[Objective] To prevent the accidental release of toxic gases in urban environments, it is imperative to measure the concentration field quickly and accurately. The response factor method (RFM) based on the linear response relationship between the sources and the sensors for a steady flow field and a fixed pollution source location yields good results in concentration calculations after pollutant release.It is necessary to convolve the response factor with the release function to calculate the concentration of different pollutant-release conditions. RFM can save considerable calculation time. However, current studies primarily emphasize indoor environments and lack outdoor environment simulations. Thus, we aim to address the problem of using RFM to simulate the concentration when time-varying pollution release occurs around buildings.[Methods] To confirm the reliability of the simulation results, two examples of wind tunnel data (CEDVAL experiment No. A1-5 an isolated building, and No. B1-1 building arrays) were simulated under the same conditions, and the velocity profile results were consistent. RFM consisted of the following steps:(1) The calculation domain was constructed, and the flow field was solved in the steady state. (2) The concentration field under the pulse pollutant release was solved in the transient state, the time series of the concentration response factor was obtained, and the component transfer matrix A represented by the response factor sequence was constructed. (3) The concentration at any time was calculated from the time series of the component transfer matrix A and the release intensity. RFM and computational fluid mechanics (CFD) transient simulation method were employed to simulate three types of pollutant-release scenarios:constant, periodic, and triangular releases. Based on the structural characteristics of the flow field around buildings, sensor positions of different heights and distances were selected for comparison. Next, the RFM and CFD results were compared using the commonly used statistical indexes:fractional bias (FB), normalized mean square error (NMSE), and the ratio between ratio 2 (FAC2). (1) The results reveal that the concentration RFM can effectively simulate the diffusion of pollutants around buildings. The variation trend of the concentration field aligns with the transient simulation results of CFD, and the FAC2 value is >0.95. The running time of this method is 1/30 of the CFD transient simulation. (2) For the time-varying pollution-release function, RFM can simulate the concentration field well and the concentration at any time in the process well. In this paper, RFM is applied to the case of pollutant diffusion around urban buildings. Compared with the CFD transient simulation results, the method can be applied to the rapid simulation of outdoor pollutant concentration and the influence of different time-varying pollution source release functions can be obtained. These results provide a foundation for the rapid calculation of the concentrations of multiple cases of pollutant diffusion around buildings using RFM.

[Objective] Spontaneous coal combustion is one of the major natural disasters in coal mining; thus, accurate prediction of the risk of spontaneous coal combustion is crucial to prevent and control coal fire disasters. However, the complexity of the physicochemical process of spontaneous coal combustion and its various influencing factors poses a challenge to the risk prediction of spontaneous coal combustion. Strengthening research on spontaneous coal combustion hazard prediction technology using deep learning is crucial for improving the intelligent control level of coal mine safety production.[Methods] In this study, CO volume fraction was chosen as the index for spontaneous coal combustion evaluation. A dataset was constructed, and the field observation data were visualized. Next, the dataset was tested for the distribution of eigenvariables, normalized for the distribution of eigenvariables, and normalized for the dataset using kernel density estimation, logarithmic transformation, and maximum-minimum normalization. Finally, three algorithms, namely recurrent neural network (RNN), long short-term memory (LSTM) network, and gated recurrent unit (GRU), were applied to the data mining of spontaneous coal combustion feature information, and a dynamic sequence prediction model of spontaneous coal combustion CO volume fraction was established. During the model construction process, the full connectivity layer and Dropout class were added to prevent overfitting, and the mean square error and three model performance test indicators were introduced to analyze and optimize the model parameters and test the model performance.[Results] The results were presented as follows:(1) The CO volume fraction sequence dataset was established based on the field data of the Dafosi Coal Mine, the model generalization capability was enhanced, and the training time of the model was shortened by preprocessing the dataset. (2) The RNN, LSTM, and GRU models achieved the dynamic prediction of CO with an error of less than 1 %. (3) The optimal parameters of the three models were determined from the mean absolute error (MAE), the root mean square error (RMSE), and R² of the training and validation sets. A comparative study using the model performance evaluation metrics revealed that the LSTM model had the highest prediction accuracy under the same sequence data, followed by the RNN and GRU models.[Conclusions] Using 285 sets of field data, the spontaneous coal combustion CO volume fraction sequence prediction models based on the RNN, LSTM, and GRU algorithms were established. The experimental values of the CO volume fraction were highly consistent with the predicted values, and the prediction error was less than 1 %. The model can predict the change in the CO volume fraction in future moments using the dataset. The results reveal that the dynamic time series prediction of CO volume fraction from spontaneous coal combustion using sequence models is possible compared with conventional static models. Moreover, the process of constructing the three models and optimizing the parameters can be employed as a basic study for developing sequence prediction models for other indicator gases.

