Introduction

With the continuous development of the rail transit industry, the ever-increasing operating speed of trains has made aerodynamic effects a key factor restricting high-speed operation. When a train runs at speeds above 200 km/h under strong crosswind conditions, its operational safety faces severe challenges. Under the combined influence of a 30 m/s crosswind and an operating speed of 200 km/h, the lateral aerodynamic load on the rail vehicle will increase abruptly, causing the wheel–rail contact force to drop by about 40%. At this point, the train’s derailment coefficient will exceed the safety threshold of 0.8, posing a significant safety hazard1. In engineering practice, a commonly used flow-guiding optimization strategy is to install streamlined inter-car fairing devices at the connections between EMU cars, which can effectively reconstruct the distribution characteristics of the train’s surrounding flow field. This device can reduce the maximum turbulence intensity in the inter-car gap region by about 30%. Such a flow field reconstruction mechanism can lower the aerodynamic lateral force under crosswind conditions by 18%–22%, thereby significantly enhancing the continuity of the train’s aerodynamic profile and improving operational stability2. The design of inter-car fairings is crucial to the aerodynamic performance of trains. Therefore, studying the effects of different fairing designs on train aerodynamics can effectively enhance operating speed, ensure safety and comfort, and play an important role in drag reduction and noise suppression. Currently, aerodynamic performance studies of high-speed trains are primarily conducted through full-scale field tests3,4,5, numerical simulations6,7, and wind tunnel experiments8,9,10.Existing research indicates that fairing configurations significantly influence the aerodynamic drag and lift distributions of trains. For instance, fully enclosed fairings can reduce aerodynamic drag by up to 22.5% compared to conventional semi-enclosed designs and can effectively stabilize the flow field at inter-car connections11. The degree of closure and gap width of fairings alter surface pressure distributions; fully sealed fairings can reduce pressure drag at inter-carriage joints to approximately 30% of that observed with typical fairings12. Moreover, differences in fairing designs can lead to variations of up to 12.9% in crosswind-induced overturning resistance, and streamlined fairings can reduce the size of wake vortices on the leeward side by up to 40%. Although these findings underscore the importance of fairings, investigations into their aerodynamic coupling effects under varying speeds and complex wind environments remain limited13,14,15,16. Mohebbi et al. analyzed the influence of barrier height on the stability of high-speed trains under crosswind conditions by employing the lattice Boltzmann method17,18.Li Tian et al. investigated aerodynamic performance at inter-carriage connections and concluded that fully enclosed external fairings further optimize train aerodynamics19. Sun Zhikun et al. studied the effects of varying wind directions on flow fields and aerodynamic loads around trains equipped with different fairing structures under crosswind conditions20. Zhang Ye et al. examined the influence of external wind deflectors and train surface pressure distributions across different track structures at 200 km/h21. Tang Mingzan et al. found that aligning fairings flush with the train body’s exterior surface effectively reduces aerodynamic drag22. Zuo Xiong et al. employed numerical simulations to explore unsteady aerodynamic characteristics of high-speed trains under crosswind conditions23.

The results reveal that the exterior fairing structure of high-speed trains plays a crucial role in reducing aerodynamic drag and enhancing aerodynamic performance, with fully enclosed exterior fairings achieving the best optimization effect. Among high-speed train fairing structures, the most common fully enclosed types are the capsule-shaped fairing and the folded-canopy fairing. At present, most studies focus on the capsule-shaped fairing, while research on the folded-canopy fairing remains relatively limited. Studies comparing the differences in aerodynamic performance between these two types of fully enclosed fairings are relatively scarce.

Therefore, this study employs three-dimensional numerical simulations to develop an aerodynamic model of a high-speed train operating under crosswind conditions. The influences of varying train speeds and crosswind velocities on the two fairing structures are comprehensively analyzed, providing theoretical insights and data support for the structural optimization of inter-carriage fairings.

