Subway Tunnel Fire Research Articles

A deep learning–based surrogate model for spatial-temporal temperature field prediction in subway tunnel fires via CFD simulation

The smoke temperature and toxic gas generated by subway tunnel fires threaten trapped personnel and hinder rescue work. In this study, with a view to ensuring that smoke movement is acquired quickly and contributing to the future development of subway tunnel fires to guide emergency rescue, a deep learning surrogate model based on a transposed convolution neural network (TCNN) was proposed to predict the spatial-temporal temperature field. A numerical database of subway tunnel fires was constructed based on the computational fluid dynamics (CFD) method. Considering different fire locations, heat release rates (HRRs), and ventilation velocities, 135 scenarios were conducted to collect fire data. An adaptive up-projection unit (AUPU) and axial attention fusion block were proposed to improve the TCNN. The results showed that, with the inputs of fire location, HRR, and ventilation velocity, the surrogate model can quickly identify given features and reproduce the spatial-temporal temperature field with 94 % accuracy within 0.034 s. The proposed deep learning surrogate model can aid emergency control and rescue efforts in the context of subway tunnel fires.

Temperature prediction based on long short‐term memory convolutional neural network Bragg grating sensing

AbstractTo address the constraints associated with conventional fitting techniques for temperature demodulation in the context of subway tunnel fires, a new method of demodulation grating sensing spectrum using long short‐term memory convolutional neural network (LSTM‐CNN) is proposed in this paper. Build the monitoring platform based on LSTM‐CNN ultra‐weak fiber grating temperature measurement system, predict its sensing signals by LSTM‐CNN algorithm, select 18000 spectra as sample data for training, use AdamW stochastic optimization algorithm for training, and carry out the temperature calibration and demodulation error analysis of the Fiber Bragg Grating within the temperature range of 25–75°C. Compared with GRU algorithm, LSTM algorithm and traditional maximum peak method, the algorithm of this paper is good and can effectively improve the measurement accuracy, the experimental results show that: the demodulation accuracy of temperature wavelength prediction in this paper can be up to 99.27%, and the root mean square deviation is 0.08528°C, through the experiments, it is verified that the method proposed in this paper has a certain reference and support in terms of theories and technology significance. It is suitable for the identification and monitoring of fire hazards in underground tunnels, and also has application value in the signal processing of grating array sensing demodulation system.

Investigation of the fire hazard of underground space fire scenarios in urban metro tunnels under natural ventilation: Analysis of the impact of tunnel slope on smoke back-layering length

The smoke back-layering length is a crucial parameter for evacuating people in both road and subway tunnel fires. This study investigates the fire hazard induced by carriage fire in inclined metro tunnels under natural ventilation. The parameter ‘transition slope’ is defined to measure the smoke flow from the carriage head in the upstream direction to the tunnel or not due to the stack effect of the tunnel slope. The aim of this paper is to analyse the effects of changes in cross-section, downstream length, tunnel slope, and carriage side-door coupling on smoke behaviour characteristics by experiment and simulation methods. A piecewise function expression between dimensionless smoke back-layering length, downstream length, and tunnel slope for carriage fires in an inclined tunnel under natural ventilation is proposed by theoretical analysis. At the same time, a 1:15 scale model experiment was conducted to initially analyse the characteristics of smoke movement. Following this, full-scale numerical simulations were employed to complement the model experiment and quantify the principles governing smoke movement. The experimental results show that the tunnel slope has a significant effect on the smoke back-layering length. In contrast, the influence of the heat release rate was found to be relatively minor. In addition, simulation results show that the tunnel slope has no significant effect on the smoke back-layering length when the fire location is approximately 20 m from the train head, and the tunnel slope is in the range of 2.29° ∼ 3.43° (4% ∼ 6%). For small tunnel slopes, smoke spreads in the tunnel, and the smoke back-layering length produced by the virtual fire source shows a different law from the previous study model. Finally, the correlation coefficient of the piecewise function in theoretical analysis is fitted by combining the experimental and numerical simulation results. Practical application This study provides valuable insights into the practical implications of controlling and mitigating the impact of fires in inclined metro tunnels. By understanding the critical role of tunnel slope and providing a quantitative tool for smoke spread law assessment, this study contributes to the enhancement of safety measures and the protection of lives in tunnel environments during fire incidents.