[Objective] As a green energy resource, hydrogen-doped compressed natural gas (HCNG) is progressively gaining attention. A potential large-scale method of hydrogen transportation is mixing hydrogen with natural gas using the natural gas supply pipeline. However, owing to the difference in the properties of natural gas and hydrogen mixture, leakage and explosion accidents possibly occur during the transportation and utilization of HCNG. To ensure the safe promotion and utilization of HCNG, it is necessary to mix inert gas suppressants into HCNG. Therefore, this study aims to investigate the explosion characteristics of HCNG in a 0.5 m×0.5 m×0.5 m square explosion vessel. The suppressing explosion experiments are conducted using the explosion suppressants carbon dioxide and heptafluoropropane.[Methods] First, a study on the explosion characteristics of HCNG with a hydrogen volume fraction in the range of 0 % to 50 % is conducted. Subsequently, the HCNG with the 50 % volume fraction hydrogen is selected to investigate the explosion suppression based on the suppressants, carbon dioxide and heptafluoropropane. The profiles of critical explosion parameters such as maximum explosion overpressure, maximum overpressure delay time, and flame propagation velocity are analyzed under different volume fractions of suppressants.[Results] Results show that the maximum explosion overpressure and flame propagation velocity increase with increasing hydrogen doping ratio. With increasing volume fraction of suppression gas, the maximum explosion overpressure rapidly decreases, and the delay time to reach the maximum overpressure is prolonged. The high-speed image acquisition results indicate significant stretching of the flame surface. The spherical flame propagation velocity is considerably decreased by the suppressant. Subsequently, the typical volume fraction interval of the two suppressants is selected for comparing their explosion suppression effects. The explosion suppression effects of carbon dioxide and heptafluoropropane on the HCNG are quantified.[Conclusions] The suppression effect of heptafluoropropane is found to be generally comparable with that of carbon dioxide at twice the volume fraction. Complete explosion suppression is attained using 10 % volume fraction heptafluoropropane or 20 % volume fraction carbon dioxide, and the maximum explosion overpressure is decreased by 98.0 % and 94.4 %, respectively. The reasons for the suppression effect of heptafluoropropane being considerably superior to that of carbon dioxide are discussed based on the physical and chemical differences between the two suppressants. In particular, heptafluoropropane demonstrates better abilities of oxygen isolation, combination of explosive reactive radicals, and absorption of reaction heat than those of carbon dioxide. The current results can serve as a guide for improving the transportation safety of HCNG.

[Objective] Coal continues to be a vital component of China's energy infrastructure, but the inherent explosion risk associated with coal dust significantly hampers industrial safety and sustainable economic growth. Given these challenges, conducting in-depth research on efficient explosion inhibition technology for coal dust and understanding its underlying mechanism are urgent and necessary. Researching efficient explosion inhibition technology and its mechanisms represents a scientific challenge and an essential step toward achieving sustainable economic development.[Methods] In this study, we adopted a thorough and systematic approach to investigate the thermal decomposition characteristics and product distribution of raw coal and phytic acid (PA)-containing coal samples at high temperatures. Our research methodology combined thermogravimetric experiments, mass spectrometry analysis, and other techniques. During the thermogravimetric experiment, we accurately recorded the mass change of the coal samples during the heating process. The data served as the foundation for subsequent mass spectrometry analysis. By monitoring the thermal decomposition process of the coal samples, we revealed in detail the thermal decomposition characteristics of the raw coal and PA-containing coal samples in the high-temperature environment. We paid particular attention to the dynamic changes in the gases and volatiles produced during the heating process of the coal samples. Following the thermogravimetric experiment, we conducted a mass spectrometry analysis using in-situ pyrolysis time-of-flight mass spectrometry (in-situ-Py-TOF-MS). With this technique, we monitored the gases and products generated during the pyrolysis of the coal samples in real time. This provided us with high-resolution data that supported our analysis of the relative abundance and the generation mechanisms of these gases and products.[Results] The research results indicate that the inclusion of PA-containing inhibitors can significantly reduce the content of light combustible gases (mainly CH₄ and CO) and organic volatiles (mainly alkenes, aromatic compounds, phenolic compounds, and dihydroxy aromatic hydrocarbons) during coal dust pyrolysis. The main chemical components generated by the pyrolysis of PA water mist include[(HO)₂PO]₂O, PO(OH)₃, and HOPO₂. These small phosphorus-containing molecules can effectively terminate the chain reaction of explosions by consuming key flame radicals such as H and OH. Additionally, the addition of PA water mist results in a 15.8 % decrease in the heat release of coal dust samples. Concurrently, the activation energy increases from 11.3 kJ/mol to 16.4 kJ/mol. The inhibition mechanism of PA-containing fine water mist on coal dust explosion is further analyzed by measuring the pyrolysis products of coal samples with and without detonation inhibitors.[Conclusions] PA pyrolysis yields[(HO)₂PO]₂O, PO(OH)₃, HOPO₂, and other small molecules of phosphorus-containing components. These phosphorus-containing explosion inhibition components can be achieved through the removal of flame radicals, bond breaking, and free radical extraction reactions involving oxygen or hydrogen atoms. Additionally, these components can initiate hydrogen atoms or hydroxyl addition reactions, all of which collectively work to inhibit the chain reaction of dust explosion. This, in turn, reduces the concentration of key flame radicals.

