Optimization of Ventilation Performance in Large-Section Highway Tunnels: The Role of Deflector Shields in Jet Fan Systems

Abstract

The jet fan system is a widely adopted form of longitudinal ventilation due to its cost-effectiveness, operational flexibility, and high reliability. However, in large-section highway tunnels with a low height-to-span ratio, the limited clearance between the tunnel ceiling and surrounding structural boundaries imposes significant constraints on improving ventilation performance by adjusting the installation height or pitch angle of the jet fan. To address this limitation, this study proposes a deflector shield system to enhance the aerodynamic efficiency of jet fans. A total of thirteen test cases, including a control group, three deflector plate quantities, and four deflector pitch angles, were tested in a full-scale field test conducted in a large-section tunnel. The objective of this study was to evaluate the influence of the number and pitch angle of deflector plates on tunnel ventilation efficiency and to identify the optimal parameter combination for application in large-section tunnels. The results show that static pressure along the tunnel initially rises with distance from the fan, peaks, and then declines sharply. The pressure rise coefficient is significantly enhanced under several configurations, particularly with four deflector plates at 8° and 10° pitches, and with five plates at 4° to 10° pitches. When the number of deflector plates is five, a sharp drop in average wind speed is observed 15 m downstream of the fan, and extensive low-velocity regions appear further downstream. In contrast, the configurations with four deflector plates at 8° and 10° exhibit better wind speed uniformity in the downstream flow field. Considering both the pressure rise coefficient and wind speed uniformity, the optimal ventilation performance of the jet fan system is achieved with four deflector plates at a pitch angle of 8°.

1. Introduction

With the rapid development of highway transportation in China, the scale of tunnel construction has been expanding [1,2,3,4]. The safety, comfort, and health of highway tunnel operations are receiving more and more attention [5,6]. The interior of the tunnel is a relatively enclosed area. Vehicle emissions tend to be accumulated in tunnels under natural ventilation conditions. Vehicle emissions contain a variety of harmful components such as CO, NO, SO₂, organic compounds, hydrocarbons, and so on [7,8,9]. Pollutants pose significant threats not only to human health but also to tunnel visibility, compromising traffic safety and maintenance operations. Furthermore, under prolonged exposure, these pollutants accelerate the corrosion of tunnel structures, lighting systems, and electrical equipment, substantially reducing their service life. Therefore, ventilation equipment must be installed in tunnels to utilize fresh air to dilute and remove harmful gases, fumes, and dust emitted by vehicles in the tunnels, so as to ensure a relatively safe, healthy, and comfortable environment inside the tunnels [10,11].

The jet fan system is a commonly used longitudinal ventilation system [12,13,14,15,16], and it has been widely applied due to its low cost, high flexibility, and high reliability. Jet ventilation is realized by the induction and pressure rise effect of jet fans on the airflow. The jet is blown out from the jet fan at high speed and collides with the low-speed airflow in the tunnel. Under the action of the jet, the tunnel airflow presents a gradual, non-uniform adverse pressure flow along the longitudinal direction. At the same time, the pressure rises. The combined induced effect and boosting effect of the jet fan drive the airflow to move longitudinally within the tunnel, thereby achieving the purpose of ventilation. Fundamentally, jet ventilation involves the conversion of the jet’s kinetic energy into potential energy. At the outlet section of the fan, the velocity distribution becomes uniform, turbulence intensity stabilizes, and the jet fan group achieves its maximum pressure rise [17,18].

The Coanda effect states that when there is surface friction between a fluid and the surface of an object it flows over, the fluid will flow along the surface of that object [19]. Jet fans are usually installed at the tunnel ceiling. Due to the Coanda effect, the jet will cling to the tunnel ceiling, leading to significant friction losses [20,21]. This ultimately reduces the ventilation efficiency of the fan. In order to improve the ventilation efficiency of jet fan systems, many scholars have carried out related research. Yu et al. [22] conducted an optimization simulation analysis on the installation height of jet fans in road tunnels and the angle between the fan axis and the tunnel axis. Their results indicated that both parameters significantly influence ventilation performance, exhibiting a parabolic trend in their effects. Gao et al. [23] conducted an analysis of the influence of fan arrangement parameters on ventilation efficiency by combining numerical simulations with an orthogonal experimental design. Lu et al. [24] developed a three-dimensional CFD model of longitudinal ventilation in a long highway tunnel beneath Taihu Lake using Fluent and studied the influence of jet fan installation positions on ventilation efficiency. Zhao et al. [25] used CFD simulations to evaluate fan performance under different installation heights—specifically at distances of 15 cm, 30 cm, 45 cm, and 60 cm from the ceiling of the tunnel. The results showed that both the pressure rise generated by the fan and the overall impact coefficient decreased as the distance from the ceiling increased.

