Search

An improved multiple imputation based on R language is proposed to deal with the miss of data in a fire prediction model, which can affect the accuracy of the prediction results. Hazard and operability (HAZOP) is used to accurately find the data related to the research purpose, and exclude data with a missing rate greater than 80% and small differences in characteristics. Then, by changing the m value in the mice package under the R language (R-mice), the relevant parameters of the complete filling factor set under different m values are obtained. The value of m is determined after observing and comparing the parameters. The proposed method fully considers the randomness of filling and the difference between the generated dataset. Taking Hubei Province as an example, the data processed by this method are used as the input of the Bayesian network, and the fire trend is used as the output. The results show that the improved multiple imputation based on R-mice can solve the problem of missing data very well, and have a high prediction effect (AUC = 94.0800). In addition, the results of the predictive reasoning and sensitivity analysis show that the government’s supervision has a vital influence on the trend of fires in Hubei Province.

Guangzhou rail transit lines have intercity and subway trains in the same line, and some lines operate suburban EMUs. Obtain the best train fire evacuation rescue mode through theoretical analysis; Through numerical simulation, the available safe evacuation time and necessary safe evacuation time are obtained under different fire source positions and evacuation opening distances, and the evacuation safety of personnel is analyzed to provide decision-making basis for their safe evacuation facility setting plan. The results show that: in the case of fire, The subway evacuation mode should be adopted; When the fire occurs, the fire source position of the train should be stopped between the two evacuation openings as much as possible; The distance between the horizontal passages of the emergency rescue point is set to 20 m to meet the safety evacuation requirements of train personnel.

The widely used alkyd resin suffers not only the flammability but also the autoignition of resin waste. Its autoignition should result from the air-drying exothermic effect due to autoxidation and conversion of double bond. However, the air-drying exothermic effect is seldom studied. In this study, five soybean oil-based alkyd resins were synthesized. The content of double bond measured by the titration method increased linearly with increasing the molar fraction of maleic anhydride. After air-drying, the infrared spectroscopy absorption peak for the double bond disappeared. Considering that the curing originated from the conversion of chemical bonds, the exothermic effect was theoretically analyzed to be 1163–1455 kJ mol−1 double bond based on the air-drying mechanisms and the differences among chemical bond energies. According to thermodynamics principles, the air-drying exothermic effect was also derived from the measured heat of combustion. When the content of double bond increased from 2.19 mol kg−1 resin to 2.56 mol kg−1 resin, the exothermic effect linearly increased from 1352 kJ mol−1 double bond or 2.96 kJ g−1 resin to 2508 kJ mol−1 double bond or 6.42 kJ g−1 resin. The steady and unsteady autoignition theoretical analyses implied that due to the air-drying exothermic effect the alkyd resin waste with a certain thickness would catch fire within one working day.

X-ray markers have been applied for diagnostic purposes in the medical imaging field. However, they were not flexible enough to accommodate the fast evolution of biomedical applications leading to a considerable performance gap. EHD-inkjet printing technique as a high-resolution, low-cost, and flexible approach was used to fabricate a silver X-ray marker to fill the gap in the biomedical field. Even though X-ray characterization of bulk silver had been studied, silver nanoink has presented a different X-ray absorption response. To understand the X-ray characterization of silver microstructures, multiple samples were fabricated. The EHD-inkjet printing system was applied to fabricate silver microstructures under optimized printing parameters, including voltage amplitude, pulse width, frequency, and printing speed. To understand X-ray characteristic differences between bulk silver and nanoink silver, experiments were designed to print a single layer and multiple layers of silver nanoink lines. Microstructures, which were e-jet printed by silver nanoink, presented different X-ray absorption rate to the bulk silver under the same radiation condition and thickness. Sub-30 μm microstructures of silver nanoink were achieved by e-jet printing; part of them will serve as the X-ray marker in bio-scaffold. This research proposed a method to bridge the e-jet printing technique with a customizable X-ray marker in bio-scaffold, which could effectively resolve the problem of flexibility, improve the resolution, and reduce the cost of manufacturing. Moreover, it supported the development of biomedical products that were fabricated by silver nanoink material, such as in vivo embedded biosensors in computed tomography applications.

