Experimental investigation of the protective effect of wind barriers on high-speed railway bridge in inland strong wind area

Abstract

This study aims to investigate the wind protective effect of wind barriers on the Lanzhou–Xinjiang high-speed railway using wind tunnel tests. Wind barriers with different heights and porosities were analyzed. Two girders, that is, a box-girder and a trough-girder, each with 1:30 and 1:8 scales were experimentally investigated. The results suggest that the protective effect of the wind barrier with a height of 4 m and porosity of 20% is better than the others. The influence of wind barriers on the aerodynamic characteristics of train vehicles and girders must be analyzed simultaneously. The aerostatic force coefficients of trains are approximately the same at different scales, and the Reynolds number effect could be neglected.

Keywords

aerostatic force coefficient girder protective effect train wind barrier wind tunnel test

Introduction

Train vehicles are subject to great risk when high crosswind conditions occur. There are strong evidences (Baker et al., 2004; Bocciolone et al., 2008) showing that the effects of crosswind contribute significantly to the occurrence of train vehicles accidents. Wind barriers are considered as the most effective device to ensure the operational safety and comfort of train vehicles (Imai et al., 2002; Liu, 2010). High number of bridges are being constructed in high-speed rail lines in China, which may reach up to 80% of the length (He et al., 2014). The aerodynamic forces of the train vehicles under crosswinds depend not only on the shapes of the vehicles but also on the infrastructure, for example, bridges (Cheli et al., 2010; Dorigatti et al., 2012; Suzuki et al., 2003). Although wind barriers could effectively reduce the wind loads on train vehicles, the height and porosity of wind barriers may greatly influence the aerodynamic forces of the train vehicles, and they have negative effects on bridges because of the changes in cross-section and the complex flow field around bridges. Thus, the optimization of wind barriers is not only required to ensure safe operation of a train vehicle but also to reduce the wind loads on the bridge girders; as a result, such optimization is of great significance for engineering application.

Many researchers have focused on the flow field behind wind barriers, such as the mean and turbulent wind characteristics (Dong et al., 2010; Kozmar et al., 2012, 2014; Lee and Kim, 1999; Procino et al., 2008; Santiago et al., 2007). Investigation of the aerodynamic forces on train vehicles and bridges may be another factor to be considered when optimizing the height and porosity of the wind barriers. Fujii et al. (1999) presented the wind-induced accidents in Japan and investigated the aerodynamic forces on train vehicles. In addition, several software and hardware preventive measures were conducted. Strukelj et al. (2005) tested the wind forces on train vehicles using a numerical simulation method and showed how the resulting wind forces could affect the vehicle dynamic responses and traffic safety issues. Based on the pressure measurement method, Barcala and Meseguer (2007) investigated the effect of solid parapets on the aerodynamic forces acting on a train vehicle, considering the configurations of the viaduct and the ground. The results showed that solid parapets appeared to be more effective on the viaduct than on the ground. In 2008, they used smoke visualization around the train vehicle model on a viaduct (Barcala and Meseguer, 2008). There was a subtle difference in the streamline impinging in the front of the bridge deck, which might have an influence on the lift force on the train–bridge system. Wind tunnel tests combining different positions of train vehicles on double tracks on the viaduct were performed by He et al. (2014); they also studied the effect of the wind barriers. Xiang et al. (2014) investigated the aerodynamic forces on a train vehicle and the wind pressure above the track via wind tunnel tests. Zhang et al. (2015) investigated the six-component aerostatic forces on train vehicles on a bridge; the influential factors of the positions of the train vehicles, the height and porosity of the wind barriers, the wind speed, and the wind yaw angles were considered in the tests.

Given the above discussion, which only considers the condition of train vehicles on a box-girder, there are two main problems in these research studies. First, little attention is paid to the influence of the bridge girders on the protective effect of the wind barriers. Second, small-scale models are usually tested because of the limitations of the wind tunnel size, the model size, and so on. This article describes an attempt to address these problems using the investigation of the Lanzhou–Xinjiang high-speed railway as a case study.

