1 Introduction

A large number of pollutants are generated during tunnel excavation and construction. In order to ensure the working environment of construction personnel, it is necessary to discharge these pollutants out of the tunnel or dilute them below the safe concentration range. The ventilation system is responsible for providing fresh air to the tunnel and discharging pollutants. The effectiveness and reliability of the ventilation system are related to the construction safety and the environment in the tunnel. Therefore, fully understanding the diffusion and migration mechanism of pollutants is an important basis for selecting and optimizing the ventilation system.

Cao made statistical analysis on the explosion smoke accidents in the project [1]. The data study found that the main cause of the accident was the poor ventilation effect and the failure of ventilation time prediction. Harmful gas accumulates in the construction area and becomes the root cause of explosion smoke accidents. Therefore, it is necessary to study the migration and dilution laws of harmful gases, which are the main components of pollutants. In order to acquire the the gas dispersion evaluation in tunnel ventilation, Widiatmojo et al. [2] proposed a Lagrangian based Particle Tracking method.This technology is less restrictive to the applicability of large ventilation networks. Yang, Shen, and Liang [3] conducted tests in a Shanghai underground tunnel to investigate the effects of air velocity and traffic distribution. Their results show that complementary ventilation can improve tunnel ventilation effect and reduce potential pollution. Zhou et al. [4] explored the influence of various factors in the tunnel on the performance of forced ventilation. They put forward an empirical formula for calculating the air supply volume required for pollutant emission. Chang et al. [5] used numerical simulation method to derive the spatial distribution function of pollutants in forced ventilation tunnel. On this basis, the estimation function of tunnel safety ventilation time is deduced. Liu et al. [6] made an analysis of the effects of pressure airflow rate and fan’s exhaust airflow rate on the diffusion and pollution behaviors of dust. They further improved the ventilation performance of the tunnel through numerical simulation technology and field measurements. Li et al. [7] developed a one-dimensional steady-state pollutant mass transfer model to determine the ventilation control. The model can quantitatively evaluate the distribution of the CO pollutant concentration and the ventilation system control effect.

Computer numerical simulation is widely used in the research of tunnel safety. Computational fluid dynamics (CFD) can provide effective solutions to many complex flow and heat transfer problems in modern science [8,9,10,11,12]. Zhang, Sun, Lin, and Wang [13] evaluated the optimal position of the fan in the tunnel through CFD method. Their results achieved rapid emission of pollutants. Nie et al. [14] studied the influence of air volume on dust diffusion during tunnel excavation process through CFD software. Their conclusions provide an important reference for the protection of workers’ health and the optimization of tunnel ventilation systems. Huang et al. [15] combined CFD methods with a monodimensional temperature prediction model to study the the performance of Saccardo Nozzle in tunnel. Their design is considered to be a feasible solution to be adopted in tunnels. Zhang et al. [16] applied numerical model to study the influence of rough elements on the resistance coefficient along the tunnel wall. They obtained a correction formula for calculating the resistance coefficient and the rationality was verified by a field test. Liu et al. [17] studied the dust control effect of a construction tunnel under the different parameters of ventilation system. The results show the optimal range obtained regarding the distance between the secondary air pressure inlet and the tunnel face was 30–40 m. Zhao et al. [18] used CFD method to verify that the jet fan with a pitch angle can significantly alleviate vortices caused by the jet flow. Liu et al. [19] developed a three-dimensional computational model to study the carbon dioxide concentration and thermal environment of the subway tunnel. They proposed that the optimal train speed can minimize the volume of mechanical supply air and optimize the carbon dioxide concentration and thermal environment of the tunnel. Besides, the impacts of the main vent location, the air baffle length and the exhausting air flow quantity on flow field distribution and dust flow behavior were analysed through simulation models using CFD software [20]. Moreno et al. [21] performed experiments in the Barcelona Metro system to understand the relationship between ventilation and air quality. Their research proved that the combination of fan power and forced airflow direction has a key impact on air quality.