This study aims to examine the risk of each category of behavior related to different accident types and consequences by introducing the Monte Carlo algorithm. Totally about 234 cases were collected to construct a structured dataset, and 12 categories of passengers’ unsafe behaviors related to the four types of metro operation accidents including escalator injury, crowd stampede, pinched by a shielding barrier and fall injury were considered for this analysis. Results showed that although the total behavioral risk of each accident type obeys normal distribution with means between (4, 6] in the Moderate risk level, the variances of pinched by a shielding barrier (1.29) and crowded stampede (1.08) are much higher than escalator injury (0.60) and fall injury (0.59). The top four contributors that resulted in the accident behavioral risks were irregular behavior on the escalator, go on boarding even after the door-close alarm sounded, push each other, and pay no attention to the ground. It was concluded that due to the uncertainty of the behavior occurrence and its consequences, there are differences of the probabilities in different risk levels of the four accidents caused by the various passengers’ unsafe behaviors. Based on the results, risk mitigation measures for different accident types are recommended. This paper provides a quantitative method to compare and analyze the effect of passengers’ unsafe behaviors related to different accident types and severities, which has relative comparative significance and improves the accuracy of judging the probability distribution of risk degree that can provide basis for risk mitigation in metro operation.

This paper investigates the influence of reduced pressure on flame characteristics of elevated n-heptane fires in an aircraft cargo compartment, including the flame height driven by the weak plume ...

Surface topography and surface finish are two significant factors for evaluating the quality and dimensional accuracy of additive manufacturing (AM) parts. In general, compared with traditional subtraction and forming manufacturing techniques, the nature of the rough surface and the geometric complexity make the surface of AM parts "another surface," and traditional methods such as coordinate machine measurement may not be applicable. Most research on surface extraction focuses on regular surfaces, such as flat, cylindrical, or spherical surfaces, but less has been done with irregular surfaces. This paper presented an approach for extracting irregular surfaces based on micro-computed tomography (μ-CT) data of AM parts. The extracted data sets were then compared with data sets obtained by a structured light system (SLS). Areal surface texture parameters, cloud comparison methods, and statistical methods were applied to evaluate the difference between the surface data obtained by the two systems. The results showed that the two surface data correlated well and confirmed the capability of the proposed method for surface extraction from irregular surfaces. This work will contribute to the further multi-sensor data fusion for metrology and provide valuable information for near-surface defects prediction with real-time in-situ metrology method in additive manufacturing.

The Direct Energy Deposition (DED) process utilizes laser energy to melt metal powders and deposit them on the substrate layer to manufacture complex metal parts. This study was applied as a remanufacturing and repair process to fix used parts, which reduced unnecessary waste in the manufacturing industry. However, there could be defects generated during the repair, such as porosity or bumpy morphological defects. Traditionally the operator would use a design of experiment (DOE) or simulation method to understand the printing parameters’ influence on the printed part. There are several influential factors: laser power, scanning speed, powder feeding rate, and standoff distance. Each DED machine has a different setup in practice, which results in some uncertainties for the printing results. For example, the nozzle diameter and laser type could be varied in different DED machines. Thus, it was hypothesized that a repair could be more effective if the printing process could be monitored in real-time. In this study, a structured light system (SLS) was used to capture the printing process’s layer-wise information. The SLS system is capable of performing 3D surface scanning with a high-resolution of 10 µm. To determine how much material needs to be deposited, given the initial scanning of the part and allowing the real-time observation of each layer’s information. Once a defect was found in-situ, the DED machine (hybrid machine) would change the tool and remove the flawed layer. After the repair, the nondestructive approach computed tomography (CT) was applied to examine its interior features. In this research, a DED machine using 316L stainless steel was used to perform the repairing process to demonstrate its effectiveness. The lab-built SLS system was used to capture each layer’s information, and CT data was provided for the quality evaluation. The novel manufacturing approach could improve the DED repair quality, reduce the repair time, and promote repair automation. In the future, it has a great potential to be used in the manufacturing industry to repair used parts and avoid the extra cost involved in buying a new part.

In this research, ready-to-print clay was employed to 3D print cylindrical constructs using two different nozzle geometries. The cylinders were printed using a direct ink writing (DIW) 3D printer with similar printing parameters such that the impacts of using different nozzle geometries in contour deviation and mechanical properties of the printed constructs can be quantified. Choosing the right nozzle geometry before printing is critical as it affects the surface finish as well as the mechanical properties of the constructs. This study utilizes point cloud data of the printed samples obtained from a scanning system to measure the contour deviation and surface roughness. According to the results from the point cloud analysis, for cylindrical constructs, circular nozzle imparts less surface roughness and contour deviation whereas square nozzle imparts higher compression strength but with comparatively higher contour deviation and surface roughness. The study provides framework to determine the deviations and mechanical properties of the free form constructs. The work has provided a guideline of nozzle selection in 3D printing of clay constructs for civil infrastructures.

