Impacts of opening baffle of city road tunnels on natural ventilation performance

Abstract

Based on the opening baffle mode for natural ventilation of city road tunnels, this paper studies the impacts of opening baffle on natural ventilation performance by verifying numerical simulation through model tests. By analyzing the impacts of installation angle, dimension, location, and quantity of opening baffle on ventilation performance, the paper reached the conclusions as follows: 1) When installation angle is larger than 45° and tunnel ventilation is well operated, the baffle exhaust could increase by at least 30% compared to when there is no baffle. 2) The baffle reaches its optimal performance when the length of the baffle is equal to the width of the city road tunnels. 3) Baffle exhaust could increase by 30% when it is installed in the downstream of openings. 4) The performance of a single baffle is better than that of multiple baffles.

Keywords

city road tunnel / numerical simulation / baffle / top opening / natural ventilation

Introduction

The construction scale of tunnels is growing with the improvement of urban transportation system in China. Vertical ventilation is the mainstream for tunnel ventilation. However, it results in excessive pollutions at the tunnel exits when used in long city tunnels. Therefore, natural ventilation with opening is utilized in city tunnels to solve this pollution problem. However, there are many problems of the natural ventilation with opening, such as the larger area of the opening, the higher construction engineering costs and so on. The previous studies are in terms of theoretical calculation of the flow, model experiment theory and smoke control of fire.

Yoon from Inha University of Korea used thermodynamic method to calculate the natural ventilation pressures in a large tunnel in Korea that has opening baffles for both winter and summer seasons. He concluded that the baffle height and the temperature difference between inside and outside of tunnels are the major factors that influence the natural ventilation pressure [1]. Sambolek of Croatia verified the marginal Reynolds value by conducting simulations and field measurements. He also researched the wind speed and volume created by a car model of different speeds during transportation under the condition of natural ventilation where inflow is the result of the pressure difference between inlet and outlet [2]. Medic of the United States wrote a program to calculate the appropriate length of tunnels based on transportation volume and basic parameters of tunnels. Kim of Korea analyzed the impact of baffle location on the ventilation capacity and fluid fields inside the subway tunnels, also by simulations and field measurements [3].

Shanghai University of Science and Technology [4] also conducted laboratory experiments and simulations to study the quantity, distance and sectional shape of the opening baffles. The results show that the quantity of openings has little impact on the total inflow under the condition of same opening area. The volume of exhaust increases with the quantity of openings. Zhong et al. used a one-dimension constant flow model to analyze the formation and function of pressure at the openings, and designed a calculation method for this format of natural ventilation [5]. Li analyzed the advantages of opening baffle natural ventilation for city road tunnels [6]. Zhu et al. studied a similar situation of opening baffle application, and constructed a similar mathematical model for velocity field of shaft natural ventilation [7]. He analyzed the principles and flow features of shaft natural ventilation for city road tunnels to and Shi build a statistical model to study the distribution of VOCs concentrations inside the tunnels [8]. Tong and Shi used dimension analysis method to obtain the similarity principle of shaft natural ventilation for city road tunnels, and tested its accuracy through the scaling model experiment [9].

There is a lot of research showing baffle could improve the situation of uneven distribution of fluid field. Xie and Zhu utilized Computer Fluid Dynamics (CFD) program to simulate the two-phase flow (gas-solid) inside a 90° square-section curved duct, and found that baffle did improve the distribution of fluids, and also provided an effective solution for the abrasion of tunnels and accumulation of dirt [10]. Wen et al. combined the design features of vertical ventilation for extremely long road tunnels, and conducted simulations and calculations for flows at duct corners to determine the optimal format for ventilation duct corner construction [11].

How to enhance the exhaust effect of the opening with respect to the ventilation performance is rare discussed all over the world. This study may raise the economy of the ventilation mode. Opening baffle ventilation could reduce the pollutions released along the tunnels, and also the total pollutions at the exits of tunnels. To enhance the performance of exhaust at the exists, this paper used both simulations and experiments to study the installation angle, dimension, location, and quantity of opening baffle, and proposed an optimization design for opening baffle construction. Based on the study of natural ventilation with opening, this paper may provide a reference to the city road tunnel ventilation.