Therefore, it is crucial to study the diffusion law of pollutants in tunnel ventilation during construction. Mastering the distribution pattern of pollutants helps to ensure the safety of workers and the surrounding environment. Understanding the spread of pollutants in tunnels can guide the design and management of ventilation systems, reduce the risk of workers being exposed to harmful substances, and maximize the protection of the health and surrounding environment.

Obviously, the application of numerical simulation methods to solve ventilation problems has been widely verified. The discharge effect of pollutants has become the top priority of ventilation engineering. The survey of the tunnel construction site show that the ventilation problem has become the bottleneck restricting the rapid progress of the project. The main reason is that the supply of effective fresh air at the working face is insufficient during tunneling. The causes of this phenomenon are as follows: (1) unreasonable ventilation system layout and ventilation scheme; (2) Improper selection of ventilation mode. In order to make the ventilation system and scheme safer and the selection of ventilation modes more reasonable, it is crucial to understand the pollutant emission mechanism of tunnels under different ventilation modes. In this paper, a series of numerical simulations were carried out to determine the airflow field and pollutant distribution in the tunnel under different ventilation modes. Through numerical calculation, the advantages and disadvantages of different ventilation modes can be simulated and demonstrated. The results of this study can fill the gaps in the existing research on pollutant distribution and provide reference for ventilation design during tunnel construction.

2 Numerical method

This study takes the Lianghekou Hydropower Station as the research object. The Lianghekou tunnel is located in the Yalong River Valley of China. The lithology of the tunnel is silty shale slate and metamorphic sandstone. The potential for the release of pollutants such as CH4 is low due to low organic content in tunnel. Therefore, relatively stable CO gas is used to represent the pollutants in the tunnel [22]. China’s ‘Design Specification for Railway Tunnels’ (2016) [23] stipulates that the CO concentration limit at the working face is 30 mg/m3. When construction workers need to enter the tunnel for inspection, the concentration can be as high as 100 mg/m3.In the calculation of the model in this study, the following assumptions are made [24]: (1) the internal friction is not considered in the airflow; (2) The fluid is incompressible; (3) The flow of pollutant is carried out at a constant temperature; (4) The air velocity in the duct outlet is evenly distributed; (5) In the initial state, the pollutant is evenly distributed in the range of smoke; (6) The interference of equipment such as lining formwork trolley is ignored.

The finite element model is established by ADINA software. The CFD module in the finite element software ADINA has the characteristic of reasonable and efficient solution for turbulence equations. It can model, solve, and analyze the results of large-scale projects through a visual interface. The turbulence simulation method of the flow field adopts the Renormalization Group - turbulence model. The turbulence model is an airflow model based on the Navier-Stokes equations. Air velocity is determined by solving the continuity, momentum, and turbulence model equations using the RNG k-ε turbulence model. And then the CO concentrations are obtained by introducing air velocity into the CO transport equations.

In this paper, the following boundary conditions are adopted.

  1. (1)

    Inlet boundary:

    It is assumed that the velocity distribution, turbulent kinetic energy and energy ratio at the duct outlet are uniformly distributed. The air velocity uin in the X direction is determined by the cross-sectional area A of the duct and the ventilation volume Q. The velocities in the Y direction and the Z direction are both 0. The X direction is the longitudinal direction of the tunnel, which is the direction of airflow flow. The Y direction is the cross-sectional direction of the tunnel. The Z direction is perpendicular to the ground.

  2. (2)

    Outlet boundary:

    The outlet boundary is the tunnel outlet. Uniform flow boundary condition is adopted. The gradient of each variable at the exit in the YZ plane is taken as 0. The relative pressure at the tunnel outlet is set to 0.

  3. (3)

    Wall boundary:

    The side wall of the tunnel and the working face are both adhere to the law of fixed walls. All the nodes located in the fixed wall use the no-slip condition.