Origami structures, which were inspired by traditional paper folding arts, have been applied for engineering problems for the last two decades. Origami-based thin-wall tubes have been extensively investigated under axial loadings. However, less has been done with radial stiffness as one of the critical mechanical properties of a tubular structure working under lateral loadings. In this study, the radial stiffness of novel thin-wall tubular structures based on origami patterns have been studied with compression tests and finite element analysis (FEA) simulations. The results show that the radial stiffness of an origami-inspired tube can achieve about 27.1 times that of a circular tube with the same circumcircle diameter (100 mm), height (60 mm), and wall thickness (2 mm). Yoshimura, Kresling, and modified Yoshimura patterns are selected as the basic frames, upon which the influences of different design parameters are tested and discussed. Given that the weight can vary due to different designs, the stiffness-to-weight ratio is also calculated. The origami-inspired tubular structures with superior stiffness performances are obtained and can be extended to crashworthy structures, functional structures, and stiffness enhancement with low structural weight.

Experiments are conducted in a model tunnel of (1:20 of a tunnel section) to study the effect of blockage ratio on critical velocity. The vehicle blockage of the model tunnel is varied at 11 different blockage ratios, 9.4%, 15.6%, 21.9%, 25.0%, 31.3%, 34.4%, 40.7%, 43.8%, 50.1%, 56.3%, 59.5%. At the same time, the influence of heat loss in the experimental model on critical velocity is considered. The test results show that the critical velocity decreases with the increase of the blockage ratio, and the reduction ratio is less than the blockage ratio. The blockage-ratio modified coefficient obtained by experiment result is 0.545, and thus the formula of the critical velocity containing blockage ratio is proposed. The test results showed excellent agreement with from the formula of the critical velocity. The relationship between the critical velocity reduction ratio and the blockage ratio of this test and Oka and Atkinson (1995), Lee and Tsai (2012) and Kang (2010) were compared and analyzed. It is found that the main factors which lead to the different relationships about the critical velocity decrease rate and the blockage ratio are brought up, which include the height of fire burning surface, the location of fire source (the relative position about the fire source and blockage vehicles and whether the fire source is inside or outside the vehicle) and the lateral wall open area of vehicle.

A series of fire tests with various water spray flow rates were conducted in a small-scale longitudinal ventilated tunnel to investigate the influence of flow rates of a water spray system and different longitudinal wind speed on maximum smoke temperature and longitudinal smoke temperature distribution. Then, the test data for the case without the water spray system were compared with previous studies and agreed well with previous model which overestimates smoke temperature of the cases with the water spray. Results show that the maximum smoke temperature decreases and the longitudinal smoke temperature decays faster with the increase of water spray flow rate. Thus an improved model for maximum smoke temperature is proposed by introducing a modified coefficient λ representing the water spray flow rate effect. It was found that longitudinal smoke temperature decays exponentially and the decay coefficient increases with increasing water spray flow rate. Therefore considering the effect of water spray flow rate, a modified coefficient Kdown is introduced into the exponential model to predict the longitudinal smoke temperature distribution.

In order to improve the training and maintenance level of military communication equipment, in view of the practical needs of communication equipment maintenance and support training, combined with the development of information technology, a communication equipment training and maintenance system based on virtual reality technology is built. Focusing on the functional analysis and module design of the system, starting from the top-level architecture of the system, the functions, system architecture, system processes, and data realization of the virtual training and maintenance system are elaborated. The functions of communication equipment maintenance support teaching, training, and assessment can be realized.

Abstarct Although the interest in statistical analysis of human stampedes has increased to explore the characteristics of accidents, more effort is needed to address the complexity of causative factors and their relations to risk, due to their variability and uncertainty in each case. This study aims to examine the causative factors of human stampedes for different groups of cluster risk by introducing a combined method, the Iterative Self-organizing Clustering Algorithm (ISODATA) and Fuzzy Theory. Totally about 215 cases were collected to construct a structured dataset, and characteristics of human stampedes including accident site, accident consequence, accident type, manmade factors, hardware factors, environment factors and management factors were considered for this analysis. Results show that factors such as accident type has similarity with single-dimensional analysis, while accident site in the highest risk cluster has a different distribution with the proportion of temples 39% and schools 28% which is contrary to the highest 23% of stadiums in single-dimensional analysis. It was concluded that the site with high incidence of stampede accidents did not necessarily have a high degree of risk, due to it had a high frequency of accidents but causes little casualties. Based on model results, control points for different sites are recommended according to the characteristics of different cluster risk. This paper provides a quantitative method to compare and analyze the causative factors of different risk clusters, which has relative comparative significance and improves the accuracy of judging the risk degree that can provide basis for human stampede prevention and management.