Temperature prediction of Bragg grating sensing based on a one-dimensional convolutional neural network.

In this paper, we propose a new (to us) way of demodulating the grating sensing spectrum using a one-dimensional convolutional neural network (1DCNN) to overcome the limitation of the traditional fitting method of temperature demodulation for subway tunnel fires. This method constructs a one-dimensional convolutional neural network model and combines it with the experimental device of a fiber Bragg grating (FBG) temperature measurement. One thousand eight hundred spectra of experimental data are selected as sample data for training. Adam's random optimization algorithm is used in training to predict the temperature of multiple periods, with an accuracy of 99.95% and a root-mean-square deviation (RMSE) of 0.0832°C. The experiment shows that the algorithm in this paper is better than the GRU and LSTM algorithms, as traditional maximum peak methods, and can effectively improve the measurement accuracy. This article aims to provide a high-speed demodulation solution for FBG-based sensing systems to meet the practical needs of large-scale real-time monitoring.

Estimating longitudinal maximum smoke temperature attenuation due to large-angle bifurcation in UTLT-like tunnels under longitudinal ventilation condition

Extensive studies have focused on single-straight tunnels or subway bifurcation tunnel fire. However, the effect of large angle bifurcation on longitudinal maximum temperature rise attenuation in Urban Traffic Link Tunnel（UTLT）needs to be further studied. In this study, based on the theoretical analysis, a double exponential prediction model for longitudinal temperature rise attenuation in bifurcation tunnels is established. To obtain the correlation coefficients, a series of full-scale numerical fire simulation experiments were carried out in an UTLT-like tunnels using Fire Dynamics Simulator. The results revealed that the bifurcation structure of the large-angle bifurcation angle tunnels impeded the longitudinal movement of the fire smoke and hot smoke diversion. Without longitudinal ventilation conditions, the longitudinal decay of the temperature rise in each region can be expressed as the sum of double exponential expressions with different coefficients. The law governing the effect of the bifurcation angle on the longitudinal decay of the temperature rise is the quadratic equation of the angle sin α. A prediction model for the longitudinal decay of the temperature rise in the tunnel downstream of the fire source under longitudinal ventilation is proposed. The findings of the numerical simulation corresponded to those of the model prediction.

A fast wavelength detection method based on OTDR and 1-DDCNN in series overlapping spectra

To improve the rapidity and reliability of the subway tunnel fire monitoring system and realize the monitoring of small fire sources, the method of regional subsection demodulation is used to group several adjacent nearly identical ultra-weak fiber Bragg gratings (UWFBGs) sensors into a group, and the 1-D dilated convolutional neural network (1-DDCNN) is used to demodulate the same group of highly overlapping sensing signals. The well-trained 1-DDCNN model can achieve extremely low signal demodulation error. The experiment shows that the UWFBG demodulation scheme proposed in this paper improves the detection accuracy of the subway tunnel fire monitoring system and shortens the detection time. The root-mean-square error (RMSE) of four highly overlapping peak wavelengths of sensing signals is less than 1.5 pm, and the average demodulation time is less than 30 ms.

Study of Blockage Effects of Metro Train on Critical Velocity in Sloping Subway Tunnel Fires with Longitudinal Ventilation

Critical velocity is very important for smoke control in longitudinally ventilated subway tunnel fires. Numerical investigations are conducted in this paper to study the impacts of metro train blockages on critical velocity in sloping subway tunnel fires by using fire dynamics simulator (FDS) tunnel models validated with the field-experiment data. Moreover, a global model of critical velocity is presented for the blocked zone of a metro train in subway tunnel fires including influencing factors of the blockage ratio and tunnel slope. The results show that the reduction ratio of critical velocity in the blocked zone is less than the metro-train blockage ratio. The correction factor between the critical velocity reduction ratio and metro-train blockage ratio is 0.545. The aerodynamic shadow zone downstream of a subway train blockage has important impacts on the critical velocity. The critical velocity in the unblocked zone of a metro train is higher than that in the blocked zone of a metro train blockage. The reason is that smoke flow is hindered by the metro train blockage in subway tunnel fires. With an increase in the blockage–fire source distance, the critical velocity first decreases and then tends to be constant. The global model presented can accurately predict the critical velocity in a sloping subway tunnel with a train blockage. The results may provide beneficial suggestions for designing ventilation systems for subway tunnels.