[Objective] Fires and explosions caused by gas leaks result in serious casualties and property damage. Therefore, the precise detection of gas leaks is crucial. The primary component of natural gas is methane, and gas leak detection aims to accurately detect methane gas. However, methane detectors may face interference from adverse environmental factors during the gas leak detection process, affecting the performance of the detectors.[Methods] This study systematically analyzed the influence of water immersion and the complex conditions of low-temperature immersion on the response characteristics of catalytic combustion methane detectors. The catalytic combustion methane detector was initially immersed for 15-180 mins. The detector's alarm concentration and response time were then tested to evaluate the impact of immersion on the alarm performance and response speed of the catalytic combustion methane detector. Subsequently, tests were conducted on the alarm concentration and response time of the catalytic combustion methane detector immersed for 15 and 30 min at environmental temperatures of 30, 15, and 0 ℃. Furthermore, an analysis was conducted on the reasons for the impacts of low temperature and water immersion on the catalytic combustion detector, considering factors such as resistance, absolute humidity, moisture content, and catalyst-specific surface area.[Results] The results revealed that immersion significantly reduced the sensitivity of the methane detector, with a 90.3 % increase in the alarm concentration after immersion. The time required for the display values of the detector to reach 90 % of the baseline value increased by 115 min. Compared with room-temperature immersion, low-temperature immersion resulted in a 23.3 % decrease in the alarm concentration. The time required for the display values of the detector to reach 90 % of the baseline value increased by 55 min, but the repeatability of the response time improved, with an average standard deviation reduced by 65.16 %. After immersion treatment, a thin film of water formed on the surface of the sensor, impeding the entry of methane gas into the sensor's interior and decreasing the internal methane concentration. Additionally, water immersion reduced the catalyst's specific surface area and led to catalyst poisoning, thereby diminishing the available adsorption sites for methane. Therefore, this increase in methane concentration was required for the detector to reach the alarm state. Moreover, the platinum wire heater of the detector showed more significant resistance changes at low temperatures, requiring less heat from the catalytic combustion reactions to achieve the necessary resistance change for triggering the alarm, thereby reducing the required methane concentration for the detector's alarm. The absolute humidity of the testing environment decreased by 92.9 % at low temperatures, reducing the impact of environmental humidity on the detector. Therefore, the low-temperature environment effectively mitigates the adverse effects of immersion on the detector.[Conclusions] This study examined the variations in the response characteristics of catalytic combustion methane detectors under low temperature and water immersion conditions. This study also analyzed the reasons for the impacts of low-temperature and immersion environments, thereby providing experimental evidence for addressing the impact of complex low-temperature immersion conditions on methane detector performance. The results are beneficial for enhancing the accurate detection of methane leaks.

[Objective] Currently, many achievements in the study of subway car fires are mainly concentrated on the full development stage of fires. However, in-depth research on early fire detection and improving detection efficiency is limited. Further research is necessary on the response time of fire detectors in subway car fire incidents, considering the influence of the air conditioning system.[Methods] To investigate the influence of fire source location on the response time of fire detectors in subway cars under an air supply environment, this study examines the smoke spread patterns and detector response times in cars with different fire source locations. Using an actual subway car as the basis, a simulation model of the car is created. In this model, air is uniformly supplied downward through the air supply port at the top of the car, while return air flows through ports at the top of the car. The passenger area contains seats, and the top of the car features four return air outlets and four waste exhaust outlets that are symmetrically arranged. The boundary conditions of the car air conditioning system mainly consist of supply air, return air, and waste exhaust boundaries. The total air intake of the whole car is 10 000 m³/h, the total waste displacement is 3 850 m³/h, and the total return air volume is 6 150 m³/h.[Results] In the fire detection experiment, the location of the fire source directly below the air supply port on the top of the car is more affected by airflow and is the most unfavorable detection position for fire detectors compared with the locations between the two air supply ports. The velocity is larger at the return air outlet. If a fire occurs near the return air outlet, smoke may enter the air supply system at the top of the car through the return air outlet. In this case, the fire cannot spread into the C-slot in time, which may prolong the detector response time. Higher wind speeds will prevent the smoke from spreading to the top of the car, thus preventing the smoke from entering the C-slot quickly enough to reach the fire detectors. Therefore, the location of the cross-section where the air supply outlet is located is less favorable for fire detection than the cross-section in the center of the car.[Conclusions] For the subway car modeled in this study, placing two detectors in the C-slot is relatively reasonable because they can detect smoke within 60 s in all the fire scenarios considered. The findings provide insights into detector placement in subway cars, aiding in enhancing fire detection strategies.