Numerical simulations are conducted to study the influence of the heat release rate, the fire source location, and the tunnel slope on the smoke spread characteristics of V-shaped slope tunnels. Theoretical analysis and simulation results show that the smoke movement in the V-shaped slope tunnel is affected by the competition effect on both sides, which leads to the fact that the smoke only overflows from the tunnel portal closer to the fire source, and a large amount of air flows into the tunnel from the other portal. Due to the competition effect, the downstream smoke back-layering length first decreases and then increases slightly with the increasing distance between the fire source and the grade change point (refering to the intersection point of two adjacent slope lines). As the tunnel slope increases, both the back-layering length and the influence of the fire source location on the smoke spread decrease gradually. Besides, the heat release rate has little effect on the smoke movement characteristics. Based on the back-layering length and the mass flow rate distribution, the smoke movement behaviors in the V-shaped tunnel can be divided into the double-slope control zone, the transition zone, and the single-slope control zone. When the vertical distance between the fire source and the grade change point is more than 2.1 m, the smoke movement can be classified in the single-slope control zone. The mass flow rate reaches the peak value when the fire source is located in the transition zone, and the dimensionless mass flow rate is proportional to the tunnel slope and the 0.3 power of the dimensionless heat release rate.

In a tunnel fire, smoke flow is often divided into two regions: the near-field and the far-field. The former is a three-dimensional radial flow region close to the fire source and the latter is a one-dimensional steady flow region generally formed a certain distance away from the fire. Although the efficiency of the smoke exhaust system is influenced by the region, limited research has been done to determine the demarcation point between the two regions. In this study, a total of 27 tunnel fire computer simulations were conducted to examine the effect of heat release rate (HRR) and the tunnel's cross-sectional dimensions on the location of the demarcation point. Rather than relying on the tunnel's height, which has been commonly used in previous tunnel fire studies to determine smoke flow characteristics, a hydraulic diameter of the tunnel's cross-section was employed in this study. An improved correlation for the location of the demarcation point is proposed by comparing a theoretical dimensionless model to numerical simulation results.

An extrusion based 3D printer was used to prepare the semi-solid tablets with different drug loading dosages (75, 100, 125 mg) under ambient temperature. The active pharmaceutical ingredient, theophylline, was uploaded within the hydrogels prepared of hydroxypropyl methylcellulose (HPMC) K4M or E4M. The HPMC concentrations were adjusted to different levels (10 and 12% w/w) to fulfill the requirements for 3D printing. Rheological and textural properties, as well as release profiles, were significantly affected by the type and concentration of excipient regardless of theophylline doses used. The printing material should exhibit shear-thinning behavior, keeping yield stress less than 4000 Pa and a loss factor (tanδ = G''/G') between 0.2 and 0.7, especially for 3D printing purposes using the current platform. The SEM images demonstrated that the hydrogel matrix exhibited a porous structure, which had the potential to encapsulate the theophylline clusters within its microstructure. The in vitro dissolution test showed that the release of all tablets was extended over 12 h, and the calculation of drug release kinetic models revealed that the 3D printed HPMC matrices release the theophylline by diffusion and erosion mechanisms. The excipient HPMC K4M 12% w/w hydrogel was optimal to load the theophylline with flexible dosage combinations due to the great extrudability and shape retention ability. The exploration of rheological properties was investigated in this study, and the results revealed that it is a feasible method to predict the SSE 3D printability and quality of hydrogel-API blend materials for the drug delivery system.