Study of Obstruction Effects of Metro Train on Critical Velocity in Sloping Subway Tunnel Fires with Longitudinal Ventilation

Study of Obstruction Effects of Metro Train on Critical Velocity in Sloping Subway Tunnel Fires with Longitudinal Ventilation

Numerical Simulation Study of Effect of Tunnel Slope on Smoke Propagation

Tunnel slope is an important factor affecting smoke spreading for a stopped subway train on fire. In this paper, a multi-scale method where one-dimensional (1D) network is coupled with three-dimensional (3D) field simulation is used to calculate smoke and temperature distributions for a fire occurred in the middle part of the subway train in tunnel. The tunnel slopes of 0‰, 12‰, 21‰ and 30‰ are considered. By comparing the temperature distribution in a condition with natural ventilation and with 2 m/s longitudinal ventilation, we comprehensively analyzed temperature, smoke spread characteristics. Results show that tunnel slope has a great influence on temperature and smoke profile when slope is higher than 12‰. The temperature at 1.8m would exceed the safe limit of 60°C at 60 s after fire ignition. The larger the slope, the higher temperature would be in the train carriage. For a train fire scenario with natural ventilation, within initial 120 s period the train compartment could keep a safe evacuation condition. This research would provide reference for making right evacuation plan of subway tunnel fire.

Real-Time Prediction of Smoke Spread Affected by Multiple Factors in Subway Tunnel Using CAERES-DNN Model

High-accuracy and real-time smoke spread models are vitally important for firefighting in subway tunnels. This paper proposes a data-driven model, named CAERES-DNN, for predicting smoke spread in subway tunnel fire. CAERES-DNN constitutes of CAERES (convolutional autoencoder with residual blocks) architecture and DNN (deep neural network) architecture. CAERES has 26 convolutional layers, 4 full link layers and 2 reshape layers in total. Specially, 6 residual blocks introduced in the CAERES to avoid gradient vanishing. DNN, a six-full-link-layer neural network, is built to regress data. Both in CAERES and DNN, binary cross entropy loss with L2-norm and stochastic gradient descent optimizer with momentum are used to update and calculate the model parameters. Therefore, CAERES-DNN performs better than conventional convolutional neural network based on the dataset (500, 100 and 48 in training, validation and testing set) from 54 FDS simulations in the fired subway tunnel. The 95.3% soot visibility values predicted by CAERES-DNN is within ± 10% of FDS prediction. In the 200 m tunnel, the prediction error of spread distance is always within ± 5 m. Meanwhile the prediction time is in seconds by the pre-trained model, which is 105 times faster than FDS. Due to its accuracy and speed, CAERES-DNN can help in firefighting during a confined and long subway tunnel in real-time.

Study on the Smoke Spread Law of a Subway Tunnel Fire under the Synergistic Action of Shaft Smoke Exhaust and Longitudinal Ventilation

Urban rail transit improves the urban spatial structure, solves the problem of urban traffic congestion, so its safety problem can not be ignored. Fire is one of the major factors threatening the safety of urban rail transit. Taking the tunnel between wengjiao road and Maqing road of Xiamen Rail Transit Line 2 as an example, the paper uses FDS software to simulate the mechanical smoke exhaust of the shaft under different wind speed conditions of the main tunnel, and deeply analyzes the smoke and temperature distribution under the fire conditions of the tunnel with the shaft. The results show that when the longitudinal wind speed v=1.7m/s and the shaft uses mechanical exhaust, there is no smoke upstream of the fire source, all the smoke is discharged through the shaft, and the height of the flue gas layer is significantly smaller than the thickness of the flue gas layer without longitudinal ventilation. The combination of longitudinal wind and mechanical smoke exhaust is more conducive to the safe evacuation of people.