  4. (4)

    Initial conditions:

    The initial CO concentration in the tunnel is calculated by the formula c = Gb/lA. Where c is the initial concentration of the pollutants in the tunnel, in the unit of mg/m3; G is the amount of blasting explosives, in the unit of kg; b is the volume of the produced toxic gas per kilogram of explosive (0.04 m3/kg in this study); l is the initial range of the smoke, in the unit of m; and A is the cross-sectional area of the tunnel, in the unit of m2.

In this paper, the tunnel model parameter settings under different ventilation modes are shown in Fig. 1.

Fig. 1
figure 1

Schematic diagram of tunnel model and parameter setting

Full size image

Figure 1 shows the longitudinal layout of the horizontal section of the tunnel under different ventilation methods. Under the forced ventilation and extracted ventilation modes, the air ducts are arranged on the side positions of the tunnel as shown in Fig. 1 (a)-(b). The fan is located on the outside of the tunnel exit. For the other ventilation modes as shown in Fig. 1 (c)-(f), both sides of the tunnel are equipped with air ducts and fans for joint ventilation.

3 Grid characteristics and model validation

3.1 Grid characteristics

Grid independence is an important criterion to evaluate the accuracy of numerical simulation. In this study, tunnel models with 5 different grid densities in the longitudinal and transverse directions are established. The length of each model is 100 m, and the distance between the duct outlet and the working face is 35 m. Forced ventilation is adopted for the system. Tunnel models are shown in Fig. 2. The measuring point P at the same location is selected as the research object. The air velocity distribution at point P along tunnel is shown in Fig. 2. A1-A5 is a grid density classification in the longitudinal direction of the tunnel. The grid density from A1 to A5 gradually decreases. Similarly, B1-B5 is a grid density classification in the cross-sectional direction of the tunnel.

Fig. 2
figure 2

Tunnel models used for the grid-sensitivity analysis

Full size image

According to the calculation results, the velocity distribution law of the point P is similar. With the increase of distance from the working face, the air velocity gradually tends to a stable value. The model error is measured by the result of variance calculation. The variance of the other models is calculated based on the calculation results of the models A1 and B1 with the smallest grid density. The error analysis of calculation results is shown in Table 1.

Table 1 Calculation results of tunnel models
Full size table

The increase of grid density will raise the running time of the model. The density selection needs to consider both the calculation accuracy and efficiency. According to the model calculation results, the difference between the tunnel outlet and the duct outlet (40–100 m) is slight. The relatively large difference in calculation results is in the area near the working face (0–40 m). With the increase of tunneling distance, the influence of calculation error caused by the difference near the working face will gradually decrease. Therefore, the grid density of A2 and B1 is used for the calculation of the area near the working face. The grid density of models A2 and B2 is selected for the other areas of the tunnel.

3.2 Model validation

The verification of the model method is based on the measured data of Micangshan tunnel in China. In this paper, a 1:1 tunnel model is established and compared with the CO concentration data measured by SUMMIT-706 detector in the project. The test section is 100 m away from the working face. The measuring point is located 1.7 m away from the ground at the center line of the tunnel. The measured air velocity is 0.65 m/s. The model air velocity is 0.652 m/s. The data error is 0.3%. The comparison of CO concentration between the simulated value and the measured value is shown in Fig. 3.

Fig. 3
figure 3

Correlation results of simulated and measured values

Full size image

It can be seen from Fig. 3 that the numerical simulation results are in good agreement with the measured data. At the Sect. 100 m away from the working face, the simulated data maintains a high degree of consistency with the measured data. The reason for the error may be that the CO gas generated by mechanical equipment and the flow field changes caused by mechanical equipment and human factors are ignored in the model.