Verification of mathematical simulation

Modeling system

The experiment platform was constructed with a rate of 1∶12 based on a city road tunnel in Tianjin. On the base of the tunnel model, one side and the top surface of the tunnel made from 1 mm thick galvanized steel sheet and the other side made from 3 mm thick Plexiglas are attached to enable immediate location and removal of potential mechanical trouble. The roughness of the galvanized steel sheet is close to the actual tunnels. And Plexiglas is chosen for the tunnel model in order to enable visual observation during the experiments. The tunnel has a dimension of 27.4 m × 0.37 m × 0.5 m, and has 6 openings, each one of which has a dimension of 0.17 m × 0.17 m. Cars were assumed to be sedans with a uniform dimension of 0.34 m × 0.17 m × 0.12 m. The traffic flow was assumed to be 110 per hour. No jet fan was considered. This is shown in Fig. 1. The velocity of the opening is measurement by multi-anemometer.

Construction of the mathematical model

Based on the modeling system, a physical model for the tunnel could be constructed, and the simulation of ventilation could be conducted under the condition where opening 1, 2, and 3 are operating. The dynamic mesh model technique and sliding interface were used to simulate the motion of cars inside the tunnel [12–13].

Mathematical simulation assumed that the fluids inside the tunnel were incompressible, regular Newton fluids. The variations of specific heat capacity of air with internal tunnel temperature were neglected, and so was the impact of temperature on the internal fluid field.

The governing equations for car-induced unsteady tunnel flow are the continuity equation, Reynolds-averaged Navier-Stokes equations, and the RNG k-ϵ turbulence model. The general format of its control function is shown in Eq. (1) [12].

(1){\displaystyle \frac{\partial \left(\rho \varphi \right)}{\partial t}}+\mathrm{div}\left(\rho u\varphi \right)=\mathrm{div}\left(\mathrm{\Gamma }\mathrm{grad}\varphi \right)+S,

where, \rhois the fluid density; \varphiis the general variable; t is time; uis the velocity; \mathrm{\Gamma }is the diffusion constant; and S is the source constant.

The governing equations are discretization on finite-volumes. The PISO algorithm is applied to solve the pressure-velocity coupling. The QUIK scheme is employed in the discretization for momentum, turbulent kinetic energy and turbulent dissipation rate equations. As to the pressure corrective equation, the PRESTO scheme is used, which provides improved pressure interpolation in situations where the large pressure gradients exist. For unsteady analysis, the time derivatives are discretized using the first-order implicit scheme.

For modeling the moving boundaries of the car, the dynamic mesh model is adopted. Considering the characteristics of cars motion in a city road tunnel, the dynamic layering method is employed in this study. With the dynamic layering method layers of cells adjacent to a moving boundary can be created or removed based on the size of the layer adjacent to the moving surface in association with the motion of a movable bloke.

The boundary conditions set are as the followings:

1) The pressure-inlet boundary condition is applied at the tunnel inlet, P = 0 Pa.

2) The pressure-outlet boundary conditions are assumed at the tunnel outlet and at the outlet of each ventilation opening, P = 0 Pa.

3) The speed of the cars is 20 km/h.

4) The sliding interface is implemented at the region that dynamic mesh and static mesh junction.

5) The convergence criterion is 10^-6 for all the variables.

Verification of mathematical simulation

A comparison of experimental data and simulated data for average velocity at tunnel openings is shown in Fig. 2. The two sets of data have similar trends, which verified that the mathematical model built for this paper is accurate enough for a further study.

Impact of opening baffle on the performance of ventilation

The concentration of pollutants in the tunnel is closely related to the ventilation of the opening. Both airflow intake and exhaust will dilute the pollutants. The exhaust can distribute the concentration of pollutants, and takeout the pollutants. Though this paper evaluate the performance of the natural ventilation based on the exhaust rate (the exhaust rate is assumed to be K_{\mathrm{x}}=Q_{\mathrm{x}}/Q_{\mathrm{o}}, in which Q_{\mathrm{o}} is the exhaust when there is no baffle at a cycle time and Q_{\mathrm{x}}is the exhaust when there is a baffle at a cycle time).