[Objective] With the development of airworthiness certification for domestic commercial aeroengines, research in domestic civil aviation has continued to deepen. As one of the core components of the aeroengine, the combustor case has been subjected to long-term testing using high-temperature and high-pressure flames. The safety of the aircraft would be compromised if the combustor case is burned-through. The Federal Aviation Administration (FAA) has included burn-through safety as one of the criteria for engine airworthiness certification in its regulations. From the perspective of simulation and experimentation, foreign scholars have conducted a series of studies on the combustor case burn-through phenomenon. However, few relevant studies have been conducted in China.[Methods] The development of a set of jet flame test bench systems is presented in this paper. The system is designed to simulate the emitted flame from the combustion hole after the combustor case is burned. The test bench system can meet the following specifications: First, the temperature of the generated jet flame at the nozzle should not be below 1 648 ℃. Second, the internal pressure of the burner must be 0.74 MPa based on the standard operating conditions. The effects of the nozzle aperture, oil-gas mass ratio, outer air velocity, and internal pressure on the axial temperature and heat flux distribution of the jet fire are studied using the system through the single variable method.[Results] Experimental results are as follows: (1) The jet flame takes on a concentrated form, revealing a rapid decrease in the temperature and heat flow in the axial direction. (2) The oil-gas mass ratio notably affects flame temperature and heat flux, demonstrating a positive correlation. Specifically, at distances of 127 mm, the increases in temperature of 18.6 % and heat flux of 13.0 % are observed when the oil-gas mass ratio increases from 0.035 to 0.050. (3) The flame temperature and heat flux are also affected by the outside air flow rate. The temperature and heat flux have a considerable impact at distances of 254 to 508 mm and 127 to 254 mm, respectively. (4) The internal pressure of the burner considerably influences the flame temperature and heat flux. However, the data changes cannot be observed due to the limitations of the test bench system. (5) The size of the nozzle aperture has a significant impact on the shape and temperature of the flame. The flame concentration increases when the aperture is small. At distances of 127 to 381 mm, the temperature initially increases with aperture size but eventually declines. However, at distances of 508 to 762 mm, a high temperature at the corresponding position is observed under a large aperture. The trend of heat flux follows the same pattern as that of temperature.[Conclusions] To prevent the hazards of possible combustor case burn-through, aeroengine design should focus on the provision of a fire barrier within 254 mm outside the normal direction of the weak point of the combustor case.

[Objective] In practice, conductors are often arranged in multiple parallel rows. When twin conductors are closely arranged in parallel, air entrainment can be considerably affected during fire spread, resulting in changes in fire spread characteristics and molten and dripping behaviors. However, most previous studies have mainly focused on the fire spread characteristics of a single conductor. Limited studies exist regarding fire spread over twin parallel conductors, especially with different inclination angles. This study aims to provide a superior understanding of the effects of the inclination angle on the fire spread of the conductors and compare the differences in the fire spread characteristics of the single and adjacent twin conductors.[Methods] Herein, various closely comparative experiments are conducted to investigate the fire spread and molten and dripping behaviors of the single and adjacent twin conductors under various inclination angles (0°, 30°, 60°, and 90°). Polyethylene-insulated conductors with an inner copper core diameter of 0.97 mm and an insulation thickness of 1.15 mm are selected as test samples. The essential parameters, including flame pattern, flame length, flame height, and flame spread rate, are simultaneously obtained and analyzed.[Results] The main results of this paper are as follows:(1) Whether for single or twin conductors, the molten and dripping behaviors will occur during the flame spread process, resulting in the flame pattern experiencing cyclic changes in the stable, growth, attenuation, and adjustment stages. The frequency of molten and dripping behaviors for the adjacent twin conductors is greater than that for a single conductor. (2) The average flame length, length of the bare wire wrapped by the flame, and length of the insulation layer heated by the flame increase with the inclination angle; however, the average flame height first increases and then decreases with the increase of the inclination angle. Except for the 0° cases, the average flame height and length of adjacent twin conductors are larger than those of a single conductor. (3) The flame front shows a periodic change over time owing to molten and dripping behaviors, and the fire spread rate increases with the inclination angle. However, the fire spread rate for adjacent twin conductors is consistently less than that for a single conductor mainly because for the adjacent twin conductor arrangements, an overlapped area of air entrainment and heat transfer exists, air entrainment is restricted, and consequently, the burning rate is weakened, which reduces the average heat flux transferred by the flame to each conductor.[Conclusions] Based on the above analysis and investigations, the inclination angle and arrangement of adjacent twin conductors have considerable effects on the fire spread and heat transfer over conductors. Moreover, the flames over twin conductors with a certain spacing may propagate with a high fire spread rate owing to the enhanced flame interactions, which is an interesting question worthy of further study.