A Study on Ceiling Temperature Distribution and Critical Exhaust Volumetric Flow Rate in a Long-Distance Subway Tunnel Fire with a Two-Point Extraction Ventilation System

Smoke control is a crucial issue in a long-distance subway tunnel fire, and a two-point extraction ventilation system is an effective way to solve this problem, due to the characteristics of controlling the smoke in a limited area and removing high-temperature and toxic smoke in time. In this study, the ceiling temperature distribution and the critical exhaust volumetric flow rate to control the smoke in the zone between two extraction vents were investigated in a long-distance subway tunnel fire with a two-point extraction ventilation system. Experiments were carried out in a 1/20 reduced-scale tunnel model based on Froude modeling. Factors, including the heat release rate (HRR), the extraction vent length, the internal distance between two extraction vents and exhaust volumetric flow rate, were studied. Smoke temperature below the ceiling, exhaust volumetric flow rate and smoke spreading configurations were measured. The ceiling temperature distribution was analyzed. Meanwhile, an empirical equation was developed to predict the critical exhaust volumetric flow rate based on the one-dimensional theory, experimental phenomenon and the analysis of forces acting at the smoke underneath the extraction vent. The coefficients in the empirical equation were determined by experimental data. Compared with the experimental results, the developed empirical equation can predict the critical exhaust volumetric flow rate well. Research outcomes in this study will be beneficial to the design and application of two-point extraction ventilation system for a long-distance subway tunnel fire.

Critical roof opening longitudinal length for complete smoke exhaustion in subway tunnel fires

A series of experiments were conducted to investigate the smoke exhaustion in a subway tunnel under natural ventilation with a wide roof opening (its width is equal to tunnel width) based on a 1/10 reduced-scale tunnel model. The effect of the roof opening longitudinal length on the smoke exhaustion was addressed both theoretically and experimentally. The results indicated that there existed a critical opening longitudinal length which allows a complete smoke exhaustion under fire conditions. The critical opening longitudinal length was related to the tunnel aspect ratio and the distance between fire source and roof opening. A theoretical model was also developed to predict the critical opening longitudinal length for subway tunnel. The research outcomes of this study provide a technical guide for the future subway tunnel design.

Effects of ambient pressure on smoke back-layering in subway tunnel fires

This paper investigates the effects of ambient pressure on smoke back-layering in subway tunnel fires with and without train blockage. A series of numerical simulations were conducted in a 1/4 small-scale tunnel with different heat release rates (40–160 kW), longitudinal ventilation velocities (0.2–0.8 m/s) and ambient pressures (60–100 kPa). The smoke back-layering lengths under different conditions are analyzed, and the results show that under the same heat release rate and ventilation velocity, the back-layering length increases with decreasing ambient pressure due to the weak inertial force of longitudinal airflow led by the low air density. The Li’s and Zhang’s models, which can well predict the smoke back-layering length under regular pressure, are modified for the reduced pressure. The constant increment of back-layering length between adjacent ambient pressures, which mainly depends on the heat release rate, is used to correct Li’s model. The smoke back-layering length under the low ambient pressure can be predicted by this modified model without train blockage. With train blockage considered, new models are developed by introducing both equivalent and virtual fire sources for predicting the smoke back-layering length under the low ambient pressure, which is shown to well reproduce the simulation results.

Influence of mechanical smoke exhaust on smoke spread in underground tunnel

The model of a tunnel in Beijing is established, and the smoke reflux, the smoke reflux length and the influence of longitudinal wind speed on temperature distribution of the tunnel vault are studied using numerical simulation method. The length of the tunnel is 150 m, the cross section is circular, the height of the tunnel is 6.3 m and the width is 6 m. The simulation results show that the smoke reflux length of the subway tunnel fire decreases with the increase of the longitudinal positive pressure wind speed, and the reflux length decreases linearly with the increase of wind speed when the reflux length is more than 10 m. The gradient of the reflux length decreases with the increase of the wind speed when the distance from the fire source decreases. The temperature curve of tunnel vault is consistent. Under the same fire source heat release rate, the flue gas temperature decreases with the increase of ventilation wind speed, and the distance between the highest temperature of the ceiling and the fire source decreased with the increase of the longitudinal ventilation wind speed.

The Zhengzhou 27 Subway Station Under Fire Smoke Control Multiple Ventilation Mechanism Perfect Research

As an important facility of modern urban traffic, Metro has the characteristics of large passenger volume, high speed and accurate time. Therefore, in recent years, with the rapid development of urban rail transit construction in our country, the subway line length and passenger traffic volume have been increasing. But along with the development of subway traffic, fire accidents happen frequently. The subway with ventilation conditions, lighting conditions, characteristics of the internal environment is complex, fire safety evacuation and rescue work more difficult, so the fire safety design features of subway tunnel fire and tunnel has become a new research topic.