Opening baffle could improve the uneven distribution of fluids. Setting up a baffle at the bottom of the openings could change the internal resistance of tunnels and therefore influence the ventilation volume. The model was simplified to reduce the required calculations: tunnel is assumed to be single lane, with a length of 300 m and three openings. Each opening has a dimension of 2 m × 2 m, as is shown in Fig. 3. Cars are assumed to be small-size sedan, with a dimension of 4.3 m × 1.8 m × 1.6 m. Car volume is 1000 per hour, and car speed is 50 km/h. The distance between two cars is assumed to be 53 m. This paper is targeted at opening 2 to study the impact of baffle installation angle, dimension, location and quantity on ventilation performance.

Impact of installation angle on ventilation performance

The length of the baffles is the same with that of the openings. The width of the baffles is adjusted based on the installation angle to accommodate the actual tunnel height limit: the bottom of baffle needs to be 0.8 m away from the top of the tunnel (shown in Fig. 4). The dimension of baffle is shown in Table 1. The contours of static pressure and velocity vectors colored by velocity magnitude of 45° baffles are shown in Figs. 5 and 6. The simulation results are shown in Fig. 7.

Figures 5 and 6 show that the baffles can prevent flow direction along the tunnel, but exhaust from the opening. Figure 7 indicates that increasing the number of baffle does not impact the total ventilation volume in tunnel. When the installation angle of baffle is 15°, the ventilation rate is reduced, which would result in a rise in the concentration of pollutants. When the installation angle is 30°, 45°, 60°, 75° and 90°, exhaust ratio at openings is larger than 1. When installation angle is 45°, 60°, 75° and 90°, the ventilation volume inside the tunnel is slightly larger than that when there is no baffle, the exhaust volume at openings are about the same, increased by approximately 30%.

As the increase of installation angle, the exhaust ratio increases first then decreases. This shows that the installation angle of the baffle has a great influence on its performance. While the installation angle is 15°, the area of horizontal contact between airflow and baffle is largest, but the effective area of the opening decreases, so the performance of the baffles is poor. As the installation angle is 30°, 45°, 75° and 90°, the baffles benefit the airflow exhausting.

In sum, when the installation angle of baffle 45°, 60°, 75° and 90°, eliminating the pollutants at openings could be greatly enhanced without changing the total ventilation volume in the tunnel.

Impact of baffle dimension on ventilation performance

This paper further studied the impact of baffle dimension on the ventilation performance. The width of baffle is set to be 0.8 m, and length of baffle includes 2.0, 2.4, 3.6, and 4.4 m. The simulation result shows that the exhaust volume at openings increases with the increment of baffle length. The exhaust ratios (K_x) at baffle length 2.0, 2.8, 3.6, and 4.4 m are 1.10, 1.74, 2.10, and 2.38, respectively. The ventilation performance reaches its optimum when the baffle length equals the tunnel width, and it does not affect the ventilation volume in the tunnel.

Impact of baffle location on ventilation performance

As indicated above, the performance of ventilation inside the tunnel and the exhaust at openings are optimal when the installation angle of baffle is larger than 45°. Therefore, this paper will only study the impact of baffle location on ventilation performance at an installation angle of 45°. The installation of baffle is shown in Fig. 8. The distance between baffles and the upstream of openings is shown in Table 2. The simulation results are shown in Fig. 9.

Figure 9 indicates that when the distance between baffle and the upstream of opening is 0, the back of the baffle will have a partial negative pressure, and therefore, the external air will flow into the tunnel. The ventilation volume in the tunnel is the greatest under this condition, but it does not help distribute the pollutants. When the distance between baffle and the upstream of opening is between 15 to 138 m, the ventilation performance is not as good as when there is no baffle, but the ventilation rate is increased. When the distance is between 138 to 200 m, the exhaust volume is proportionally increased, the ventilation performance is improved, and ventilation rate in the tunnel is also increased.

In conclusion, without affecting the ventilation requirements in the tunnel, installing the baffle around 2/3 away from the upstream of the opening is the optimal option for improving the ventilation performance. When baffle is installed in the downstream of openings, ventilation performance is optimal, and has no interruption on the total ventilation volume.

Impact of baffle quantity on the ventilation performance

The above research is focused on a single baffle, and does not consider the impact of baffle quantity on the ventilation performance. This section will combine several baffles based on the simulation results in Section 2.3. The distances are set up respectively as 15.4 and 30.8 cm. Figure 5 shows the baffle location numbers for baffles, and Table 3 indicates the various combination means for baffles.