[Objective] In real fires, the shape of the flame front on the discrete fuel bed varies greatly with fuel density. For example, A- and V-shaped flame fronts have opposite shapes. Researchers have conducted limited studies on the shape of flame fronts on discrete solid fuels. Thus, the formation mechanism of different flame front shapes needs further exploration.[Methods] In this experiment, the 5 mm (length)×5 mm (width)×60 mm (height) pine wood strip was selected as fuel. An aluminum composite plate (an open plate with hole spacing of 3, 5, and 7 mm and a support plate with a thickness of 3 mm) was designed to fix the fuel. The test results were obtained using open plates with the mentioned hole spacing. The open plate had a thickness of 2 mm, i.e., the pine wood had an embedding depth of 5 mm, and the exposed pine wood featured a length of 55 mm. In addition, two 60 frames/s cameras were used to film the fuel combustion process. One camera was placed directly above the fuel platform, and the other was situated on the side. Before the experiment, the height and focal length of the cameras were adjusted to ensure that the cameras could capture the entire flame and fuel table. An electronic balance was placed at the bottom of the fuel platform, and the mass change during combustion was recorded at a sampling rate of 1 Hz. Two heat flow sensors were installed side-by-side at a certain spacing of wood strips at the end of the array to measure the total and radiant heat fluxes during the spread process, with a maximum range of 10 kW/m² and an acquisition frequency of 5 Hz. The convective heat flux was obtained through the subtraction of the radiant heat flux from the total heat flux. In this experimental study, a linear ignition source was used, i.e., the first row of fuel was ignited simultaneously. During the ignition process, a stainless steel plate was placed between the first and second rows of wood strips to avoid the effects of preheating.[Results] Through measurement of the heat flux and mass loss during the fire spread process, flame front changes in discrete horizontal fuel fire spread showed the following: (1) The wood strip arrays with 3 and 5 mm spacing exhibited a V-shaped flame front, and the sparse wood strip array with 7 mm spacing presented an A-shaped one. (2) In the stable combustion stage, the mass loss rate approximately reached a balance, and the mass loss rates of 5 and 7 mm wood strip arrays were close and more significant than that of 3 mm wood strip array. (3) With the increase in spacing, the peak value of the total heat flux released by combustion of the wood strip arrays increased.[Conclusions] With the increase in spacing, the shape of the flame front gradually changed from "V" (faster flame spread on both sides) to "A" (faster flame spread on the center line). Based on the heat transfer theory, the shape of discrete fuel fire spread front was analyzed via convection and radiation, and the radiation heat transfer model was established. The experimental results are in reasonable agreement with those of calculations.

[Objective] Evacuation simulation is an important method for studying tunnel safety. Previous studies on highway tunnel evacuation simulation have assumed a uniform distribution of various vehicle types across each lane. However, this overlooks the potential impact of differences in vehicle type distribution across lanes (DVTDLs) and random vehicle distribution on evacuation time. In practice, different vehicle types tend to gravitate toward specific lanes, and the volumes as well as personnel loads vary significantly among these vehicles. We propose an algorithm for the random arrangement of vehicles within a tunnel, considering DVTDLs and random vehicle distribution. Concurrently, we develop a system with randomly arranging vehicle types (RAVT).[Methods] Our proposed method randomizes the arrangement of vehicles using parameters such as the spatial coordinates of the numerical model of the tunnel, vehicle sizes, vehicle model ratios, and vehicle type distribution differences. We apply this algorithmic process to develop the RAVT system. This system, based on HTML5, CSS, and JavaScript, can quickly calculate the parameters of vehicle arrangements and personnel loads and automatically generate the corresponding fire dynamics simulator (FDS) codes. This feature enables efficient and accurate vehicle arrangement in the numerical model of fire and evacuation of road tunnels. The FDS codes serve as a tool for modeling fires and can be directly imported into Pyrosim (a graphical tool integrated with FDS) and Pathfinder (a graphical tool for evacuation simulation). Using RAVT and Pathfinder, we established 12 scenarios featuring two- and three-lane tunnels as the subjects of our research. A total of 240 simulations were conducted. Our analysis focused on the effects of DVTDLs and random vehicle distribution on evacuation time under various conditions, ranging from extremely congested (vehicle spacing of 1.5 m), and normally congested (vehicle spacing of 5 m), to noncongested (vehicle spacing of 60 m). The lane model distribution ratios were set according to both uniformly and nonuniformly distributed proportions of each lane model, the latter being more reflective of real-world situations.[Results] The results reveal several key insights:(1) DVTDLs significantly affect evacuation time under extremely congested and normally congested conditions, with the mean difference of evacuation time reaching 43.51 s at the maximum. (2) Under noncongested conditions, DVTDLs have little effect on the average evacuation time. However, a uniform distribution of vehicle models across each lane increases the variability of evacuation times. (3) Large vehicles are likely to become stranded at the pedestrian crossing entrances, particularly in congested conditions and when vehicle types are uniformly distributed across lanes. This is especially true for large vehicles in overtaking lanes, leading to potential congestion during evacuation and, subsequently, longer evacuation times. (4) Given the same number of vehicles and vehicle type proportions, different vehicle distributions can cause significant differences in evacuation time. The maximum difference observed between the longest and shortest evacuation times was 59.76 %.[Conclusions] By applying RAVT and Pathfinder software and analyzing the results, we have gained valuable insights into the effects of DVTDLs and random vehicle distribution on evacuation simulation results. We recommend that researchers and engineers consider these factors when conducting tunnel evacuation simulations. The algorithm and system developed in this study, along with the simulation method employed, provide a more scientifically robust reference for tunnel evacuation simulations.