Computational model, domain, and boundary conditions

Computational model

The numerical simulations were conducted using the Fuxing high-speed train model. A three-car formation is used, consisting of a head car, a middle car, and a tail car. The model excludes components such as pantographs, door handles, and headlights, but retains parts like the train bogies. Two different full-coverage inter-car fairing structures are employed for the fairing sections: the capsule-type and the folded-bellows-type. As shown in Fig. 1 is the train model with capsule-type inter-car fairing, and Fig. 2 shows the train model with folded-bellows-type inter-car fairing. The theoretical basis of this modeling strategy lies in the fact that the overall aerodynamic characteristics of the train are primarily determined by the head and tail cars: the head car governs flow separation, flow-around behavior, and the generation of pressure waves, while the tail car dominates the wake structure and rear resistance. In contrast, the middle car has a relatively regular geometry, and its aerodynamic characteristics are mainly manifested as the cumulative effect of repetitive surface friction drag. Therefore, a single middle car can effectively represent the central characteristics of the whole train. This approach, while ensuring that the major aerodynamic features of the train are captured, significantly reduces computational scale and resource consumption. It has been extensively validated to accurately reflect the overall aerodynamic performance of actual trains, and thus has become a commonly adopted modeling method in numerical simulation studies.

High-speed trains operate on various track foundations, including flat ground, embankments, and bridges. Among these, flat ground operation is the most common, so this study focuses solely on the flat ground running condition.

Fig. 1
Fig. 1
Full size image

Capsule-type inter-car fairing train model. (a) Overall Train Model, (b) Capsule-type inter-car fairing model.

Fig. 2
Fig. 2
Full size image

(a) Overall Train Model. (b) Folded-bellows-type inter-car fairing model.

Computational domain setup

The computational domain is shown in Fig. 3. The length of the domain in the direction of the train’s length is 240 m, the width of the domain in the direction of the train’s width is 120 m, and the height of the domain in the direction of the train’s height is 40 m. The velocity inlet 1 is located 40 m from the train’s nose, while pressure outlet 1 is located 120 m from the tail. The velocity inlet 2 is 40 m from the center of the train body, and pressure outlet 2 is 80 m from the center of the train body. The distance from the ground to the top boundary of the computational domain is 40 m, and the distance between the train’s bottom wheels and the ground is 0.18 m.

Fig. 3
Fig. 3
Full size image

Computational domain setup.

In the flow field simulation, the train surface is set as a no-slip wall boundary condition. To reduce the ground effect on the train simulation, the ground boundary is set as a slip wall with a sliding velocity equal to the train’s operating speed of 400 km/h. The inlet is located directly in front of the train and to the left of the train’s moving direction. The wind speed is synthesized from the train’s operating speed (400 km/h) and the lateral airflow speed (15 m/s), resulting in a combined velocity of 112.12 m/s at an angle of 7.7° to the horizontal plane, as shown in Fig. 4. The outlet is set behind the train and to the right of the train’s moving direction, with a pressure outlet boundary condition. The top boundary is set as a symmetry boundary to reduce computational effort.

Fig. 4
Fig. 4
Full size image

Schematic diagram of the synthetic wind direction.

Computational method

The aerodynamic coefficients are key factors in describing the aerodynamic performance of high-speed trains. The expressions for the aerodynamic drag coefficient, lift coefficient, and side rolling moment coefficient are as follows:

Aerodynamic Drag Coefficient:

$${C_d}=\frac{{2{F_d}}}{{\rho v_{{ref}}^{2}S}}$$
(1)

Aerodynamic Lift Coefficient:

$${C_1}=\frac{{2{F_1}}}{{\rho V_{{ref}}^{2}S}}$$
(2)

Side Rolling Moment Coefficient:

$${C_M}=\frac{{2M}}{{\rho V_{{ref}}^{2}Sh}}$$
(3)

In the equations:

\({F_d}\)— Aerodynamic Drag.

\({F_1}\)— Aerodynamic Lift.