A large-section tunnel in China, with a total length of approximately 24 km and a low height-to-span ratio (Figure 1), is equipped with jet fans for ventilation within the submerged portion. According to Chinese regulatory codes [26], the installation height of the fans must not encroach upon the structural clearance envelope, and the minimum allowable distance between the fan edge and the tunnel’s structural clearance limit must be no less than 15 cm. After the fans were installed, testing revealed that the ventilation efficiency was not as effective as expected. Due to the limited space between the tunnel ceiling and the structural clearance boundary, it was not feasible to improve ventilation efficiency by adjusting the installation height and pitch angle of the jet fans [12,14,27]. To avoid discarding the already-installed jet fans, a deflector shield device was proposed to improve ventilation efficiency. The deflector shield contains multiple internal deflector plates, and its airflow guiding performance can be enhanced by adjusting either the pitch angle or the deflector plate quantity.

In this paper, a deflector device was proposed to enhance the ventilation efficiency of jet fans in a large-section tunnel. A total of thirteen test conditions were designed, including a control group, three variations in internal deflector plate quantities, and four variations in deflector pitch angles. The field tests of ventilation were conducted in a large-section highway tunnel equipped with a deflector shield to investigate the effects of deflector plate quantity and pitch angle on ventilation performance. Based on the test results, an optimal deflector shield configuration suitable for jet fans in the large-section tunnel was identified.

2. Field Test

2.1. Engineering Background

The large-section tunnel is a twin-tube, bidirectional structure, with each standard tube having a width of 21.1 m (internal width: 18.0 m) and a height of 6.6 m. The field test was carried out in a standard section of the immersed tube portion of the tunnel. Two jet fans (manufactured by Howden Hua Engineering Co., Ltd., China) are installed symmetrically on either side of the tunnel’s transverse centerline, positioned at distances of 1.5 m and 4.5 m from the centerline, respectively. The layout of the fans is illustrated in Figure 2, and their specifications are summarized in

Table 1. Performance parameters of jet fan.

Equipment	Impeller Diameter (mm)	Outlet Flow (m3/s)	Outlet Flow Rate (m/s)	Axial Thrust (N)	Motor Power (kW)
APR-37 kW	1120	34.5	35.5	1260	37

.

2.2. Test Design

Based on the engineering conditions, 13 test conditions were designed, as detailed in

Table 2. Field test conditions.

Number	Number of Deflector Plates (Piece)	Pitch Angle of Deflector (°)
1	3	0
2	3	4
3	3	6
4	3	8
5	3	10
6	4	4
7	4	6
8	4	8
9	4	10
10	5	4
11	5	6
12	5	8
13	5	10

. These included a control group with three deflector plates set at a pitch angle of 0°, four groups with varying deflector pitch angles (4°, 6°, 8°, and 10°), and three groups with different numbers of deflector plates (3, 4, and 5). All field tests were conducted exclusively during nighttime hours (22:00–04:00), when temperature, humidity, atmospheric pressure, and natural air flows in the tunnel remained relatively stable, thereby minimizing environmental variability in the measurements. Meanwhile, the interval between test groups was kept within 1.5 h to minimize long-term environmental drift.

2.3. Distribution of Measuring Points

The arrangement of measurement points was based on the Log–Tchebycheff method [28]. A total of 30 measurement points were arranged within each test section, with locations detailed in Figure 3.

The layout of the test sections is shown in Figure 4, comprising a total of six sections (A~F). The furthest section is located 30 m downstream from the fan. This distance was determined based on preliminary measurements using an anemometer, which indicated that the wind speed at this location had stabilized to a relatively low and steady value.

2.4. Selection and Arrangement of Test Equipment

The equipment used for the field test included fan deflector plates, Pitot tubes, data collection devices, and thin and thick hoses, among other necessary components.

To minimize test cost and time, and to prevent potential damage to the jet fans from repeated installation and removal, this study adopted four sets of modular deflector shield models with 3, 4, and 5 deflector plates, respectively. These temporary deflector plates were designed to maintain a safe clearance from the jet fan during installation and removal, thereby avoiding physical contact or damage. Moreover, each deflector plate allowed for adjustment of the pitch angle through side-mounted bolts. This design significantly reduced the frequency of disassembly and reassembly during the test process, thereby enhancing testing efficiency. The deflector shield setup is illustrated in Figure 5.