This article describes the investigation of the optimization of wind barriers on the Lanzhou–Xinjiang high-speed railway based on wind tunnel tests. The line, which was opened on 26 December 2014, passes through five strong wind regions, with the wind speed reaching up to 60 m/s. A detailed introduction of the project is provided in section “Engineering background.” The test methods and working conditions are described in section “Experimental setup and method.” The aerostatic force coefficients of the train vehicle and the bridge girder with 1:30 and 1:8 scales are compared in section “Wind tunnel test results and discussion.”

Engineering background

The Lanzhou–Xinjiang high-speed railway (Figure 1) is 1776-km long and passes through five strong wind regions (An-xi wind region, Yan-dun wind region, Bai-li wind region, San-shi-li wind region, and Da-ban-cheng wind region), the length of which is 32.6% of the total length of the line. In the Xinjiang Uygur Autonomous Region, the length of the line is 710 km, and the length of the line that passes through the wind regions could reach up to 462.409 km, which is 65.1% of the length of the line in Xinjiang. The total length of the bridge in the rail lines could reach up to 206 km, and most of the girder sections are box-girder and tough-girder.

Figure 1.

Lanzhou–Xinjiang high-speed railway.

Table 1 shows the basic parameters of the five wind regions. There are approximately 100 days every year that the wind speed is higher than levels 8 in these wind regions. During the year of 2002–2007, the wind speed exceeded levels 8 for 131 days in the Bai-li wind region and for 121 days in the San-shi-li wind region. The probability of the wind speed exceeding levels 12 is approximately 10%–15%. Therefore, it is of great significance to test the wind protect effect of the wind barriers.

Table 1.

Basic parameters of the five wind regions.

Positions	Wind regions	Length (km)
Gansu Province	An-xi	117.19
Xinjiang Uygur autonomous region	Yan-dun	129.325
	Bai-li	170.114
	San-shi-li	86.398
	Da-ban-cheng	76.572

Experimental setup and method

Models

In the following experimental studies, the CRH2 passenger vehicle, which is widely used in China, was selected as the prototype model. The dimensions of the middle vehicle are 3.38 m in width and 3.5 m in height (Figure 2). The bottom of the vehicle model was simplified as a flat plane without consideration of the effects of bogie, wheels, and so on. The vehicle model was constructed using wood.

Figure 2.

Cross-section of the middle trailing vehicle (unit: m).

Two different types of bridge girders were tested in the wind tunnel tests. One type was the simply-supported box-girder (13.4 m wide and 3.5 m high) with a span length of 32 m (Figure 3(a)). The other type was the simply supported trough-girder (14.2 m wide and 2.0 m high) with a span length of 16 m (Figure 3(b)).

Figure 3.

Cross-sections of the bridge girders: (a) box-girder and (b) trough-girder (unit: m).

In Figure 4, the wind barrier porosity is illustrated, and the shape of the holes is circular. In 2013, Xiang et al. (2013) investigated the simulation method of a scaled model for a porous wind barrier, and it is found that the hole shape of a wind barrier has some effect on the aerodynamic coefficients of the train and the circular hole size of a wind barrier at the scale of 1:20 should be in the range of 8–12 mm. Figure 5 shows the arrangement of the wind barriers on the girders. Only two working conditions, that is, wind barriers at the windward side and wind barriers at both sides of the girder, were considered.

Figure 4.

Illustration of the wind barriers with different porosities: (a) porosity 0% and (b) porosity 20%.

Figure 5.

Wind barriers on the bridge girder: (a) case A: wind barrier at the windward side of the box-girder, (b) case B: wind barrier at both sides of the box-girder, (c) case C: wind barrier at the windward side of the trough-girder, and (d) case D: wind barrier at both sides of the trough-girder.