[Objective] A spill fire can cause considerable property damage and casualties. Although research on spill fires in open spaces is abundant domestically and internationally, research on spill fires in confined spaces is limited.[Methods] To investigate the combustion spread law and temperature field distribution characteristics of spill fires in confined spaces, a 1:10 reduced-size confined space model was constructed. Seven types of spill fire tests with different mass leakage rates and 12 types of oil pool fire tests with varying sizes were carried out. The analysis focused on the fire source diameter, mass combustion rate, and temperature field distribution characteristics in the confined space. The dimensions of the confined space model were 9 m long, 0.6 m wide, and 0.45 m high. The walls were covered with 2 cm-thick asbestos panels. A fireproof glass with a length of 1 m and a width of 0.45 m was placed horizontally in the center of the model as a spilling burning platform for leaking fuel. The platform was 4 cm away from the bottom of the model. To simulate fuel leakage, a stainless-steel pipe with a diameter of 1 cm was used as a leakage pipe, extending vertically from the ceiling down to the flowing platform. In addition, an electronic balance with an accuracy of 0.1 g was placed under the fuel barrel to measure the fuel leakage rate in real time.[Results] The spill fires in the confined space undergo three stages in accordance with the changes in the fire source diameter over time. These stages are combustion diffusion, stabilization, and extinction. The transition between stages is mainly determined by the relative sizes of the fuel mass leakage rate and mass combustion rate. Equilibrium between the mass combustion rate and mass leakage rate leads to complete fuel combustion before reaching the flame front, the fire source diameter stops increasing, and the combustion enters the stabilization stage. The mass combustion rate of ethanol fuel in a confined space demonstrates a linear correlation with the fire source area because the change in mass combustion rate per unit area exerts less influence on the fuel mass combustion rate than the change in fire source area. During the combustion stabilization, the temperature field distributions of the spill and oil pool fires are similar. The temperature at the root of the fire source gradually decreases as the power of the fire source increases, while the temperature at the top rises continuously. Moreover, the temperature of the ceiling in the confined space increases with the power of the fire source and decreases longitudinally. Finally, a prediction model of ceiling maximum smoke temperature rise during the steady combustion stage of spill and oil pool fires is established.[Conclusions] The findings can provide theoretical support for the detection design of spill fires in confined spaces.

[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.

[Objective] Firefighting gear serves as a crucial barrier for firefighters to prevent skin burns and reduce life-threatening risks during fire emergencies. Scientifically equipping firefighters with appropriate gear and enhancing the capability to assess on-site risk of disaster incidents are crucial for reducing casualties among firefighters. During a fire, the core of the heat transfer process revolves around the system of environment, clothing, and human body. A comprehensive analysis of the heat transfer mechanisms of the system is a scientific approach to improve the thermal protection level of the human body. This study aims to reveal the heat transfer mechanisms of the clothed human body to enhance occupational protection for firefighters.[Methods] The flame manikin system was employed in physical experiments to establish and validate numerical models. A geometric model of the combustion chamber with a clothed human body was established using a three-dimensional body scanner and reverse engineering technology. The turbulence, combustion, and radiation models were determined based on combustion chamber characteristics. The initial conditions, boundary conditions, and solution methods for the model were determined to calculate the system's heat transfer process. A skin heat transfer model with actual skin thickness was developed to predict skin burns. This model was combined with the flame manikin model to conduct virtual flame manikin tests under various thermal conditions. The fire environment and clothing are crucial elements in human thermal protection research. Herein, heat flux intensity and heat exposure duration were selected as environmental factors. Clothing thickness and surface emissivity were selected as clothing factors. All factors were set at nine levels, and a four-factor, nine-level orthogonal experimental design was used with 81 simulation calculations.[Results] The results of the simulation research indicate the followings. (1) The increase in heat flux intensity has the most substantial impact on the proportion of skin burns. (2) The proportion of skin burns generally increases with the increase in heat exposure duration. When the heat exposure duration is long, burn injuries tend to concentrate on the upper body because of the accumulation and increase in hot airflow. (3) With the increase in clothing thickness, the variation in the proportion of first-degree burns on the body's skin surface exhibits fluctuations. The proportion of second-degree burns considerably decreases between clothing thickness levels 1 and 3. (4) No clear pattern was observed in the proportions of skin burns with increasing clothing surface emissivity.[Conclusions] Determining a safe working distance or safe working time is important for the occupational safety of firefighters. The impact of clothing factors on burn proportions differs from the results of fabric studies. In the clothed state, different angles between the local body surface and heat source affect the actual incident heat flux on the clothing surface. Systematic analysis of burn degree distribution reveals that body areas with clothing openings, such as the neck, forearm, and calf, are the most susceptible to burns. Enhancing the design of clothing openings and improving the thermal protective performance of firefighter gear can benefit firefighters.

[Objective] In firefighting and rescue scenarios, firefighters often confront a combination of high-temperature, heat radiation, and physical exertion, frequently leading to heat stress reactions among firefighters and resulting in fainting, injuries, and casualties, which affects their physical and mental well-being. Moreover, thermal sensation and comfort significantly impact the operational efficiency of firefighters and the duration of effective rescue efforts. This analysis focused on the effectiveness of cooling vests in alleviating heat stress and improving thermal comfort for firefighters. Herein, we mitigate heat stress levels among firefighters, enhance the operational efficiency and the duration of emergency rescue operations, and validate the effectiveness of cooling vests.[Methods] This study utilized a high-temperature environmental chamber and silicon carbide radiant panel to construct a high-temperature and heat radiation experiment setup. Experiments were conducted in a composite environment at 35 ℃ and 2.5 kW/m² heat radiation at two physical activity intensity levels, 4.5 and 7.0 km/h. This study involved physiological experiments on the human body and the use of subjective questionnaires to investigate the impact of cooling vests on the physiological and subjective responses of firefighters.