Pitot tubes [29] were used to measure static and total pressure, as shown in Figure 6. Two ports were installed at the base of a Pitot tube, each connected to a thin hose via a thick hose and a plug. To ensure airtight connections, paper tape was repeatedly wrapped around both the Pitot tube and the plugs. The port aligned parallel to the inlet of the Pitot tube was used for measuring static pressure, while the port perpendicular to the inlet was used for measuring total pressure.

The MPS4264 pressure scanner from Scanivalve, along with the accompanying ScanTel software, was used for data acquisition (Figure 7). The MPS4264 was connected to the Pitot tube assembly by a thin flexible tube. The pressure data collected by the scanner was processed using the Scantel software and subsequently visualized on a computer for analysis.

2.5. Testing Process

① According to Case 1, a deflector shield with three deflector plates was installed using a forklift and a lift platform. The deflector plates were adjusted and fixed at a pitch angle of 0°. The installation process and final configuration of the deflector shield are shown in Figure 8.

② Five Pitot tubes were mounted on each of the two steel pipes in accordance with the measurement point layout plan. The measurement point positions were marked using a tape measure and paper tape. The Pitot tubes were then secured to the steel pipes using tape and zip ties, as illustrated in Figure 9.

③ The locations of the measurement points on the ground were accurately marked with a tape measure and spray paint.

④ The data acquisition device was positioned at the test section midpoint, with the steel pipes at opposite ends. This ensured adequate hose length and prevented disconnections at the Pitot tubes or scanner ports during pipe movement. Left-side Pitot tubes measuring static pressure (arranged low to high along the pipe) connected to scanner ports 1–5, while those measuring total pressure connected to ports 11–15. Similarly, the right-side Pitot tubes measuring static pressure (low to high) were connected to ports 6–10, and those measuring total pressure were connected to ports 16–20. The thin hoses were organized and bundled together, with each hose labeled using paper tape according to the wiring scheme. The wiring configuration between the Pitot tube ports and the data acquisition device is illustrated in Figure 10.

⑤ The computer was connected to the data acquisition device, after which the device was powered on, and the analysis software was configured. The two calibration ports were placed inside a sealed box to perform calibration of the ambient air pressure and zero adjustment of the data processing software. Figure 11 illustrates the fully connected and calibrated data acquisition system.

⑥ The fan was started and allowed to reach a stable wind speed. The steel pipe was then positioned vertically at the two leftmost measurement points in Section A. The test software was run for a 3 min measurement period, during which data were recorded synchronously. The test system is shown in Figure 12.

⑦ Moved the Pitot tube holder to the center and right halves of Section A. Repeated step 6 to complete the measurements for the right half of Section A and recorded the corresponding data.

⑧ Transferred all equipment sequentially to Sections B through F and repeated steps 6 and 7 to complete all tests under Case 1.

⑨ Turned off the fan, adjusted the deflector pitch angle as shown in Figure 13, and repeated steps 6 through 8 to conduct tests for each working condition corresponding to the current number of deflector plates.

⑩ Replaced the deflector and repeated the above procedures to complete the testing for test conditions 6 through 13.

2.6. Evaluation Index

Researchers have proposed a variety of indexes to evaluate the tunnel ventilation efficiency, such as pressure rise coefficients, mean velocity in a section, and air age [30,31,32]. In this paper, the pressure rise coefficient and the mean velocity in a section are used to evaluate the ventilation efficiency of fans.

The pressure rise coefficient $\eta$ is an important index to measure the ventilation efficiency of the jet fan, and its calculation method is shown in Equations (1)–(3).

$$\eta ={\displaystyle \frac{\Delta p_{aj}}{\Delta p_j}}$$

(1)

$$\Delta p_{aj}=p_{\max }-p_{\min }$$

(2)

$$\Delta p_j=m\rho v_j^2{\displaystyle \frac{A_j}{A_t}}\left(1-{\displaystyle \frac{v_t}{v_j}}\right)$$

(3)

where $\Delta p_{aj}$ is the actual rise pressure of the jet fan and it is obtained by on-site measurement or numerical simulation; $\Delta p_j$ is the theoretical rise pressure of the jet fan; $p_{\max }$ is the maximum mean static pressure in the tunnel cross-section downstream of the jet fan; $p_{\min }$ is the minimum mean static pressure in the tunnel cross-section upstream of the jet fan inlet; m is the number of jet fans, 4; $\rho$ is air density, 1.2 kg/m³; $v_j$ is the jet fan outlet wind speed, 35.50 m/s; $A_j$ is the fan cross-sectional area, 1.0 m²; $A_t$ is the tunnel cross-sectional area, 127.4 m²; and $v_t$ is the tunnel design wind speed, 6.75 m/s.