1:30 models in the XNJD-1 wind tunnel

The XNJD-1 wind tunnel, located at Southwest Jiaotong University, is a closed-circuit wind tunnel with two tandem closed-test sections; the dimensions of the second section are 2.4 m × 2.0 m × 16.0 m (width × height × length), with a wind velocity of 0.5–45.0 m/s. The test section is equipped with a force balance system. The attack angle can be adjusted from −20° to 20° at a minimum increment of 0.1°. There is a 2- to 3-mm gap between the train vehicle and the bridge girder model. The tests are conducted in the smooth flow, as shown in Figures 6 and 7. In the 1:30 scaled wind tunnel tests, the characteristics of the simulated flows were measured by Cobra probe, which was capable of resolving three components of mean and fluctuating velocities with a frequency response up to 2 kHz. In order to examine the uniformity of the simulated flow characteristics, wind data were measured repeatedly along the test model. No significant difference was observed between the results measured at different positions. The wind speed of smooth flow is 20 m/s.

Figure 6.

Box-girder model in the XNJD-1 wind tunnel: (a) porosity 0% and (a) porosity 30%.

Figure 7.

Trough-girder model in the XNJD-1 wind tunnel (case C).

1:8 models in the XNJD-3 wind tunnel

The XNJD-3 wind tunnel, located at Southwest Jiaotong University, Chengdu, Sichuan province, is the largest wind tunnel for wind-engineering application in the world. It is a closed-circuit-type wind tunnel with a test section that is 4.5 m high, 22.5 m wide and 36 m long. The wind speed for the empty tunnel ranges from 1 to 16.5 m/s. In the 1:8 scaled wind tunnel tests, the wind speed is 15 m/s. The oncoming turbulent intensity is less than 1.0%. In addition, the characteristics of the simulated flows were measured by Cobra probe and the uniformity of the simulated flow characteristics were examined. No significant difference was observed between the results measured at different positions.

Figure 8 shows the models in wind tunnel. A force balance system for static wind loading testing of the vehicle model was installed in the vehicle models. In addition, the leading vehicle and tail vehicle were considered in the tests, but only the middle vehicle was the test model.

Figure 8.

Models of 1:8 scale in the XNJD-3 wind tunnel.

Aerodynamic force and the moment coefficients

In reference to Figure 9, the mean aerodynamic force and the moment coefficients are defined as follows

\begin{array}{l} F_{H} = \frac{1}{2} ρ U^{2} C_{H} D \\ F_{V} = \frac{1}{2} ρ U^{2} C_{V} B \\ M = \frac{1}{2} ρ U^{2} C_{M} B^{2} \end{array}

(1)

where $ρ$ is the air density, U is the mean wind speed, and D and B are the height and width of the structure, respectively. C_H, C_V, and M are the three component force coefficients. F_H is the lateral force. F_V is the lift force. M is the torsion moment. O is the centroid of the bridge girder cross-section, and O" is central of the train model. These definitions are separated for train vehicle and bridge deck.

Figure 9.

Schematic diagram of the three component forces.

Wind tunnel test results and discussion

The mean three components force coefficients of the train vehicle and bridge girder models in XNJD-2 and XNJD-3 wind tunnel were tested several times in each working condition and the results were essentially the same which proves that the test results are reliable. The results listed in the article are the mean values of wind tunnel tests.

1:30 wind tunnel tests

Case A and case B

First, taking Case A as an example, the wind-protective effect of the wind barriers on the box-girder was investigated. Wind barriers with porosities of 0%, 10%, 20%, 30%, and 40% and heights of 0, 2, 3, 4, 5, 6, and 7 m were tested. In the wind tunnel tests, the train vehicle models were still on the box-girder. Barcala and Meseguer (2008) reported that a stationary vehicle model subjected to wind perpendicular to the vehicle side does not exactly reproduce the real situation but appears to be suitable to analyze the relative effectiveness of wind barriers on the aerodynamic forces on train vehicles. The detailed operating conditions are listed in Table 2.

Table 2.

Wind barrier on the box-girder of 1:30 scale.