M— Side Rolling Moment

\(\rho\)— Air Density \(\rho\)= 1.225 kg/\(\:{\text{m}}^{3}\).

S— Reference Area S= 12.0\(\:{\text{m}}^{2}\)

\({v_{ref}}\)— Resultant Wind Speed.

h— Reference Height h=3 m

The train operating speed is set to 400 km/h, corresponding to a Mach number of 0.33. Since the Mach number exceeds 0.3, air is modeled as an ideal compressible gas. The numerical simulation employs the three-dimensional, steady, compressible Navier–Stokes equations coupled with the \(SST\)turbulence model. This model uses the \(k - \omega\) model near the wall region and the \(k - \omega\) model in the free stream, with a blending function to achieve a smooth transition between the two. Calculations are performed using the commercial CFD software Fluent 2021R1, adopting the SIMPLEC algorithm with second-order accuracy. The computational hardware used includes an AMD R9-7945HX CPU and 64 GB of RAM.

Grid generation and grid independence verification

Mesh generation was performed using Fluent Meshing, with both the train body and computational domain adopting Poly-Hexcore polyhedral core meshes. To minimize the impact of mesh division on the calculation results, mesh refinement was applied to the fairing, train body, and the surrounding areas of the train, as illustrated in the mesh distribution schematic in Fig. 5. The minimum mesh size for the train body was set to 16 mm, while the fairing area was refined to a minimum mesh size of 6 mm.

Fig. 5
Fig. 5
Full size image

Grid division schematic.

To reasonably select the computational mesh, a train model with a capsule-type inter-car fairing as the car-body connection was used to simulate the train running at 400 km/h under a crosswind speed of 15 m/s. Three different meshes with varying numbers and layers of boundary layers were generated for mesh independence verification to study the influence of mesh variation on the lateral force and lift of the head car. The simulation results are shown in Table 1. Using the fine mesh as a reference, under the same operating conditions, the differences in side force on the head car for the coarse mesh and optimized mesh are 0.62% and 0.46%, respectively, while the maximum differences in lift are 8.02% and 3.60%, respectively. The difference in aerodynamic lift is relatively larger compared to side force because the lift magnitude is smaller, which amplifies numerical errors; however, the maximum differences remain within an acceptable range. To balance computational efficiency and numerical accuracy, the boundary layer mesh parameters of the optimized mesh are adopted as the reference in the following simulations. The thickness of the first layer of the mesh on the car surface is approximately 0.8 mm, with surface Y + values ranging between 30 and 150.

Table 1 Influence of different boundary layer grid parameters on the computational Results.
Full size table

Numerical simulation method verification

Feng Li24 conducted experiments using a 1:8 scale high-speed train model in a wind tunnel with dimensions of 15 m × 8 m × 6 m. The model consisted of a three-car formation, including a head car, a middle car, and a tail car. The aerodynamic drag coefficient of the train was calculated at an incoming flow velocity of 65 m/s. The high-speed train model is shown in Fig. 625.

Fig. 6
Fig. 6
Full size image

High-speed train experimental model25.

In this study, the aerodynamic drag coefficients of the head and tail cars obtained from numerical simulations under the same conditions were compared with the data from wind tunnel tests to verify the accuracy of the numerical calculations. The validation using a scaled wind tunnel model is based on the principle of aerodynamic similarity. Under similar conditions for the key dimensionless parameters, such as the Reynolds number and Mach number, the flow field distribution, pressure distribution, and aerodynamic characteristics of the scaled model can represent the aerodynamic behavior of the full-scale train. The consistency between simulation and experimental results under similar conditions indicates that the numerical model is reasonably set up and the computed results are reliable, providing a basis for further analysis and optimization of the train’s aerodynamic performance. The comparison between the computational and experimental results is shown in Fig. 7. As can be seen, the maximum error between the numerical simulation and experimental results is 5.6%, which meets the requirements of numerical simulation and confirms the feasibility of the simulation accuracy.