In tunnel jet ventilation systems, the mean velocity in the section is usually used to evaluate the state of jet development due to the large area of the section and the non-uniform distribution of velocity.

However, a higher $\eta$ does not directly equate to “good” ventilation. Excessively high $\Delta p_{aj}$ often involve significant local energy losses. This leads to uneven airflow distribution across the cross-section and can even create “dead zones”. These zones significantly reduce actual pollutant removal efficiency. In engineering practice, areas with wind speeds below 0.5 m/s are generally considered dead zones [33], which should be avoided. Therefore, this study concludes that a larger pressure rise coefficient and more uniform airflow distribution indicate a superior deflector shield configuration, provided the local minimum wind speed requirement (≥0.5 m/s) across the cross-section is satisfied. This principle serves as the evaluation criterion for the subsequent analysis.

3. Results and Discussion

3.1. Pressure Rise Coefficient Analysis

(1)

Static pressure analysis for each working case

Figure 14 shows the variation in static pressure with distance for different deflector pitch angles and deflector plate quantities. As shown, regardless of deflector plate quantity and pitch angle—including the control group—the static pressure referenced to Section A (upstream of the jet fan) initially increases with distance from the fan, peaks, then decreases rapidly. At a distance of 30 m, the measured static pressure within the tunnel drops below the reference value at Section A. Furthermore, the figure indicates that, for most cases, the static pressure reaches its maximum at approximately 10 m downstream of the fan. Exceptions are observed for two cases: the 3-deflector/4° configuration peaks at 15 m and the 4-deflector/4° configuration peaks at 5 m.

Observing Figure 14a,b, it is evident that when the number of deflector plates is 3 and 4, the static pressure in the test section undergoes an initial rise followed by a decline with increasing downstream distance from the jet fan outlet. Notably, significant differences are observed in both the magnitude and pattern of static pressure variation under different pitch angle conditions. In contrast, Figure 14c shows that when the number of deflector plates is 5, the peak values and trends in static pressure variation across different pitch angles are relatively consistent. It is worth noting that the maximum static pressure values for two test conditions (three and four deflector plates at a pitch angle of 4°) appear at 20 m and 5 m downstream of the fan, respectively.

(2)

Analysis of pressure rise coefficient for each working condition

The actual pressure rise values for each working case are presented in

Table 3. Actual pressure rise value of each working condition.

Deflector Plates Quantity	Pitch Angle (°)	Actual Pressure Rise Value (Pa)
3	0	3.68742
3	4	4.77789
3	6	3.79822
3	8	6.49818
3	10	4.68862
4	4	4.56317
4	6	4.63719
4	8	7.88234
4	10	7.11923
5	4	6.85527
5	6	7.58705
5	8	7.34637
5	10	7.85065

, and the corresponding pressure rise coefficients are shown in Figure 15. It can be observed that all working cases exhibit higher pressure rise coefficients compared to the control group. The variations in pressure rise coefficients for configurations with three and four deflector plates display similar trends. When the deflector pitch angle is small, the pressure rise coefficient remains within a narrow range and shows only slight fluctuations. These minor fluctuations are likely attributed to the weak deflection effect and airflow instability near the jet fan outlet. As the deflector pitch angle increases to 8°, a significant increase in the pressure rise coefficient is observed, indicating improved ventilation efficiency of the jet fan at this configuration. However, when the pitch angle is further increased to 10°, the pressure rise coefficient decreases. This reduction is likely due to intensified friction between the airflow and the tunnel floor caused by the excessive downward deflection, which adversely affects jet fan performance. Additionally, for configurations with five deflector plates, the pressure rise coefficient remains relatively stable across different pitch angles, fluctuating within a higher range. This suggests that increasing the number of deflector plates to five helps maintain a consistently high ventilation effect, with less sensitivity to pitch angle changes.

The essence of the change in the number of deflector plates is actually a change in deflector plate spacing. A deflector shield with a suitable pitch angle and deflector plate spacing is favorable for the ventilation efficiency of the jet fan. When the pitch angle is too small or the deflector plate spacing is too large, excessive pitch angles or insufficient deflector plate spacing cause the jet to strike the ground, resulting in significant energy loss.