Arrangement	Porosity (%)	Height (m)
Case A	0, 10, 20, 30, 40	0, 2, 3, 4, 5, 6, 7

Figure 10 presents the aerostatic force coefficients of the train vehicle at the windward side. These coefficients are decreased with the increase in height of the wind barriers. When the wind barrier is higher than 3 m, the lateral force and moment coefficients of the train vehicles are approximately the same. When the height of wind barriers is less than 3 m, the lateral force and moment coefficients of the train models were decreased as the height of wind barriers decreased. While, for the lift force coefficients, the critical height of the wind barrier is 4 m. The lateral force and moment coefficients of the train vehicle increase with the increase in porosity of the wind barrier. When the wind barrier is 4 m high and the porosity is 0%, the lateral force and moment coefficients of train vehicles are 0.022 and 0.016, respectively; they are increased to 0.89 and 0.516, respectively, for a wind barrier with a porosity of 40%. When the porosity of the wind barrier is less than 20%, the lift force coefficients decrease as the porosity increases. Thus, the lift force coefficients of the train vehicles increase as the porosity increases. The height and porosity of the wind barrier have effect on the aerodynamic characteristics of trains on girder. But the aerostatic force coefficients of girder and economic factors must also be taken into account.

Figure 10.

Aerodynamic force coefficients of the train vehicle vs the wind barrier height H (Case A): (a) lateral force coefficient, (b) lift force coefficient, and (c) moment coefficient.

Therefore, the protective effect of the wind barrier that is 4 m high and has a porosity of 20% is better than the other cases. Some further investigations could be conducted to obtain the optimal scheme of the wind barriers and the detailed working conditions of which are listed in Table 3. Use of a wind barrier at one side and at both sides of the box-girder is investigated. The wind barrier is composed of part 1 and part 2. Wind barrier in case A2 as an example, the arrangement is showed in Figure 11. For part 1, the height of the wind barrier is 1 m and the porosity is 10%, this part is at the bottom. For part 2, the height of the wind barrier is 2.5 m and the porosity is 20%.

Table 3.

Working conditions of the wind barriers on the box-girder.

Case	Height of the wind barrier	Styles (part 1 + part 2)^a
A1	0	NO
A2	3.5	1 m 10% + 2.5 m 20%
A3	4	4 m 20%
A4	4	1 m 10% + 3 m 20%
A5	4	2 m 10% + 2 m 20%
A6	4	2.3 m 0% + 1.7 m 20%
B1	3.5	1 m 10% + 2.5 m 20%
B2	4	1 m 10% + 3 m 20%
B3	4	2 m 10% + 2 m 20%
B4	5	3.4 m 0% + 1.6 m 20%
B5	7	3.4 m 0% + 3.6 m 20%

(1 m 10% + 2.5 m 20%) represents that, for part 1, the height of wind barrier is 1 m and the porosity is 20%; for part 2, the height of wind barrier is 2.5 m and the porosity is 20%.

Figure 11.

Arrangement of the wind barrier in case A2 (unit: m).

Figure 12 shows the lateral force coefficients and moment coefficients of the train vehicles. The lift force coefficients are not listed here because they are all very small and approximately the same. The protective effect of the wind barriers at both the windward and leeward sides of the box-girder is relatively better than that only at the windward side. For heights of the wind barrier of 5 and 7 m, the aerostatic force coefficients of the train vehicle are the smallest and the protective effect is the best. Generally, there is no explicit advantage for the case of the wind barriers at both sides of the girder, considering the positive effect on the train vehicles, the negative effect on the box-girder and the economic factors. For cases A3, A4, A5, and A6, the lateral force coefficients of the train vehicles increase as the porosity increases, and they are all much smaller than the lateral force coefficients of the train vehicle on the box-girder without the wind barrier. In condition A2 and A4, the lateral force coefficient are decreased 86% and 87%, respectively, comparing to that without wind barriers. This may be due to the differences of the wind barriers at part 2 in Figure 11. The wind protection of wind barriers in condition A6 is better than that in condition A1–A5. Because the porosity of wind barriers is 0% in part 1. In condition A4, A5, and A6, the moment coefficients of train model are roughly the same. In addition, the lift force coefficients of the train vehicles are much smaller than the lift force coefficient under the condition that the box-girder without wind barriers.

Figure 12.