Fig. 7
Fig. 7
Full size image

Experimental validation of head and tail car drag coefficients.

Results and discussion

Train aerodynamic characteristics

When a train operates under crosswind conditions, it experiences significant overturning moments that tend to cause the train to tilt toward the leeward side. Therefore, studying the lateral force and overturning moment acting on the train is of great importance for its operation under crosswinds. The axis of the overturning moment is selected at the contact surface between the wheel flange on the leeward side and the rail. Under the influence of lateral wind, the train undergoes a tilting motion along the leeward rail direction due to the lateral rolling moment, which may cause the train body to lean toward the leeward side. Positioning the overturning moment axis at the contact surface between the leeward wheel flange and the rail more accurately reflects the actual situation. The coordinate system for the moment is shown in Fig. 8.

Fig. 8
Fig. 8
Full size image

Schematic of moment coordinates.

In the analysis of aerodynamic characteristics, there is a clear convention for defining lateral force. When a car-body is subjected to a force directed toward the leeward side, the resulting lateral force is considered positive; conversely, if the force is directed toward the windward side, the lateral force is negative.

As shown in Fig. 9, during train operation under crosswind conditions, trains with different fairing structures exhibit distinct and noticeable patterns in the lateral forces they experience. In the operation of a three-car trainset, the head car experiences a particularly significant aerodynamic lateral force, with values far exceeding those acting on the middle and tail cars. The lateral force on the head car is positive, indicating a force directed toward the leeward side. The middle car experiences a slightly smaller lateral force, ranking second in magnitude, and its force direction is consistent with that of the head car, with values only slightly lower. As for the tail car, it experiences the smallest lateral force compared to the head and middle cars, and its lateral force is negative, meaning the force is directed toward the windward side.

Because the fairing area is much smaller compared to the overall car-body area, there is a significant difference in the lateral force borne by the fairing, which is much smaller than that borne by the car-body. The lateral forces on the fairings are all positive. Among the two fairings, Fairing 1 experiences a greater lateral force than Fairing 2. Additionally, for trains equipped with two different fairing structures, the lateral force on the head car shows almost no difference. However, for trains using the folded-bellows inter-car fairing structure, the lateral forces on the middle car, tail car, and Fairing 2 are all smaller than those on trains with the capsule-type inter-car fairing. The specific lateral force values are presented in Table 2.

Table 2 Lateral force results on the train.
Full size table

Figure 10 provides a detailed illustration of the overturning moments experienced by trains with different inter-car fairing structures under crosswind conditions. Regarding the overturning moments on the head car, middle car, and tail car, the head car experiences the largest overturning moment, followed by the middle car, with the tail car experiencing the smallest. Meanwhile, Fairing 1 endures a greater overturning moment than Fairing 2. The specific values of the overturning moments are listed in Table 3.

Under the same operating conditions, the overturning moment experienced by the head car of the train with the folded-bellows inter-car fairing is reduced by 20.1%, that of the middle car by 23.5%, and that of the tail car by 27.7%. Since the surface of the capsule-type inter-car fairing is smoother, under crosswind conditions, the airflow easily rises along the windward side to the train roof, forming a high-speed zone, while a large and stable low-pressure area forms on the leeward side, resulting in a greater pressure difference. In contrast, the folded-bellows inter-car fairing features slight concave-convex folds, where local vortices and flow reattachment occur in the folds. This weakens the size and intensity of the low-pressure zone, reducing the pressure difference. Therefore, the overturning moment experienced by trains with the folded-bellows inter-car fairing is smaller than that experienced by trains with the capsule-type inter-car fairing.

Table 3 Overturning moment results on the train.
Full size table
Fig. 9
Fig. 9
Full size image

Comparison of side forces on trains with different fairing structures under crosswind conditions. (a) Car body lateral force, (b) Fairing side force.

