495.

    Article  Google Scholar 

  • Komer B, Bergstra J, Eliasmith C (2014). Hyperopt-Sklearn: Automatic Hyperparameter Configuration for Scikit-Learn. In: Proceedings of the 13th Python in Science Conference, Austin, TX, USA.

  • Lee D, Lim M, Park H, et al. (2017). Long short-term memory recurrent neural network-based acoustic model using connectionist temporal classification on a large-scale training corpus. China Communications, 14: 23–31.

    Article  Google Scholar 

  • Li YZ, Ingason H (2018). Overview of research on fire safety in underground road and railway tunnels. Tunnelling and Underground Space Technology, 81: 568–589.

    Article  Google Scholar 

  • Li J, Liu J (2020). Science mapping of tunnel fires: A scientometric analysis-based study. Fire Technology, 56: 2111–2135.

    Article  Google Scholar 

  • Mcgrattan K, Mcdermott R (2015). Fire Dynamics Simulator User’s Guide (FDS Version 6.3.0).

  • Morales G, Huamán SG, Telles J (2018). Cloud detection in highresolution multispectral satellite imagery using deep learning. In: Proceedings of the 27th International Conference on Artificial Neural Networks and Machine Learning (ICANN 2018). Cham, Switzerland: Springer.

    Google Scholar 

  • Nævestad TO, Meyer S (2014). A survey of vehicle fires in Norwegian road tunnels 2008–2011. Tunnelling and Underground Space Technology, 41: 104–112.

    Article  Google Scholar 

  • Ngoc Thach N, Bao-Toan Ngo D, Xuan-Canh P, et al. (2018). Spatial pattern assessment of tropical forest fire danger at Thuan Chau area (Vietnam) using GIS-based advanced machine learning algorithms: A comparative study. Ecological Informatics, 46: 74–85.

    Article  Google Scholar 

  • Oka Y, Atkinson GT (1995). Control of smoke flow in tunnel fires. Fire Safety Journal, 25: 305–322.

    Article  Google Scholar 

  • Pei Y, Gan F (2009). Research on data fusion system of fire detection based on neural-network. In: Proceedings of 2009 Pacific-Asia Conference on Circuits, Communications and Systems (PACCS), Chengdu, China.

  • Ren R, Zhou H, Hu Z, et al. (2019). Statistical analysis of fire accidents in Chinese highway tunnels 2000–2016. Tunnelling and Underground Space Technology, 83: 452–460.

    Article  Google Scholar 

  • Salhi L, Silverston T, Yamazaki T, et al. (2019). Early detection system for gas leakage and fire in smart home using machine learning. In Proceedings of 2019 IEEE International Conference on Consumer Electronics (ICCE), Las Vegas, NV, USA.

  • Sayad YO, Mousannif H, Al Moatassime H (2019). Predictive modeling of wildfires: A new dataset and machine learning approach. Fire Safety Journal, 104: 130–146.

    Article  Google Scholar 

  • Sun M, Raju A, Tucker G, et al. (2016). Max-pooling loss training of long short-term memory networks for small-footprint keyword spotting. In: Proceedings of 2016 IEEE Workshop on Spoken Language Technology (SLT 2016), San Diego, CA, USA.

  • Sun P, Bisschop R, Niu H, et al. (2020). A review of battery fires in electric vehicles. Fire Technology, 56: 1361–1410.

    Article  Google Scholar 

  • Tetko IV, Livingstone DJ, Luik AI (1995). Neural network studies. 1. Comparison of overfitting and overtraining. Journal of Chemical Information and Modeling, 35: 826–833.

    Google Scholar 

  • De Vasconcelos MJP, Silva S, Tomé M, et al. (2001). Spatial prediction of fire ignition probabilities: Comparing logistic regression and neural networks. Photogrammetric Engineering and Remote Sensing, 67: 73–81.

    Google Scholar 

  • Weng Y-C, Ke M-T, Chen S-L (2009). Safety reliability method applied to evaluate smoke control system of fire accident in the tunnel of mass rapid transit system. In: Proceedings of the World Congress on Engineering 2009 Vol. II (WCE 2009), London, UK.

  • Wu Y, Liu Y, Li J, et al. (2013). Traffic sign detection based on convolutional neural networks. In: Proceedings of the International Joint Conference on Neural Networks (IJCNN), Dallas, TX, USA.

  • Wu X, Park Y, Li A, et al. (2021). Smart detection of fire source in tunnel based on the numerical database and artificial intelligence. Fire Technology, 57: 657–682.

    Article  Google Scholar 

  • Xue CJ (2010). The road tunnel fire detection of multi-parameters based on BP neural network. In: Proceedings of 2nd International Asia Conference on Informatics in Control, Automation and Robotics (CAR 2010).

  • Yao Y, Yang J, Huang C, et al. (2010). Fire monitoring system based on multi-sensor information fusion. In: Proceedings of the 2nd International Symposium on Information Engineering and Electronic Commerce (IEEC 2010), Ternopil, Ukraine.

  • Zhai C, Zhang S, Cao Z, et al. (2020). Learning-based prediction of wildfire spread with real-time rate of spread measurement. Combustion and Flame, 215: 333–341.

    Article  Google Scholar 

  • Zhang X, Wu X, Park Y, et al. (2021). Perspectives of big experimental database and artificial intelligence in tunnel fire research. Tunnelling and Underground Space Technology, 108: 103691.

    Article  Google Scholar 

Download references


This work is funded by the Hong Kong Research Grants Council Theme-based Research Scheme (T22-505/19-N) and the PolyU Emerging Frontier Area (EFA) Scheme of RISUD (P0013879).

Author information

Authors and Affiliations


Corresponding author

Correspondence to Xinyan Huang.

Electronic Supplementary Material

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Wu, X., Zhang, X., Huang, X. et al. A real-time forecast of tunnel fire based on numerical database and artificial intelligence. Build. Simul. 15, 511–524 (2022).

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI:


  • tunnel fires
  • smart firefighting
  • critical event
  • CFD
  • deep learning