Aerostatic force coefficients of the train vehicles on the box-girder at a wind attack angle of 0°: (a) lateral force coefficient and (b) moment coefficient.

Figure 13 shows the aerostatic force coefficients of the bridge girder. Wind barriers have some negative effects on the bridge box-girder. The drag force coefficients of the girder are all approximately 3.5, which are larger than that in case A1, considering the influence of the wind barrier. In case B5, the drag force coefficients of the girder could reach 4.89. In case A3, the drag force and moment coefficient of the girder, which are 3.00 and 0.12, respectively, are both the smallest. In other cases, the moment coefficients could reach up to approximately 0.3.

Figure 13.

Aerostatic force coefficients of the box-girder at a wind attack angle of 0°: (a) drag force coefficient and (b) moment coefficient.

Case C and case D

The detailed working conditions of the vehicles on the trough-girder with wind barriers are listed in Table 4. Generally, wind barriers at a single side and at both sides of the trough-girder are considered, and the arrangement of wind barriers is similar to that in Figure 11.

Table 4.

Working conditions of the wind barriers on the trough-girder with 1:30 scale.

Cases	Height of the wind barrier (m)	Styles (part 1+part 2)
C1	0	NO
C2	4	4 m 20%
C3	4	2 m 0% + 2 m 20%
D1	3.5	1.5 m 10% + 2 m 20%
D2	5	2.3 m 0% + 2.7 m 20%
D3	7	2.3 m 0% + 4.7 m 20%

In Figure 14, the lateral force and moment coefficient of the train vehicles on the trough-girder without wind barriers are 1.572 and 0.84, respectively. Considering the influence of the wind barriers, the aerostatic force coefficients of the train vehicle are greatly decreased. In case C3, the lateral force coefficients of the train vehicles are 0.117, which is less than half of those in case C2. The moment coefficients are 0.163 and 0.129. It could be inferred that part 1 of the wind barrier is of great significance for the protective effect on train vehicles. For case D, there is no explicit advantage for the protective effect of the wind barrier at both sides of the girder.

Figure 14.

Aerostatic force coefficients of train vehicles on the trough-girder: (a) lateral force coefficient and (b) moment coefficient.

Figure 15 shows the aerostatic force coefficients of the trough-girder. The drag force of the girder without a wind barrier is 1.133, and it is increased to 4.513 in case C2, which is approximately four times that in case C1. In other cases, the drag force coefficients are all greater than 5. There is no great difference for the moment coefficients. Again, it is stressed that the influence of the wind barrier on the aerodynamic characteristics of train vehicles and the girder must be considered simultaneously.

Figure 15.

Aerostatic force coefficients of the trough-girder: (a) drag force coefficient and (b) moment coefficient.

Discussion of the influence of girder

The aerostatic force coefficients of the train vehicles on the box-girder and the trough-girder that without wind barriers are shown in Figure 16. The lateral force and lift force coefficients of the train vehicles on the box-girder are relatively larger than those on the trough-girder, whereas the moment coefficients of the train vehicles on the box-girder are smaller. Note that the ratio of the width to height of the box-girder is 3.83, whereas the ratio of the width to height of the trough-girder is 7.10. Therefore, the cross-section of the box-girder is blunt and the acceleration effect of the flow around box-girder leads to the drag on train vehicles being relatively higher.

Figure 16.

Aerostatic force coefficients of the train vehicles on girder without a wind barrier.

Figure 17 shows the protective effect of the wind barrier on the box-girder and the trough-girder at wind angle of attack from −3° to +3°. The wind barrier is 4 m high, and the porosity is 20%, which is indicated by case A3 in Table 3 and case C2 in Table 4. Note that the wind angle of attack has little influence on the aerostatic force coefficients of the train vehicles. The lift force coefficients of train vehicles on the box-girder and the trough-girder are approximately the same. The drag force coefficients of the train vehicles on the box-girder are slightly smaller than those on the trough-girder. The absolute values of the moment coefficients of the train vehicles on the trough-girder and box-girder are 0.06 and 0.2, respectively.

Figure 17.