Fig. 10
Fig. 10
Full size image

Comparison of overturning moments on trains with different fairing structures under crosswind conditions. (a) Car body overturning moment, (b) Fairing overturning moment.

Pressure distribution

During the operation of high-speed trains, the impact of different inter-car fairing structures on the pressure distribution on the train body is shown in Fig. 11. From the figure, it can be seen that on the windward side, the nose tip of the head car and the fairing areas exhibit distinct pressure distribution characteristics. The pressure on the fairing area of trains with the capsule-type inter-car fairing is greater compared to those with the folded-bellows inter-car fairing. In the smooth car-body regions, including the head car, middle car, and tail car, the pressures are roughly at the same level. The pressures on the windward side fairing areas are listed in Table 4, showing that the pressures on the folded-bellows fairings are lower than those on the capsule-type fairings.

In the pressure distribution on the leeward side, the pressures in the fairing areas differ. As shown in Table 5, in the Fairing 1 area, the pressure on the folded-bellows fairing is lower; in the Fairing 2 area, the capsule-type fairing experiences higher positive pressure, while the folded-bellows fairing experiences higher negative pressure.

For the pressure distribution along the longitudinal section of the train body, the overall pressure patterns for the two different fairing structures are basically consistent. The fairing pressures along the longitudinal section are listed in Table 6, where the folded-bellows inter-car fairing experiences higher pressure than the capsule-type, with a maximum pressure difference of 428 Pa.

Table 4 Pressure results on the windward side fairing area.
Full size table
Table 5 Pressure results on the Leeward side fairing area.
Full size table
Table 6 Pressure results on the fairing area along the train body longitudinal section.
Full size table

The results indicate that under crosswind conditions, for trains with full-coverage inter-car fairing structures, changing the fairing design mainly has a significant impact on the surface pressure in the fairing areas, while its influence on the head car, middle car, and tail car regions is relatively minor.

Fig. 11
Fig. 11
Full size image

Pressure distribution along the vehicle body profile. (a) Windward side pressure, (b) Leeward side pressure, (c) Vehicle roof pressure.

Flow field pattern around the train

The cross-sectional slices of the head car, middle car, tail car, Fairing 1, and Fairing 2 are extracted respectively, as shown in Figs. 12, 13, 14, 15 and 16. Subsequently, an analysis is conducted on the flow field distribution and pressure distribution around the trains with the two different inter-car fairing structures.

Fig. 12
Fig. 12
Full size image

Pressure and streamline diagram of the head car cross-section (Pa).

Fig. 13
Fig. 13
Full size image

Pressure and streamline diagram of the middle car cross-section (Pa). (a) Capsule-type inter-car fairing structure, (b) Folded-bellows-type inter-car fairing structure.

Fig. 14
Fig. 14
Full size image

Pressure and streamline diagram of the tail car cross-section (Pa). (a) Capsule-type inter-car fairing structure, (b) Folded-bellows-type inter-car fairing structure.

Fig. 15
Fig. 15
Full size image

Pressure and streamline diagram of fairing 1 cross-section (Pa). (a) Capsule-type inter-car fairing structure, (b) Folded-bellows-type inter-car fairing structure.

Fig. 16
Fig. 16
Full size image

Pressure and streamline diagram of fairing 2 cross-section (Pa). (a) Capsule-type inter-car fairing structure, (b) Folded-bellows-type inter-car fairing structure.

The analysis of the streamline diagrams reveals distinct regional characteristics of the flow field and consistent pressure distribution patterns around the train. For the head car, the pressure distribution is asymmetric. On the windward side, directly exposed to the incoming airflow, the pressure is positive; on the leeward side, due to flow separation and acceleration effects, the pressure is negative. Along the leeward side of the head car, the pressure transitions gradually from negative near the body to positive further away. This pressure gradient change relates to the airflow behavior as it moves around the front of the train. The middle car’s flow field exhibits different features, with pressure variations mainly concentrated on the car roof, forming a low-pressure region. This is due to the influence of the front and rear cars, resulting in a specific pressure distribution pattern in the local flow field. Around the tail car, the pressure distribution differs from the head car: the entire area shows negative pressure, which increases in magnitude closer to the body. This is mainly caused by the separation zone formed by the airflow at the rear of the train, where flow velocity decreases and pressure drops.