4 Results and discussion

This study conducts numerical simulations on a 500 m long construction tunnel using different ventilation modes. By observing the flow field distribution and pollutant discharge process in tunnels under different ventilation modes, a general pattern can be obtained. Based on the model calculation, the flow field changes in the tunnel under different ventilation modes are obtained. (1) The model calculation shows that after 120s of ventilation, the flow field in each area of the tunnel no longer changes under forced ventilation. This indicates that the flow field in the 500 m long tunnel is stable after 120s of ventilation. And the closer the area to the working face, the earlier the flow field reaches stability. The final flow field distribution is characterized by channel flow, i.e. the large intermediate flow rate gradually decreases as it approaches the tunnel wall. From Fig. 4(a), the inflection point where the airflow structure tends to stabilize is basically near the duct outlet. The analysis shows that the airflow velocity gradually decreases as fresh air flows from the duct outlet to the working face. Air flows out from the other side of the tunnel (near AII) after hitting the wall of the working face. When airflow returns near the duct outlet, part of the air continues to flow to the tunnel outlet and the other part returns to the working face under the influence of jet air from the duct. This phenomenon occurs continuously as the air flow into the tunnel. Therefore, an vortex area near the working face is formed. (2) For extracted ventilation, the flow field in the 500 m long tunnel is stable after 90 s of ventilation. According to Fig. 4(b), the flow field distribution in the tunnel is similar to that in the tunnel under forced ventilation, except that the direction of air velocity is opposite. Fresh air enters the tunnel from the tunnel outlet and flows to the working face. The dirty air near the working face enters the duct. The airflow distributes a small part of the vortex area near the working face. (3) The extracted ventilation combined with jet fan is widely used in tunnel ventilation system. The addition of jet fan accelerates the airflow near the working face. As shown in Fig. 4(c), compared with the extracted ventilation mode, the air speed near the duct (AIII) is significantly increased. The local air velocity near the working face increases, thus expanding the distribution area of the vortex. (4) Under the far-pressing-near-absorption ventilation mode, the air velocity at each position of the tunnel is significantly lower than that of the forced ventilation (Fig. 4(d)). This is because part of the airflow is extracted by the duct, resulting in a decrease in the air volume entering the tunnel. The fresh air flows from the outlet of the forced duct to the working face. After the air hits the wall of the working face, it flows out from the side of the extracted duct. When the air meets the airflow from the forced duct, part of the air returns to the working face under the influence of the jet air. The process continues to occur to form the vortex zone. (5) The front-pressing-behind-absorption ventilation is mainly of the extracted mode. As shown in Fig. 4(e), the addition of forced ventilation leads to an increase in the distribution area of vortex in the tunnel. The air velocity at the side of the tunnel near the duct (AII) is significantly lower than that at other positions. (6) As shown in Fig. 4(f), the area distribution of vortex in the tunnel is wider due to the use of front- absorption-behind-pressing ventilation. Vortex is formed near the working face at the center of the tunnel.

Fig. 4
figure 4

Air velocity distribution of different ventilation modes

Full size image

As shown in Fig. 5, the peak value of the CO concentration curve shifts from left to right. During the movement, the span of the curve gradually increases and the peak value gradually decreases. This indicates that the distribution area of pollutant gradually moves from the vicinity of the working face to the tunnel outlet. The distribution area of pollutant in the moving process gradually expands. The CO concentration value near the duct (near AII) is relatively high. According to the comparison Fig. 4 that this is due to the low air velocity on this side. In addition, it can be clearly seen that the distribution of CO concentration in the tunnel presents two peaks. The position of the peak I near the working face does not move with the increase of the ventilation time. But its value will gradually decrease with continuous ventilation. From Fig. 4, it can be seen that the airflow field near the working face is significantly distributed with vortex. The center of the vortex is approximately 15 m away from the working face. The peak I is formed because the pollutant accumulates at the center of the vortex and is difficult to discharge. The other CO peak II gradually moves toward the tunnel outlet with the increase of ventilation time. The discharge process of CO gas can be divided into movement and diffusion. The movement moves the pollutant from the working face to the tunnel outlet. Diffusion gradually expands the distribution range of pollutant in the process of movement, and gradually dilute the concentration. Therefore, the discharge of pollutant has both the migration function of main air flow and the diffusion function of air turbulence.