[Objective] Electric power operators often encounter various safety risks that severely threaten their life and health. However, accurate identification of such risks is challenging because of the complex nature of work scenarios and the high randomness of risk factors. Currently, the analysis of electric power operation risks mostly involves the qualitative analysis of the entire operation process. Therefore, there is an urgent need to use appropriate technical means to quantitatively analyze the risks in electric power operations, providing safety assurance for electrical power operators.[Methods] This study develops a dynamic adaptive risk identification (ARI) model compatible with various typical spatial scenarios in the electric power industry. First, we build a macrolevel risk identification framework for typical electric power operations. Second, we establish a comprehensive risk factor classification system for multiple risk sources associated with typical operations. Finally, the ARI model is proposed, which incorporates the progressive honey badger algorithm, maximum entropy criterion, and constraint adjustment mechanism. Case studies validate the effectiveness of this model.[Results] The research outcomes based on case studies demonstrate that the ARI model exhibits high robustness, high cost effectiveness, and low bias risk. Compared with traditional risk identification methods, the ARI model effectively mitigates the effect of subjective judgments by integrating objective criteria. Its more balanced weight distribution reduces unwarranted subjective assumptions arising from inaccurate objective risk data. If the workspace undergoes alterations, there is no imperative need to solicit experts for a reassessment because the ARI model dynamically adjusts the distribution of factor weights based on typical spatial features using the progressive honey badger algorithm and constraint adjustment mechanism. This dynamic adaptation facilitates cost-effective and efficient risk identification in typical spaces.[Conclusions] The proposed model effectively quantifies the multisource element risks of typical operation scenarios in the electric power industry. Integrating the model with existing risk information systems further enhances precision and efficiency of existing risk control measures, providing crucial technical support for the safety assurance of electrical power operators.

[Objective] Since the proposal of the "21st Century Maritime Silk Road", countries and regions along the route have been interconnected in various fields. The continuously improving maritime transportation facilities and equipment have guaranteed the efficient development of the regional economies along the route. However, the current traffic safety situation of the Maritime Silk Road continues to face challenges. Natural conditions and geographical environment are important factors that influence the safety of ship navigation and the efficient operation of shipping routes. The stability safety of ships is subject to the dual constraints of the ship's stowage status and external wind and wave conditions. The natural conditions of navigable waters are complex and variable, and it is imperative to understand the marine climate and weather change characteristics of the route. To measure the impact of wind and wave factors on the rolling risk of ships during navigation, an analytical method to determine the environmental impact on stability safety of ship navigation is proposed.[Methods] A data processing framework for wind and wave environmental factors along the Maritime Silk Road was established based on the Kriging interpolation technology and the adaptive weighted average method. Combined with 22 years of meteorological data in the sea area, a navigation risk map for the key waters of the Maritime Silk Road was developed. Based on quantitative analysis of the influence of wind and wave environmental factors on ship heeling, the heeling risk index for evaluating ship stability safety was proposed. According to the model reconstruction of typical ship capsizing accidents, the impact of external wind and wave factors on ship stability safety was revealed.[Results] The results reveal the following:(1) The risk is high in the South China Sea, medium in the Arabian Sea, and low in the Bay of Bengal. (2) The navigation environment of the Maritime Silk Road route is significantly affected by typhoons:(a) During the typhoon season, there are high risks in the East China Sea and the South China Sea from June to August, as well as in the Arabian Sea from March to May. (b) During the nontyphoon seasons, due to the influence of monsoons and cold air, winds and waves increase. High risks appear in the Yellow Sea, the East China Sea, and the Bay of Bengal from November to December and January. (3) Based on the constructed environmental impact model for ship stability safety, the waters near the Gulf of Aden and the Strait of Malacca exhibit significant narrow tube effects, indicating a significant level of high navigation risk. Special attention should be given to the wind and wave conditions in the above sea areas to ensure the stability safety of ships. (4) According to the analysis of accident cases, when the heeling risk index is between 0.5 and 0.8, it is important to assess the heeling risk as early as possible and prepare for emergency deployment in a timely manner. When the heeling risk index is in the range of 0.8-1.1, emergency deployment should be implemented. When the heeling risk index is >1.1, evacuation should be performed in a timely manner to ensure human safety. Further comparative analysis of similar cargo transportation cases was conducted, and the impact of the internal environment of ships on ship stability safety was discovered, verifying the fundamental role of ship attributes and cargo performance in ensuring ship stability. Particularly, the stability of cargo behavior is as important as the navigation environment and should be included in the stability safety impact system of ships.[Conclusions] This study provides a reference for establishing a Maritime Silk Road traffic stability safety system with navigation safety and resilience as the core.