3.2. Wind Speed Uniformity Analysis

In addition to ensuring a sufficient pressure rise coefficient of the jet fan, maintaining a uniform wind speed distribution within the tunnel is also critical for effective ventilation during actual operation. Specifically, minimizing regions of low wind speed is essential to enhance airflow performance throughout the tunnel. Building upon the previously analyzed pressure rise coefficients, this section focuses on wind speed uniformity under operating conditions exhibiting higher pressure rise performance.

Figure 16 presents the cross-sectional average wind speed distributions for six configurations: four deflector plates at a pitch angle of 8°, four deflector plates at a pitch angle of 10°, five deflector plates at a pitch angle of 4°, five deflector plates at a pitch angle of 6°, five deflector plates at a pitch angle of 8°, and five deflector plates at a pitch angle of 10°. Areas with wind speeds no more than 0.5 m/s are defined as dead zones [33]. As shown in Figure 16, there are no dead zones in all test conditions. To evaluate the uniformity of tunnel airflow, areas with wind speeds more than 1.8 m/s are further defined as high-wind-speed zones. For configurations with five deflector plates, the high-wind-speed zones primarily distribute within 20 m downstream of the fan. In contrast, the configurations with four deflector plates demonstrate superior long-distance airflow performance, maintaining the wind speed exceeding 1.8 m/s at 24 m downstream from the fan. Moreover, these configurations allow the jet fan-induced airflow to be effectively directed downward after 15 m, highlighting the functional advantage of the deflector plates. Therefore, the configurations with four deflector plates at 8° and 10° are deemed more favorable in terms of wind speed uniformity throughout the tunnel.

4. Conclusions

This paper presented a new deflector shield device for enhancing the ventilation efficiency of jet fans. A total of thirteen cases were selected for field testing, including one control group, three groups with different deflector plate quantities, and four groups with different deflector pitch angles. The main conclusions are as follows:

(1)

Regardless of the deflector plate quantity or pitch angle, the static pressure in the test section initially increases with distance from the jet fan, reaching a peak before rapidly decreasing. For three or four deflector plates, the static pressure distribution varies with the pitch angle; higher pitch angles correspond to greater peak static pressures and more pronounced pressure attenuation. In contrast, when five deflector plates are used, the influence of pitch angle on static pressure distribution is minimal, with both peak values and attenuation amplitudes remaining relatively consistent across different test sections.

(2)

All test conditions exhibit higher rise pressure coefficients than the control group. For configurations with three or four deflector plates, the rise pressure coefficient increases with pitch angle up to 8°, after which it declines. In contrast, when five deflector plates are used, the rise pressure coefficient remains relatively stable across different pitch angles, fluctuating around a consistently high level. Overall, a higher rise pressure coefficient is observed under the cases of four deflector plates at pitch angles of 8° and 10°, as well as five deflector plates at pitch angles of 4°, 6°, 8°, and 10°. The maximum pressure rise coefficient is achieved with four deflector plates at a pitch angle of 8°.

(3)

When the number of deflector plates is five, the deflection effect is excessively strong, leading to a sharp drop in average wind speed approximately 15 m downstream of the fan and resulting in extensive low-wind-speed regions further downstream. In contrast, the configurations with four deflector plates at 8° and 10° demonstrate improved wind speed uniformity downstream, with fewer low-velocity zones and more balanced airflow distribution.

(4)

Based on the combined analysis of the pressure rise coefficient and wind speed uniformity, it is concluded that the optimal ventilation performance of the jet fan is achieved with four deflector plates at a pitch angle of 8°.

The recommended configuration in the work is the result of specific tunnel scenarios and may have certain limitations. However, for ventilation design in other space-limited tunnels or optimization of existing ventilation systems, the evaluation criteria outlined in this work may also serve as a reference. To promote the application of the deflector shield system and advance the design and operation efficiency of tunnel ventilation systems, three directions will be examined in future studies. Firstly, we plan to develop a high-fidelity CFD model coupled with machine learning algorithms for intelligent parameter optimization and to establish the relevant dimensionless calculation formula. Meanwhile, we will systematically evaluate energy consumption and the carbon footprint across different configurations of the deflector shield system. Finally, we aim to establish adaptive deflector adjustment strategies, incorporating IoT technologies, based on real-time traffic flow and pollutant concentration monitoring.