Support material plays a leading role in the application of 3D printing to avoid deformation and enhance stability. This study aimed to fabricate the support structure by using hydroxypropyl methylcellulose (HPMC), which has advantages over conventional material such as low cost, low printable temperature, and high biodegradability. Once dissolved in water over gelling temperature, the HPMC based hydrogel exhibited shear-thinning behavior with decreasing apparent viscosity values at higher shear rates. The shear-dependent viscosity makes the HPMC hydrogel extrudable throughout the printing process and the printed structure stable enough without deformation. As concentration increased, apparent viscosity and storage modulus both subsequently increased. These rheological properties indicated that the concentration of HPMC K4M hydrogel significantly influenced the printability and shape retention ability, which is associated with the mechanical strength of printed filaments. The highest concentration, 12% w/v, should have the best ability to hold the printed shape over time due to the highest G' and lowest loss tangent. The printability test also showed that K4M 12% w/v could be printed into different fill density (100%, 75%, and 50%) with different patterns, i.e., rectilinear and Hilbert curve. The selection of fill density and pattern both have an effect on surface roughness and porosity. The printed support material was compatible with acrylonitrile butadiene styrene (ABS), which is the material to fabricate the main structure for 3D printing. The support material made of HPMC can be easily removed by peeling off from the main structure without visible residual.

The vertical shaft which can discharge fire smoke through the driving force of stack effect, is an effective natural smoke extraction method to control tunnel fire smoke and extend the available personnel evacuation time. Its natural smoke extraction efficiency would be affected by many factors such as the shaft structure. In order to search the optimum shaft structure, a series of tunnel fires with five kinds of shafts (the common shaft, 45 degree angle shaft, shaft groups, shaft with rectangular hood and board-coupled shaft (BCS)) are simulated by Fire Dynamic Simulator (FDS). Results show that the optimization design of the shaft structure, to a certain extent, can improve the smoke extraction efficiency, which are BCS shaft, shaft groups, 45 degree angle shaft, shaft with rectangular hood and common shaft respectively from large to small. There into, the smoke extraction efficiency of the BCS shaft can reach 70% due to the diversion and deflection of the smoke flow by BCS baffle, while the smoke extraction efficiency of ordinary shaft is 49%. Next the smoke extraction efficiency of shaft group can reach 57%, which makes the suction and penetration less likely to occur and directs the flow of fire smoke.

A series of fire experiments were carried out to investigate the plug-holing in the road tunnel with central smoke extraction system. The smoke layer height and the temperature under the exhaust vent as well as the velocity in the upstream and downstream are measured and analyzed. The results show that with the increase of smoke exhaust rate, no plug-holing, incompletely plug-holing and completely plug-holing will occur in turn for the smoke layer under the exhaust vent. The height of smoke layer under the vent increases rapidly, while the temperature decreases quickly with the increase of smoke exhaust rate, until they become stable when the completely plug-holing occurs. The velocity difference between upstream and downstream of the vent firstly increases with the larger smoke exhaust rate, and then it starts to decrease in the transitional region of plug-holing. Similarly, it becomes stable when the completely plug-holing occurs. When Fr is Fr is larger than 1.8, the completely plug-holing occurs. It is transitional state for plug-holing when 1.2 Fr

The lag time which fire plume flows from the fire origin at the ignition time to the location of interest on the ceiling where the first heat gas front reaches is important for the activation of a fire detector, especially for the early fire detection in an aircraft cargo compartment. This lag time of a fire plume was measured in a simulated aircraft cargo compartment at four atmospheric pressures (100 kPa, 90 kPa, 80 kPa, 70 kPa). Results showed that the heat release rate is the major factor in the determination of the lag time, while the ambient pressure can also have a slight effect on it. Based on Alpert’s theory and our experiments, a new correlation predicting the lag time of a fire plume has been developed in an air craft cargo compartment.

In the point smoke extraction system, induced airflow velocity which can restrain smoke back-layering is an important parameter to evaluate smoke control efficiency of smoke extraction systems. New formula for the induced airflow velocity and smoke back-layering length is proposed for the tunnel fires with different heat release rates through small-scale tunnel model experiments. The experiment results show that when the dimensionless back-layering length is more than four, the influence of smoke outlet area on the induced airflow velocity and smoke back-layering length can be negligible. The relationship between the dimensionless back-layering length and induced airflow velocity is greatly affected by the heat release rate. In point extraction system, the smoke back-layering length increases gradually with the increasing of heat release rate, then tends to be smooth.