[Objective] During takeoff of civil aviation aircraft, pilot operational errors can easily lead to accidents, such as tail strikes. The aircraft takeoff attitude is closely related to flight operations. Therefore, focused on the decoupling problem between aircraft takeoff attitude and flight operations, guided by the "Safety II" mode, the coupling mechanism between aircraft takeoff attitude and flight operations is studied. This investigation considers the risk response performance throughout the entire process of aircraft takeoff.[Methods] The research is based on the flight quick access recorder (QAR) data of a domestic airline's A319 fleet. A coupling model of aircraft takeoff attitude, flight operations, aircraft performance, and flight environment is established using dynamic Bayesian networks (DBN) and Genie software for parameter learning and modeling. Daily flight data are deeply explored and fully utilized to study the causal-time series coupling mechanism between flight operations and aircraft takeoff attitude. Initially, based on the theory of safety resilience, the causal-time series coupling (CTC) model is developed to analyze the aircraft takeoff attitude and flight operations. Then, based on the QAR data of 1 359 flight segments of a domestic airline's A319 fleet, the CTC-DBN model is quantified and validated using Genie software. Results show that the CTC-DBN model can effectively analyze the dynamic formation mechanism of aircraft takeoff attitude. Finally, single and combined scenarios, such as heavy load, light load, downwind, and headwind, are selected to determine the optimal flight operation mode by adjusting the probability distribution of key flight operation nodes. The coupling mechanism between aircraft takeoff attitude and flight operations under different scenarios is studied, ultimately improving the pilot's ability to respond to risks in advance.[Results] The results indicate that the model can effectively analyze the dynamic formation mechanism of aircraft takeoff attitude. (1) The difference in aircraft weight is mainly reflected in the different throttle commands in the moment of rotation and the varied pitch commands in the moment of takeoff. A relatively large weight of the aircraft indicates considerable throttle in the moment of rotation and rapid pitch command in the moment of takeoff to obtain sufficient lift for the aircraft. (2) The different wind directions are mainly manifested by varied throttle commands at three distinct moments. The throttle of the headwind scenario is greater than that of the tailwind scenario in all three instances, thereby overcoming the wind speed and obtaining sufficient airspeed to ultimately ensure sufficient lift for the aircraft. (3) Compared with the optimal flight operation modes of four single scenarios, the combination of two scenarios increases the throttle commands due to heavy weight and headwinds. The throttle command exhibits a decreasing trend owing to its small weight and downwind. The pitch command at the time of takeoff in various scenarios has immediately become the main mode.[Conclusions] The causal-time series coupling mechanism between flight operations and aircraft takeoff attitude is studied using the CTC-DBN model. This research ultimately provides guidance for pilot operations and improves the risk response ability of pilots during the aircraft takeoff process. Subsequent research should combine other data, such as terrain, meteorology, and pilot characteristics, to conduct in-depth studies on different types of takeoff and landing.

[Objective] Emergency relief supplies are crucial for dealing with disasters, and their reasonable and timely distribution relates to people's health and safety. Emergency relief supplies at rescue centers are limited and cannot meet the emergency needs of all affected areas simultaneously. Post-disaster emergency relief supplies face double challenges in this regard due to short supplies and limited transportation capability, resulting in the needs for medical rescue and materials of a disaster area in a short period. To develop a scientific and efficient post-disaster emergency response, we studied the constrained vehicle routing of emergency relief supplies based on demand urgency. Restrictions in traditional path planning, such as single objective, single depot, single distribution, undifferentiated supply, and closed scheduling, were considered.[Methods] The analytic hierarchical process was applied to measure the demand urgency index from personnel, facilities, and disaster resistance, considering the overall efficiency and key disposal. Furthermore, this study had multiple objectives, including the following: minimization of deprivation cost and response time, and maximization of demand satisfaction rate in the emergency rescue process. A constrained model of emergency vehicle routing was constructed, and a two-stage genetic algorithm was designed to deal with comprehensive distribution conditions, such as open scheduling, soft time windows, and demand splitting. The effectiveness and feasibility of the model and algorithm were verified using examples.[Results] The results revealed that the model effectively coped with the material distribution problem resulting from scarce transportation capacity and various degrees of disaster. The splitting strategy and open scheduling of vehicles guaranteed multiple services at disaster sites and optimal route combination. Moreover, relief progress in disaster sites (splitting demand, batch distribution, and service time) and vehicle dispatch schedules (distribution order, work duration, and resupply depot) were generated. During the planning period, the system loss was reduced by 40.3 %, and a 99.4 % material demand was obtained. When disaster derivation caused changes in road conditions, fluctuation parameters were inputted into the model. The model and algorithm adjusted the scheme with a low risk of service failure, and the adjusted scheme reduced the demand and supply by 1.5 % in the decision period.[Conclusions] Constrained route planning is implemented for flexible distribution conditions, such as demand splitting, soft time windows, and open scheduling, based on the dynamic change characteristics of demand and supply during sudden natural disasters. This study considers the demand urgency of key disaster areas and the efficiency of global relief to accommodate unexpected road events and maximize resource availability. With the circulation of distribution vehicles, the needs of disaster sites are gradually met within the decision-making cycle, which provides full play to the time utility of emergency relief supplies and transportation resources. The proposed model can form scientific and reasonable material distribution and vehicle scheduling schemes and evaluate the workload of each rescue center and vehicle to deploy work in advance, providing a theoretical basis and a decision-making reference for vehicle route planning of emergency relief supplies.