Aerostatic force coefficients of the train vehicles on the box-girder and the trough-girder with a wind barrier at the windward side (4 m high, porosity 20%): case A3: 4m porosity 20%, box girder; case C2: 4m porosity 20%, tough girder.

In order to compare the wind protect effect of wind barriers, parameter $ξ$ are defined as follow

ξ = | \frac{Δ C_{i, case}}{C_{i, without}} |

(2)

Here $i = H, V, M, Δ C_{i, case}$ represents difference of the lateral force, lift force, or moment coefficients of the train on girder with and without wind barriers. $C_{i, without}$ represents the aerostatic forces and moments coefficients of the train on girder without wind barriers.

From Figure 18, it could be noted that the protective effect of the wind barrier differs for different cross-sections of girders. For the drag force coefficients, the parameter $ξ$ of the case A3 is larger than that in case C2. For moment coefficients, the parameter $ξ$ of the case A3 is a little smaller than that in case C2.

Figure 18.

Wind-protect effect of wind barriers in case A3 and case C2.

1:8 wind tunnel tests

Figure 19 shows the results obtained from 1:8 wind tunnel tests. The results listed in the figures are all for the middle train vehicle, considering the influence of the leading and tail vehicles. The lateral force and moment coefficients of the train vehicles on the girder with wind barriers (case A3—4 m high and porosity of 20%) are approximately the same at a wind yaw angle of 90°. When the wind barrier is removed (case A1), the lateral force and moment coefficients of vehicles with 1:30 scale are relatively higher. This could be explained by the Reynolds number $(Re = (U \cdot D) / υ)$ effect. The cross-section of models with a wind barrier is more blunt, and the Re effect may be small.

Figure 19.

Aerostatic force coefficients of the train vehicles on the box-girders with 1:8 and 1:30 scales: (a) lateral force coefficient and (b) moment coefficient.

Figure 20 shows the aerostatic force coefficients at different wind yaw angles. The lateral force and moment coefficients are roughly the same at different wind yaw angle from 70° to 90°. Figure 21 shows the aerostatic force coefficients of train vehicles on the trough-girder in 1:8 and 1:30 scale. The results are relatively smaller for models in case C1 with 1:8 scale than that with 1:30 scale. While, in case C3, the lateral force and moment coefficients are roughly the same for train models with different scales. Comparing the results in Figures 19 and 21, the Re effect in case C1 is relatively larger than that in case A1. This may be due to the section of train on box-girder is more blunt than the section of train on tough-girder.

Figure 20.

Aerostatic force coefficients of the train vehicles on the box-girders with 1:8 and 1:30 scales at different wind yaw angle: (a) lateral force coefficient and (b) moment coefficient.

Figure 21.

Aerostatic force coefficients of the train vehicles on trough-girders with 1:8 and 1:30 scales: (a) lateral force coefficient and (b) moment coefficient.

Conclusion

This study investigated the wind protective effect of wind barriers on the Lanzhou–Xinjiang high-speed railway using wind tunnel tests and three-dimensional (3D) numerical simulation. The major conclusions of this study are summarized as follows:

The protective effect of the wind barriers at both the windward and leeward sides of the box-girder is relatively better than that only at the windward side. Moreover, considering the negative effect on the box-girder, there is no explicit advantage for the protective effect of the wind barriers at both sides of the girder.

When a train vehicle is on a trough-girder with wind barriers, the drag force coefficient of the train vehicle is decreased to 0.4 (results in case C) from 1.572 (no wind barriers) and the moment coefficient is decreased to 0.2 from 0.8.

The protective effects of the wind barrier on the aerodynamic characteristics of the train vehicles and girder must be considered simultaneously. The cross-section of the models with the wind barrier is more blunt, and the Re effect may be slight.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The research described in this article was financially supported by the Natural Science Foundation of China (grant no.: 51778545) and the Project of Science Technology Research and Development Program of China Railway Corporation (grant no.: 2010G004—I).