Figure 10 indicates that increasing the distance between baffles discourages the air diversion at the bottom of the openings. The exhaust volume at the openings is increased with the installation distance. When the distance between baffles is 15.4 cm, and the baffles are installed at 138 cm away from the upstream of openings, the ventilation performance is better than when there is no baffle. When the distance is 30.8 cm, and baffles are installed 123 cm away from the upstream, the ventilation performance is slightly improved.

In a conclusion, when the combination of two baffles is installed 2/3 away from the upstream of openings, the ventilation performance is slightly improved, but still not as good as when there is only a single baffle.

Conclusions

Based on the natural ventilation mode for tunnel openings, this paper verified a mathematical model through simulations and experiments, and used it to study the impacts of baffle installation angle, dimension, location, and quantity on the ventilation performance. The conclusions are as follows:

1) As the baffle installation angle increases, the exhaust volume first increases to an optimum, and then decreases. When the installation angle is larger than 45°, the exhaust effect at openings and the ventilation performance in the tunnel are optimized. The exhaust volume under this condition could be increased by at least 30%.

2) Under the condition that the vertical installation height is maintained the same, the diversion effect of baffle is better when the baffle length equals the tunnel width.

3) Exhaust effect is improved when a 45° baffle is installed 2/3 away from the upstream of openings. Exhaust volume could be increased by 5% to 30%, depending on the installation location. The exhaust effect is optimized when baffle is installed in the downstream of openings.

4) Exhaust performance is slightly improved when the combination of two baffles is installed 2/3 away from the upstream of openings, but it still does not exceed the performance of a single baffle.

References

[1] Yoon C H, Kim M S, Kim J. The evaluation of natural ventilation pressure in Korean long road tunnels with vertical shafts. Tunnelling and Underground Space Technology, 2006, 21(3–4): 472–477

[2] Sambolek M. Model testing of road tunnel ventilation in normal traffic conditions. Engineering Structures, 2004, 26(12): 1705–1711

[3] Kim J Y, Kim K Y. Effects of vent shaft location on the ventilation performance in a subway tunnel. Journal of Wind Engineering and Industrial Aerodynamics, 2009, 97(5–6): 174–179

[4] Huang Y, Gong X, Peng Y, Lin X, Kim C N. Effects of ventilation duct arrangement and duct geometry on ventilation performance in a subway tunnel. Tunnelling and Underground Space Technology, 2011, 26(6): 725–733

[5] Zhong X C, Gong Y F, Tong Y. Similarity study on model experiment for natural ventilation in highway tunnels. Heating Ventilation & Air Conditioning, 2008, 38(5): 13–18 (in Chinese)

[6] Li C X. Advantages of natural ventilation of urban tunnel with opening. China High Technology Enterprises, 2008, 1:€146 (in Chinese)

[7] Zhu P G, Zhu D N , He C H. The analogical theory research on natural ventilation in city tunnel with vertical well. Contamination Control & Air-Conditioning Technology, 2008, 4: 12–15

[8] Zhu P G, Luo C C, Fan L Y, Mao Y F, He C H, Zhu D N. Study on the natural ventilation in the city tunnel with vertical-shaft. Refrigeration Air Conditioning & Electric Power Machinery, 2009, 30(1): 11–14 (in Chinese)

[9] Tong Y, Shi M H. Experimental study of natural ventilation for road tunnels with shafts. Heating Ventilation & Air Conditioning, 2009, 39(9): 61–65 (in Chinese)

[10] Xie Z H, Zhou Y R. Optimization research on guide plate installed in quadrate 90° curved duct. Journal of Basic Science and Engineering, 2009, 17(4): 566–572

[11] Wen Y H, Xie Y L, Li N J. Optimization research on the structure of 90° curved duct and analysis of guide plate state. Guangdong Highway Survey and Design, 2011, 144(1): 20–22 (in Chinese)

[12] Wang F J. Computational Fluid Dynamics: The basics and Applications. Beijing: Tsinghua University Press, 2004 (in Chinese)

[13] Fluent Inc. Fluent 6.1 User’s Guide. NH: Fluent Inc, 2003

RIGHTS & PERMISSIONS

Higher Education Press and Springer-Verlag Berlin Heidelberg