Experimental investigation on the characteristics and propagation of fire inside subway train
Introduction
The research of fire in complex confined structures has gradually attracted the considerable attention of researchers for decades. For example, Du et al. (Du et al., 2015) used CFD to perform the simulation of the whole urban traffic link tunnel, and studied the design method of smoke control. Yang et al. (Yang et al., 2018) conducted brine water experiments in the inclined tunnel and analyzed the back-layering flow under this simulated fire. Ng Y W et al. (Ng et al., 2019) performed experiments in train car model to explore the flame color under poorly ventilated condition. As one of the complex confined structures, subway system exists widely around the world due to its great convenience for the people in daily life. If a fire breakout in the subway train with lateral openings, it is hazardous for the life of the commuter and the infrastructure. In fact, many major subway fire accidents happened in the history (Li and Ingason, 2018, Peng et al., 2020). Hence, it is important to study the characteristics of the fire inside the subway train, especially the flame length and temperature distribution under ceiling, which are considered to be the two most crucial and representative characteristics for fire detection and protection system.
Numerous research work have been carried out on the flame extension in the ceiling jet under various confinement conditions. For unconfined horizontal ceiling, You and Faeth (You and Faeth, 1979) were the first to experimentally measure the impinging flame lengths, and a formula for predicting the flame length under the ceiling was developed. Then, Zhang et al. (Zhang et al., 2014, Zhang et al., 2017) measured and correlated the impinging flame lengths beneath the horizontal and inclined ceiling for a free fire jet impingement. Further, the study (Zhang et al., 2019a) on the fire impingement scenario of the wall-attached fire were carried out, and the flame lengths in different directions were correlated based on the physical nature of such impinging flow structure. Moreover, Poreh and Garrad (Poreh and Garrad, 2000) studied the measurements of flame heights under different confined conditions. The effect of wall on the flame height was discussed and the proposed model overestimated the experimental value of the flame length induced by fires near the wall. Lattimer et al. (Lattimer et al., 2013) reviewed the flame lengths in various confined configurations and compared the dependence of the flame length on the HRR in different configurations. Recently, Gao et al. (Gao et al., 2019) conducted multiple experiments to characterize the ceiling jet under different flame-wall interactions. They found that the ceiling impingement flames under different confinement strengths followed different patterns, and obtained the correlations to predict the ceiling flame for different impinging flows. The above researches clearly show that different scenarios, especially different confinement conditions, can have an effect on the flame length under the ceiling. However, the flame length in a long - narrow confined structure with lateral openings has been rarely studied systematically.
In terms of smoke temperature distribution under the ceiling, many correlations under various conditions were formulated. Alpert (Alpert, 1975, Alpert, 2004) and Heskestad and Delichatsios (Gunnar et al., 1979) proposed the classical models of the temperature distribution underneath the large flat ceiling, and the dimensionless temperature could be represented as a function of the non-dimensional location . Oka et al. (Oka et al., 2010) carried out experiments in an inclined unconfined ceiling apparatus to study the effect of the inclination angel of the ceiling on the temperature distributions in different directions along the ceiling. Zhang et al. (Zhang et al., 2019b) performed experiments of wall-attached fire in a configuration consisting of wall and inclined ceiling, and the correlation for the temperature profile underneath the ceiling was developed for various ceiling inclination angles. In addition, there are also many studies on fire scenarios in long-narrow space. Zhao et al. (Zhao et al., 2019) performed a mass of experiments in tunnels with various scales to research the ceiling temperature distribution. To study the effect of the tunnel width, Ji et al. (Ji et al., 2016) carried out experiments in a model tunnel with variable aspect ratio. The models for predicting the temperature distributions involving the tunnel width and HRR were proposed. Zhao et al. (Zhao et al., 2017) established a reduced-scale corridor to analyze the effect of openings on longitudinal and vertical smoke temperature, and concluded that the longitudinal temperature under the ceiling in the far-field for different openings also submitted to the exponential decay. These studies perform an important role in the exploration of temperature distribution in confined space fires. However, studies focusing on fires in long -narrow structures with lateral openings are rarely reported, although the existence of lateral openings may have an effect on the propagation of smoke in the structure. In addition, the current temperature prediction models are generally one-dimensional relationships such as transverse or longitudinal, and less attention is paid to the prediction of two-dimensional temperature distribution.