Fig. 5
figure 5

Distribution of CO concentration in the tunnel under forced ventilation

Full size image

According to the Fig. 6, the migration process of pollutant under Far-pressing-near-absorption ventilation is similar to that under forced ventilation. The position of peak I remains unchanged. And the value decreases gradually with continuous ventilation. The value of peak II gradually moves towards the tunnel outlet. In addition, the CO peak at the side of the forced duct (AII) and the top of the tunnel (AIV) moves slowly. And the reduction speed is relatively slow. The analysis shows that the slow diffusion and discharge of pollutants are caused by the low air velocity. Therefore, in the presence of a pressure jet airflow, the discharge of pollutants inside the tunnel is divided into movement and diffusion processes. This pattern is consistent with the mass transfer process of turbulent airflow. The discharge of harmful gases has both the transport effect of the main airflow and the diffusion effect of the turbulent airflow.

In order to increase the air velocity in the tunnel, the Far-pressing-near-absorption ventilation is optimized. The optimized layout is shown in Fig. 7(a). By shortening the length of the extracted ventilation duct, the airflow flow inside the tunnel has been increased. Comparing Fig. 7 (a) and Fig. 5 (d), it can be seen that the airflow velocity inside the tunnel has significantly increased. The increase in flow velocity accelerates the rate of pollutant discharge. Comparing Fig. 6 (c) and Fig. 8 (b), under the traditional far-pressing-near-absorption ventilation mode, the highest CO concentration in the tunnel is 300 mg/m3 when ventilated for 900s. The highest CO concentration in the tunnel under the optimized far-pressing-near-absorption ventilation mode is only 100 mg/m3 at 900 s of ventilation. The optimized layout can not only accelerate the discharge of pollutants near the working face, but also increase the migration of pollutants in the tunnel. The distribution of air velocity and CO concentration is shown in Fig. 7. The air velocity in the tunnel increases obviously from the fan outlet section. The outlet jet of the fan accelerates the airflow movement. But there is an obvious peak of CO concentration at this position. However, compared with the Far-pressing-near-absorption ventilation, the optimized arrangement saves the ventilation time and improves the ventilation efficiency.

Fig. 6
figure 6

Distribution of CO concentration in the tunnel under Far-pressing-near-absorption ventilation

Full size image
Fig. 7
figure 7

Distribution of air velocity and CO concentration under optimized Far-pressing-near-absorption ventilation

Full size image

Figure 8 shows the distribution of CO concentration in the tunnel with extracted ventilation as the main method. For tunnels with such ventilation, pollutants do not flow through the whole tunnel, but are only gradually diluted and discharged in the distribution area. It can be seen from the figure that the time required for the CO gas to reach the safe concentration standard (30mg/m3) is different. (1) The tunnel area (AIand AIII) between the duct outlet and the working face forms a stagnant area of pollutants (as shown in Fig. 8 (a)). This region is also the location where vortex is generated. This shows that the vortex blocks the discharge of pollutants. A stagnant area of pollutants under extracted ventilation is formed In the center of the vortex. (2) The addition of jet fan can obviously reduce the ventilation time required for the pollutants transport under the extracted ventilation mode(as shown in Fig. 8 (b)). There are two different peaks in the CO concentration distribution. One is located near the working face (x = 4 m); The other is located at about x = 12.5 m, i.e. near the duct outlet. The two CO peaks are caused by the retardation of vortex and low air velocity. (3) Under Front-pressing-behind-absorption ventilation, the emission rate of pollutants inside the tunnel is relatively fast. The concentration of pollutants in the upper area of the tunnel is large. It indicates that the low air velocity in this area makes it difficult for pollutants to be discharged and diluted. (4) Under Front-absorption-behind-pressing ventilation, the distribution area of high concentration pollutants is different in different positions. In the range of x = 0 m to x = 15 m, the CO concentration in the central area of the tunnel (AI,AIV, and AV) is high due to vortex block. Between x = 15 m and x = 35 m, the CO concentration (AIII) near the extracted duct is high. It indicates that the vortex is located in the area near the extracted duct at this position.