An experimental investigation on blockage effect of metro train on the smoke back-layering in subway tunnel fires

This paper investigates the influence of metro train on the smoke back-layering in the subway tunnel fires. Due to the reduction of cross-sectional area of subway tunnel, the ventilation velocity will increase when the ventilated air flows across the metro train. Both the inertia force of ventilated air and the heat loss of back-flowed smoke gas will be enhanced by the ventilation with higher velocity. Then the smoke back-layering will be different from that in the subway tunnel without metro train. A series of small-scale experiments were conducted to study the smoke back-layering in the tunnel with metro train inside and the results showed that the smoke back-layering length was greatly influenced by the metro train. Compared to that in the tunnel without metro train, the back-layering length in the current study was much reduced due to the enhanced inertia force and heat loss. Based on the smoke gas temperature distribution and fire dynamics theory, a new physical model including the factor of metro train length was developed to predict the length of smoke back-layering in the tunnel.

Smoke Control in Subway Tunnel Fires

This paper concerns the smoke control modes and the critical ventilation velocity when the subway tunnel on fires. The standard for the smoke control mode is making sure the smoke exhausting in the shortest way. The critical ventilation velocity means it is just sufficient to prevent the smoke spreading upstream. The critical velocity in different heat release rates obtained though theoretical analysis and computer simulation. In the end, a simple formula to calculate the critical velocity can be fitting out.

Study of a proposed tunnel evacuation passageway formed by opposite-double air curtain ventilation

Subway tunnel fires often result in catastrophes and heavy casualties. To help people evacuate from a tunnel fire, an opposite-double air curtain ventilation assisted tunnel evacuation system (OTES) is introduced. It can be used to create a safe evacuation passageway that is free of smoke through out the length of the tunnel. The performance of the OTES is compared with that of the traditional ventilation systems: longitudinal ventilation and natural ventilation. The effect of the heat release rate, fire source location, and fire detection time are also discussed.This study also shows that compared with natural ventilation and longitudinal ventilation, the carbon monoxide, CO, concentration with OTES is significantly lower. Given the same HRR, the CO concentration values with OTES are only 0.58–2.41% of natural ventilation or 0.52–3.12% of longitudinal ventilation at the back end of the tunnel. An obviously clear evacuation passageway is created by OTES. The effects of changes in fire source location and fire detection time on tunnel ventilation are tested. Also the effect of these changes on the formation of an evacuation passageway by OTES is reported.

A model experimental study on backdraught in tunnel fires

Backdraught is a special fire phenomenon in a limited-ventilation space. Research on the occurrence of backdraught in compartment fires has been extensive, but backdraught in subway tunnel fires, which occur in underground spaces with a large slenderness ratio, has received insufficient attention. In the present study, 31 cases, divided into two groups under conditions of natural ventilation and mechanical ventilation, were examined using a 1/8 reduced-scale model tunnel to investigate the critical conditions and characteristics of backdraught occurrence, as well as the differences between tunnel backdraught and compartment backdraught. The existing critical values of the mass fraction for different ventilation conditions are also discussed. The results indicated that the key parameter determining the occurrence of backdraught in subway tunnel fires is the mass fraction of the volatilized unburned fuel in the tunnel. The critical values of the mass fraction in natural ventilation and mechanical ventilation were 8.78% and 11.71%, respectively, with a humidity of 15% in fresh air. With natural ventilation, backdraught occurrence in a single tube tunnel configuration was similar to that in a compartment, and the ignition delay depended on the velocity of the gravity current; with mechanical ventilation, backdraught occurrence in a twin-tube tunnel configuration was different from that in a compartment, and its ignition delay was determined mainly by the ignition source delay. In addition, as the velocity of air flowing into the tunnel increased, so did the intensity of the backdraught. The experimental data were qualitatively validated and analyzed based on the flammability diagram of the fuel, and these results were similar to those obtained in previous experiments. Also, the humidity of the fresh air flowing into the tunnel affected the occurrence of backdraught; it would not occur under conditions of higher humidity even if the mass fraction mentioned above was higher. Furthermore, it may be economical and feasible to install a humidification device in the tunnel ventilation system to humidify the fresh air flowing into the tunnel to inhibit backdraught.