In the fairing regions, the negative pressure shows a specific spatial pattern, mainly concentrated in two areas: the upper part of the fairing and parts of the leeward side of the fairing. This indicates that the fairing structure has a significant impact on disturbing the local flow field.

Regarding vortex distribution, there is an increasing trend along the length of the train. The head car generates the fewest vortices, while the leeward area of the tail car forms larger vortices. As airflow moves along the train, vortex generation intensity gradually increases. When comparing different fairing structures, the pressure distribution around the main body of the train (including the head, middle, and tail cars) remains largely consistent. However, pressure fields near the fairings differ: with the folded-bellows fairing, the flow field pressure is higher in the Fairing 1 area; whereas in the Fairing 2 area, the capsule-type fairing exhibits higher pressure values.

The impact of train speed on aerodynamic performance

Under a crosswind speed of 15 m/s, the aerodynamic performance of trains with two different fairing structures was studied at varying train speeds.

Figure 17 shows the variation of the lateral force with train speed for trains with two different inter-car fairing structures. As shown, the lateral force characteristics of the leading car are basically the same for both structures. At 400 km/h, the leading car of the train with the capsule-type fairing experiences a lateral force of 58,233 N, while the leading car of the train with the folded-fairing structure experiences 58,304 N. When the train speed is 320 km/h, the maximum difference is 417 N. In contrast, the middle and trailing cars exhibit significant differences under the two fairing structures.

Comparing the folded-bellows and capsule-type structures, the folded-bellows fairing demonstrates better aerodynamic performance on the middle car, with the lateral force experienced by the middle car being consistently lower than that of the capsule-type fairing at different speeds. Similarly, the folded-bellows structure also shows advantages for the tail car. The lateral force on the tail car increases with speed for both fairing structures, but the folded-bellows structure maintains lower lateral forces compared to the capsule-type.

Fig. 17
Fig. 17
Full size image

Variation of lateral force with train speed. (a) Leading car, (b) Intermediate car, (c) Trailing car.

Figure 18 shows the comparison curves of the overturning moment experienced by the head car, middle car, and tail car under two different fairing structures as the train speed varies. It can be seen from the figure that, for the head car, middle car, and tail car alike, the overturning moment values with the folded-bellows structure are all lower than those with the capsule-type structure.

From the perspective of speed influence, the overturning moments under both structures show an increasing trend as the speed increases. The head car and tail car are more affected by speed changes than the middle car, indicating that the train’s front and rear ends are more susceptible to aerodynamic effects during high-speed operation.

Fig. 18
Fig. 18
Full size image

Train overturning moment vs. train speed. (a) Leading car, (b) Intermediate car, (c) Trailing car.

The impact of crosswind speed on train aerodynamic performance

At a train speed of 400 km/h, different crosswind speeds were selected to investigate the influence of crosswind speed on the aerodynamic performance of trains with two different inter-car fairing structures. Figure 19 shows the comparison curves of lateral force on the leading car, intermediate car, and trailing car under the two fairing structures as the crosswind speed varies. It can be seen that, under different crosswind speeds, the lateral forces on the leading car and intermediate car are generally consistent for both the folded-pleat type and capsule type structures, and their magnitudes increase with crosswind speed. The lateral force on the trailing car first decreases and then increases. Specifically, when the crosswind speed is between 5 m/s and 10 m/s, the lateral force on the trailing car decreases; when the speed increases to the range of 10 m/s to 25 m/s, the lateral force increases with crosswind speed. As the crosswind speed increases, the direction of the lateral force on the trailing car changes: when the crosswind speed is below 22 m/s, the force is directed toward the windward side; as the wind speed exceeds 22 m/s, the force direction gradually shifts toward the leeward side. When the crosswind speed is lower than 22 m/s, the trailing car of the folded-pleat fairing train consistently experiences smaller lateral forces than that of the capsule-type fairing train.