Natural ventilation by shafts is widely applied in urban shallow tunnels due to its simple structure, steady smoke extraction efficiency and environment-friendly means. In order to investigate the influence of vehicle blockage on the smoke extraction efficiency of the shaft in tunnel fires, a series of n-heptane fire experiments were conducted in a 1:15 small-scale tunnel model with seven blockage ratios (0, 0.16, 0.24, 0.32, 0.40, 0.48, 0.56). The smoke extraction efficiency is calculated by the measured smoke temperature and CO volume fraction in the shaft and in the tunnel near the exhaust vent. Results show that the smoke extraction efficiency of the tunnel with shafts decreases firstly, then increases and drops lastly with the increasing blockage ratio. The smoke temperature in the shaft and at the downstream tunnel change similarly, which indicates that the vehicle blockage can affect the thermal energy discharged by a shaft. The main influence factors are the blocking effect, the necking effect caused by the blockage and the stack effect of the shaft, resulting in the variability of the smoke extraction efficiency. The smoke extraction efficiency reaches the maximum around the blockage 0.48 in the present tunnel, which can be referred to design the tunnel cross section in engineering applications.

Reasonable simulation length is of great significance to improve the computational efficiency. However, few research about the critical length for simulation have been conducted. In this study, critical length for simulation in tunnel fires was investigated numerically and theoretically. Through analyzing the variation of smoke layer temperature and smoke mass flow rate under different tunnel length, a criterion for selecting critical length is presented. Effect of tunnel cross-section and heat release rate on the critical length for simulation will be clarified. The linear growth relationship between the nondimensionless critical length L* and nondimensionless heat release rate Q* was obtained. This model perform good agreement with the full-scale tunnel fire test values of predecessors.

Integrated studies of accumulated data can be performed to obtain more reliable information and more feasible measures for investigating the potential diagnostic and prognostic biomarkers of breast cancer and exploring related molecular mechanisms. Our study aimed to explore the GATA family members involved in breast cancer by integrating data from The Cancer Genome Atlas (TCGA), Gene Expression Omnibus (GEO) and other online databases. We performed an integrated analysis of published studies from GEO and analyzed clinical data from TCGA and GTEx to evaluate the clinical significance and prognosis values of the GATA family in breast cancer. GATA3 was found to be upregulated and exhibited a favorable value in the diagnosis and prognosis of breast cancer. Through this study, we identified possible GATA3-correlated genes and core pathways that play an important role, which requires further investigation in breast cancer.

The extrusion-based 3D printing system was used to fabricate the bioscaffold with hybrid hydrogels of gelatin and alginate (G/A), with different total solid concentrations (3%, 5%, and 7%) and G/A ratios (1:2, 1:1, and 2:1). Rheological properties were related to the 3D printability and shape retention capacity of the hybrid hydrogels. For extrusion-based 3D printing using the current platform, the materials that were considered 3D printable showed shear-thinning flow behavior. Also, the printable materials demonstrated a storage modulus (Gʹ) higher than the loss modulus (Gʹʹ), with a loss factor (tan δ = Gʹʹ/Gʹ) in the range of 0.48–0.58 during the frequency sweep of 15–40 rad/s, which is the corresponding frequency that can be related to our 3D printing settings. Texture profile analysis indicated that among the optimal formulas for 3D printing, the bioscaffold fabricated with the hybrid gels of 7% 1:2 G/A had the highest hardness and adhesiveness. After freeze-drying, the hardness increased significantly (p

A series of numerical simulations about tunnel fires are carried out to study the performance of different arrangements of natural ventilation shafts in a 500 m tunnel. Two to four identical shafts are arranged in a group at the centerline and the sideline of the tunnel ceiling. The characteristics of the temperature distribution, velocity field and the total amount of smoke exhausted from each shaft group are investigated for each arrangement. The results have shown that when the fire heat release rate is 20 MW, boundary layer separation occurs in all the shafts in the series of simulations. At the same fire heat release rate, when the number of shafts, n, is ≤ 2, the smoke exhaust performance of the shaft groups arranged on the centerline of the tunnel ceiling is better than that on the sideline of the tunnel ceiling while n is >2, the shaft groups arranged on the sideline have a better overall smoke exhaust performance. The Richardson number of 1.4 proposed by Ji et al. is still applicable to determine the flow phenomena between boundary layer separation and plug-holing.

When tunnel fire smoke is exhausted using a top central exhausting system, the air entrainment at the smoke-air interface has a significant effect on the overall mass flux. Using theoretical analysis, the air entrainment coefficient in one-dimensional horizontal smoke movement stage in central exhausting was obtained. Both heat release rate and smoke extraction rate have effects on the horizontal entrainment coefficient. The experimental results show that the entrainment coefficient in central exhausting is between 0.01 and 0.03 and the horizontal entrainment coefficient increases to 0.0181 with increasing heat release rate and becomes relatively stable afterwards. The entrainment coefficient increases with the smoke extraction rate increasing as the smoke extraction process will enhance the smoke momentum in the tunnel.