Moreover, some researches on fires in confined structures with opening have also been carried out. Ji et al. (Ji et al., 2015) performed experiments in a compartment with two opposite openings to investigate the effect of wind velocity on compartment fires. For compartment fire under cross wind, Gao et al. (Gao et al., 2016a) performed experiments to focus on the spill plume and the new length scale was proposed to correlate the temperature of facade plume. Zhang et al. (Zhang et al., 2020) conducted fire experiments in compartment with circular openings, and discussed the internal temperature and the extension mechanism of the façade flame. The smoke movement in the above scenario is different from that in the subway train. Because these experimental structures are cubic compartments rather than long-narrow structures, and these studies are more focused on the characteristics of the spill plume rather than the fire characteristics inside the structure. Recently, Ingason (Ingason, 2007) conducted the fire experiments in a model railcar to study the fire development under different parameters. Li et al. (Li et al., 2014) performed the fire experiments in the train carriage models with various scales to analyze the fire development and concluded that the fire development in different scales was similar. Lönnermark et al. (Lönnermark et al., 2017) studied the impact of different ignition source-related variables on the fire development by conducting experiments in the train model. Although train fires have been qualitatively studied in many aspects, the fundamental knowledge of the quantitative research on train fire needs to be supplemented.
The above literature review shows that the state of the art on this subject provides inadequate knowledge about fire inside subway train, especially the important features that characterize the fire hazard, the flame length and temperature distribution under ceiling. Besides, most of the work was focused on the one-dimensional temperature distribution along the longitudinal or transverse centerlines, but very few studies have been conducted for overall ceiling temperature contour, even though they are the determining factors for the fire hazard. Therefore, in order to investigate the flame length and 2D ceiling temperature profile induced by fire inside the subway train, three series of experiments were carried out (fire source located in the open space or inside the train). Correlations were developed to quantitatively describe the flame length and a predictive model for the temperature rise under the ceiling files was also proposed, which could provide a fundamental base for the fire hazard assessment inside the subway train.
Section snippets
Experimental methodology
A 1:5 subway train model was used in this study to carry out experiments, the experimental setup is depicted in Fig. 1. The dimension of this train model is 4.0 m (L) × 0.55 m (W) × 0.42 m (H). Four doors with same dimension of 0.26 m × 0.38 m (W × H) are evenly distributed along the sidewall (negative Y as shown in Fig. 1(a)). The sidewall with the doors is made of fire-resistant transparent glass with a thickness of 8 mm to enable experimental observation, and the other boundaries of the
Flame length
The behavior of the flame at the steady stage for different pans (HRR) is shown in Fig. 3. For fire scenarios in open space (Fig. 3(a)), the basically symmetrical flame gradually becomes higher with increasing HRR. While for fires inside the train with open or closed doors (Fig. 3(b), 3(c)), the fire behaviors are similar. In detail, as the HRR increases, the flame gradually changes from being lower than the train ceiling to impinging the ceiling and then spreading under the train ceiling.
Fig. 4
Conclusions
In this work, reduced-scale experiments were performed on the flame length and two-dimensional ceiling temperature profile induced by methanol pool fire to characterize the performance and propagation of fire inside subway train. Some conclusions can be addressed:
- (1)
Under the experimental conditions where the sidewalls of the subway train does not hinder the transverse extension of the flame, the door status has no significant influence on the characteristics of the fire inside the train.
- (2)
The
CRediT authorship contribution statement
Min Peng: Conceptualization, Methodology, Investigation, Writing - original draft. Xudong Cheng: Conceptualization, Supervision, Funding acquisition. Wei Cong: Investigation, Methodology. Hui Yang: Investigation. Long Shi: Writing - review & editing. Richard Yuen: Funding acquisition. Heping Zhang: Supervision.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgment
This work was supported by National Natural Science Foundation of China (No.51776192), Fundamental Research Funds for the Central Universities under Grant (No. WK2320000041), the CAS PIFI Project 2015 and the Research Grant Council of the Hong Kong Special Administrative Region, China (contract grant number CityU 11301015). We sincerely appreciate these supports.
References (54)
- et al.
Experimental investigation of pool fire behavior to different tunnel-end ventilation opening areas by sealing
Tunn. Undergr. Space Technol.
(2017) - et al.
Some source-dependent effects of unbounded fires
Combust. Flame
(1985) - et al.
Radiation fire modeling
Proc. Combust. Inst.
(2000) - et al.
A method for design of smoke control of urban traffic link tunnel (UTLT) using longitudinal ventilation
Tunn. Undergr. Space Technol.
(2015) - et al.
Experimental study on transverse smoke temperature distribution in road tunnel fires
Tunn. Undergr. Space Technol.
(2013) - et al.
Fire spill plume from a compartment with dual symmetric openings under cross wind
Combust. Flame
(2016) - et al.
Influence of sidewall restriction on the maximum ceiling gas temperature of buoyancy-driven thermal flow
Energy Build.
(2014) - et al.
An investigation of the detailed flame shape and flame length under the ceiling of a channel
Proc. Combust. Inst.
(2015) - et al.
Experimental investigation on transverse ceiling flame length and temperature distribution of sidewall confined tunnel fire
Fire Saf. J.
(2017) - et al.
Experimental study of tunnel sidewall effect on flame characteristics and air entrainment factor of methanol pool fires
Appl. Therm. Eng.
(2016)