Therefore, in practical applications, when the crosswind speed in the operating environment does not exceed 22 m/s, adopting the folded-pleat inter-car fairing design can more effectively improve the aerodynamic performance at the train’s rear end.

Fig. 19
Fig. 19
Full size image

Variation of train lateral force with crosswind speed. (a) Leading car, (b) Intermediate car, (c) Trailing car.

Figure 20 shows the variation curves of the overturning moment of the leading car, intermediate car, and trailing car with crosswind speed for the two different inter-car fairing structures.

It can be seen that the overturning moments of the leading car and intermediate car increase with crosswind speed, and the trends for the two structures are generally consistent. For the trailing car, the overturning moment first increases and then decreases as the crosswind speed increases. Under the same crosswind speed conditions, the overturning moments of the train with the folded-pleat inter-car fairing are consistently smaller than those of the train with the capsule-type inter-car fairing.

Fig. 20
Fig. 20
Full size image

Variation of train overturning moment with crosswind speed. (a) Leading car, (b) Intermediate car, (c) Trailing car.

Discussion of train aerodynamic performance results

This study analyzed the aerodynamic performance differences between the capsule-type and folding canopy-type full-cover windscreen structures of high-speed trains under crosswind conditions through numerical simulation. By comparing with wind tunnel test data (1:8 scale model, yaw angle 15°), the maximum error of aerodynamic force coefficients in the numerical simulation was 4.00% (Table 2), meeting the engineering CFD accuracy standard (< 5%), which validates the effectiveness of the \(SST\) \(k - \omega\) turbulence model in capturing crosswind effects. The errors mainly arose from model simplifications (such as neglecting pantographs and other components), but these had limited impact on the prediction of global indicators like overturning moments. Despite the presence of mesh and model errors, the improvement trend of the folding canopy-type windscreen structure remains robust. The consistency in pressure distribution over the main body of the train (Fig. 11) further supports the generality of the conclusions.

Conclusion

By studying the aerodynamic characteristics of trains with two different full-cover windscreen structures under crosswind conditions, the following conclusions can be drawn:

  1. (1)

    Under crosswind conditions, the overturning moment on the folding canopy-type windscreen train is reduced by 20.1% for the head car, 23.5% for the middle car, and 27.7% for the tail car compared to the capsule-type structure.

  2. (2)

    For trains with capsule-type and folding canopy-type windscreens, the pressure distribution around the main body (including head, middle, and tail cars) remains basically consistent under crosswind conditions. However, differences appear near the windscreen areas, where the folding canopy-type structure improves the pressure distribution on the windward side of the train body.

  3. (3)

    Under crosswind conditions, the leading cars of trains with both capsule-type and folded-fairing inter-car structures experience nearly the same lateral force, regardless of changes in train speed or crosswind velocity. Under a 15 m/s crosswind and a train speed of 400 km/h, the lateral force on the leading car of the capsule-type fairing train is 58,233 N, while that on the leading car of the folded-fairing train is 58,304 N. Therefore, changing the inter-car fairing structure has little effect on the lateral force experienced by the leading car.

  4. (4)

    Compared with capsule-type inter-car fairings, folded-bellows-type inter-car fairings, by virtue of their corrugated structure, can realize gradual airflow guidance and local energy dissipation during aerodynamic interactions, thereby weakening the asymmetric vortices and lateral pressure gradients at the fairing junctions. As a result, they effectively reduce the peak lateral aerodynamic loads acting on the intermediate and tail cars, as well as the overall overturning moment of the train. This mechanism not only optimizes the distribution characteristics of aerodynamic forces along the train but also enhances its crosswind stability and operational safety.