Fig. 8
figure 8

Distribution of CO concentration in the tunnel under other different ventilation modes

Full size image

In the process of discharging the pollutants in the tunnel, the long migration path leads to the increase of ventilation time. According to the model calculation results of the above ventilation modes, it is found that the migration process of pollutants not only delays the ventilation time, but also pollutes the working environment. On this basis, this paper presents a new type of annular ventilation. The system layout is shown in Fig. 9. Annular ventilation system is a ventilation method that combines longitudinal ventilation with transverse ventilation. While the airflow flows along the longitudinal direction of the tunnel, the airflow near the working face flows along the cross-section of the tunnel by changing the airflow direction of the inlet and outlet air ducts. The pollutants in this ventilation system are transported along the direction of the tunnel cross section. Theoretically, it can shorten the migration path of pollutants and save the ventilation time. It can also reduce the distribution range of pollutants.

According to the simulation results, the pollutants do not diffuse to the area outside the duct outlet under the annular ventilation mode. There is no obvious peak of CO concentration in the area between the working face and the duct outlet. It shows that the vortex in the cross-section direction of the tunnel has no great blocking effect on the transport of pollutants. As shown in Fig. 9(b), the value of black curve (AI, tunnel bottom) is low along the tunnel. It indicates that the pollutants at the bottom of the tunnel migrate faster. In the area near the duct outlet, the CO concentration value has a small peak. It indicates that a vortex is formed, which is not conducive to the discharge of pollutants. The CO concentration at all positions in the tunnel is reduced below the safe concentration after 5 min of ventilation. Compared with other ventilation modes, the annular ventilation mode significantly shortens the time.

Fig. 9
figure 9

Distribution of air velocity and CO concentration under annular ventilation

Full size image

4.1 Comparison of different ventilation methods

In order to compare the advantages and disadvantages of each ventilation mode, the tunnel model results of different ventilation modes are sorted out. In this study, the area between the duct outlet and the working face is defined as the control body VI [25]. The study on the migration of pollutants in the control body facilitates the exploration of the working environment.The comparison results of different modes are shown in Table 2.

Table 2 Comparison of calculation results of tunnel models with different ventilation modes
Full size table

For the ventilation tunnel with small cross-section and long distance, it is recommended to adopt the ventilation mode of extracting the pollutants without flowing through other areas of the tunnel. Among them, the extracted ventilation with jet fan type, the front-absorption-behind-pressing ventilation and the annular ventilation mode can discharge and dilute pollutants faster. The difference between the entry time and the safety time is the smallest under the annular ventilation mode. This shows that the annular ventilation mode has a stronger ability to dilute pollutants. For the ventilation mode that pollutants need to be discharged through the tunnel channel, the forced ventilation time is shorter. Far-pressing-near-absorption ventilation can reduce the concentration of pollutants in the tunnel channel, but it requires longer ventilation time. The movement velocity of the CO peak of the optimized far-pressing-near-absorption ventilation is faster. It indicates that the optimized far-pressing-near-absorption ventilation will save more ventilation time when the tunnel excavation length is longer.

Figure 10 shows different times is to compare the distribution of residual CO gas in the tunnel when the tunnel reaches the safe concentration standard under different modes. The CO concentration of tunnel that allows workers to enter the tunnel face for construction directly affects the health of workers. The distribution of residual CO gas should be given sufficient attention. In order to study the construction environment near the working face, a horizontal plane with the height of the breathing zone z = 1.7 m within 30 m around the working face is taken for study. When the tunnel reaches the safe concentration standard, the pollutant distribution near the working face under different ventilation modes is shown in Fig. 10.