ORCID iDs

Qingsong Duan

Cunming Ma

Qiusheng Li

References

Baker

Jones

Lopez-Calleja

, et al. (2004) Measurements of the cross wind forces on trains. Journal of Wind Engineering and Industrial Aerodynamics 92: 547–563.

Barcala

Meseguer

(2007) An experimental study of the influence of parapets on the aerodynamic loads under cross wind on a two-dimensional model of a train vehicle on a bridge. Proceedings of the Institution of Mechanical Engineers, Part F: Journal of Rail and Rapid Transit 221: 487–494.

Barcala

Meseguer

(2008) Visualization study of the influence of parapets on the flow around a train vehicle under cross winds. Computers in Railways XI 103: 797–806.

Bocciolone

Cheli

Corradi

(2008) Crosswind action on rail vehicles: wind tunnel experimental analyses. Journal of Wind Engineering and Industrial Aerodynamics 96: 584–610.

Cheli

Ripamonti

Rocchi

(2010) Aerodynamic behavior investigation of the new Emuv250 train to cross wind. Journal of Wind Engineering and Industrial Aerodynamics 98: 189–201.

Dong

Luo

Qian

, et al. (2010) A wind tunnel simulation of the turbulence fields behind upright porous wind fences. Journal of Arid Environment 74: 193–207.

Dorigatti

Sterling

Rocchi

, et al. (2012) Wind tunnel measurements of crosswind loads on high sided vehicles over long span bridges. Journal of Wind Engineering and Industrial Aerodynamics 107–108: 214–224.

Fujii

Maeda

Ishida

, et al. (1999) Wind-induced accidents of train/vehicles and their measures in Japan. Quarterly Report of RTRI 40: 50–55.

Zou

Wang

, et al. (2014) Aerodynamic characteristics of a trailing rail vehicles on viaduct based on still wind tunnel experiments. Journal of Wind Engineering and Industrial Aerodynamics 135: 22–33.

10.

Imai

Fujii

Tanemoto

, et al. (2002) New train regulation method based on wind direction and velocity of natural wind against strong winds. Journal of Wind Engineering and Industrial Aerodynamics 90: 1601–1610.

11.

Kozmar

Procino

Borsani

, et al. (2012) Sheltering efficiency of wind barriers on bridges. Journal of Wind Engineering and Industrial Aerodynamics 107–108: 274–285.

12.

Kozmar

Procino

Borsani

, et al. (2014) Optimizing height and porosity of roadway wind barriers for viaducts and bridges. Engineering Structures 81: 49–61.

13.

Lee

Kim

(1999) Laboratory measurements of velocity and turbulence field behind porous fences. Journal of Wind Engineering and Industrial Aerodynamics 80: 311–326.

14.

Liu

(2010) Study on the system to ensure train operation safety under strong wind. Engineering Mechanics 27: 305–310.

15.

Procino

Kozmar

Bartoli

, et al. (2008) Wind barriers on bridges: the effect of wall porosity. In: Proceedings of the 6th international colloquium on bluff bodies aerodynamics & applications, Milano, 20–24 July.

16.

Santiago

Martín

Cuerva

, et al. (2007) Experimental and numerical study of wind flow behind windbreaks. Atmospheric Environment 41: 6406–6420.

17.

Strukelj

Ciglaric

Pipenbaher

(2005) Analysis of a bridge structure and its wind barrier under wind loads. Structural Engineering International 15: 220–227.

18.

Suzuki

Katsuji

Tatsuo

(2003) Aerodynamic characteristics of train/vehicles under cross winds. Journal of Fluids & Structures 91: 209–218.

19.

Xiang

Chen

, et al. (2014) Protection effect of railway wind barrier on running safety of train under cross winds. Advances in Structural Engineering 17: 1177–1187.

20.

Xiang

Zhe

, et al. (2013) Simulation method of porous wind screen scale model on bridge by wind tunnel tests. Engineering Mechanics 30: 212–216.

21.

Zhang

Guo

Xiong

, et al. (2015) Experiment study on the effect of wind barrier on aerodynamic characteristics of high-speed train on bridge. Journal of Central South University (Science and Technology) 46: 3888–3397.