For the first five ventilation modes (as shown in Fig. 10 (a) - (e), pollutants are mainly discharged out of the tunnel through the extracted duct. The ventilation time required for the tunnel is relatively short. However, the highest concentration area in the whole tunnel is concentrated near the working face. The advantage of annular ventilation is that the pollutants are evenly distributed. There is no obvious CO peak at the horizontal plane. Comparing the ventilation modes of extracted ventilation and jet fan, it is found that the addition of jet fan not only reduces the ventilation time, but also reduces the high concentration distribution area of pollutants to a great extent. Under the front-pressing-behind-absorption ventilation mode (as shown in Fig. 10 (c)), the high concentration area is distributed in the tunnel centerline area. The forced airflow and the extracted airflow effectively reduce the pollutant concentration in the corresponding area. Under the front-absorption-behind-pressing ventilation mode (as shown in Fig. 10 (d)), the high concentration pollutant area is distributed near the working face at the end of the forced airflow. This shows that the distribution area of high concentration pollutants can be reduced by appropriately increasing the forced air volume of the fan in this mode.

For the latter three ventilation modes (as shown in Fig. 10 (f) - (h)), although it takes a long time for ventilation to reach the safe concentration standard, the construction environment near working face is relatively cleaner. The highest concentration of pollutants in the tunnel is distributed near the tunnel exit.

Fig. 10
figure 10

The distribution of CO concentration in the horizontal plane near the working face with the height of the breathing zone Z = 1.7 m

Full size image

By comparing Table 2, the time cost of annular ventilation can be obtained. Under the annular ventilation method, the time efficiency savings compared to forced ventilation, far-pressing-near-absorption ventilation, and optimized far-pressing-near-absorption ventilation are 56.5%, 85.7%, and 80%. From Fig. 10, it can also be seen that compared with the extracted ventilation, extracted ventilation with jet fan, front-pressing-behind-absorption ventilation, and front-absorption-behind-pressing ventilation, the annular ventilation method significantly improves the problem of pollutant retention near the working face.

5 Conclusion

With the development of tunnel construction, the ventilation problem during the construction period has become the key to restrict the rapid construction of the project. The pollutants in the tunnel have a very important impact on the health of workers and the safe operation of equipment. In this paper, the characteristics of the airflow field and the transport law of pollutants in the ventilation tunnel during the construction of the single head tunnel are studied through the simulation model of ventilation system. The main conclusions are as follows:

  1. (1)

    Under different ventilation modes, the velocity distribution in the tunnel conforms to the distribution law of flow in the duct. Due to the influence of jet and wall constraints, vortex zones of different sizes are formed in the tunnel. The vortex zone leads to a sharp decrease of the air velocity in the local area.

  2. (2)

    There are two obvious peaks in the distribution curve of pollutants under the modes of forced ventilation and far-pressing-near-absorption ventilation. The peak near the working face is formed by the retardation of vortex. Its peak value decreases gradually with the increase of ventilation time. The other peak value gradually moves to the tunnel exit. Its migration law can be summarized as movement and diffusion. On this basis, this paper puts forward the optimization method of far-pressing-near-absorption ventilation.

  3. (3)

    Under the ventilation mode dominated by extracted type, the pollutants are only distributed in the area between the duct outlet and the working face. Due to the non-uniformity of air velocity and the retarding effect of vortex, the stagnant area of pollutants is formed near the working face under different modes. In this paper, a new annular ventilation mode is proposed. The simulation results show that the path of pollutants is shorter and the emission is smoother under this ventilation mode.

  4. (4)

    It can be seen from the comparison that the ventilation efficiency of the front-pressing-behind-absorption ventilation ventilation mode is highest. Annular ventilation takes the second place. The efficiency of annular ventilation increases with the increase of tunneling distance of small section tunnel. Under the ventilation mode dominated by forced type, the peak moving velocity of the optimized far-pressing-near-absorption pollutants is the fastest. Its ventilation efficiency is more obvious with the increase of tunneling distance.

The advantages and disadvantages of several ventilation methods commonly used in engineering are verified by numerical simulation. Reasonable parameter setting can reduce the time of pollutant transport in the tunnel. It is of great theoretical significance to master the pollutant transport mechanism under different ventilation modes for parameter setting.