Aerodynamic Behavior of Hump Slab Track in Desert Railways: A Case Study in Shuregaz, Iran

Abstract

The development of rail transport necessitates expanding environmentally friendly infrastructure. However, specific challenges arise in desert and sandy regions. One innovative solution to manage the effects of windblown sand on desert railways is the use of hump slab track superstructure. This paper develops a solid–fluid aerodynamic model based on ANSYS Fluent 2021 R2 software to simulate the hump slab track during a sandstorm. The model is validated through wind tunnel testing. A case study of a railway sandstorm in the Shuregaz region of Iran is presented, evaluating various sandstorm parameters and hump heights to determine their impact on sand concentration and particle velocity within the sand transit channels. The results indicate that increasing the sand particle diameter (from 150 to 250 µm) leads to higher sand concentration (up to 40%) and lower sand movement velocity (up to 28%). These results have been observed with a higher incremental approach concerning the sand flow rate. Conversely, increasing sandstorm velocity (from 10 to 30 m/s) decreases sand concentration and increases sand movement velocity up to 80% and 150%, respectively. Additionally, a 25 cm hump height significantly enhances sand passage by creating larger channels.

1. Introduction

Deserts cover a substantial portion of the Earth’s surface, posing challenges for establishing transportation infrastructure, particularly railway networks. Historically, nations such as France, England, and Germany developed railway systems in desert regions primarily for military purposes [1]. Examples include the Mecheria to Ain Sefra railway in the Algerian Sahara (constructed by France) [2], the Wadi Halfa–Abu Hamed railway in the Nubian Desert (built by England) [3], and the Aus–Lüderitz railway in the Namib Desert (established by Germany) [4]. In recent years, projects like the Shershah to Attock railway in Pakistan and the Dammam to Riyadh railway in Saudi Arabia have been initiated to mitigate the impacts of windblown sands [5].

Desert railways often face significant damage from windblown sand, resulting in high maintenance costs and various operational challenges [6]. Sand infiltration into the coarse-grained ballast materials leads to track solidification, increased dynamic forces during train passage, and accelerated component deterioration [7,8]. Additionally, windblown sand movement can cause track-bed clogging, reducing train speeds and posing derailment risks due to track closure [6].

Based on this, the development of railway monitoring in critical conditions has been expanded in the world [9] and several solutions have been proposed to manage and mitigate the effects of windblown sand [8,10,11]. One of the most innovative and effective measures suggested is the implementation of hump slab tracks [7], as illustrated in Figure 1. This approach entails transitioning from a ballasted to a non-ballasted track, thereby resolving the issue of solidification in the ballast layer. Additionally, the rails are elevated using concrete support structures called humps, raising the rail level. This design creates channels beneath the rails and between the humps, facilitating the free flow of sand. Consequently, the risk of track closure and sand accumulation, which may result in hazardous conditions, is diminished.

The concept of the hump slab track was first tested using computational fluid dynamics with truncated cone-shaped humps. The Eulerian method was used to determine the acceptable volumetric fraction of sand in the mixture. Sand movement was confirmed up to a height of 8 cm, with a volumetric fraction of 0.75 relative to the rail surface. The study in question showed that the designed system effectively responded only to a sand flood with a height of 8 cm [7].

A review of the technical literature on hump slab tracks reveals a significant gap in studies examining the impact of various sandstorm parameters and the geometric conditions of these tracks. Notably, there is a lack of assessments of critical sandstorm parameters such as the effective diameters of sand particles, mass flow rates of sand, and the speeds of sandstorms at different hump heights. Therefore, this paper presents numerical simulations of various sandstorms over the hump slab track. The new solid–fluid model developed based on ANSYS Fluent software is validated against a wind tunnel test. The investigation focuses on the impact of sandstorm parameters such as sand particle diameter, mass flow rate, windstorm velocity, and hump height. The paper examines sand accumulation in the inlet and outlet channels among the humps. To evaluate and compare sand accumulation in different models, values such as the Discrete Phase Model (DPM) concentration and average sand movement velocity (SMV) in the inlet and outlet channels are considered. The interpretation of the findings and discussion of results are addressed in subsequent sections.

2. The Study Region

2.1. The Shurgaz Location

The study region is located in the Shuregaz district, a subdivision of the Fahraj section of the Bam county. Due to its proximity to the Lut Desert, the largest desert in Iran, and the dry and semi-dry climate, the area is exposed to desertification, problems arising from the desert, wind erosion, the movement of flowing sands, and sandstorms. Hence, the residents of this county bear significant damages due to wind erosion, with studies indicating that over 311,000 hectares of land are severely affected [12]. Currently, a vast expanse of agricultural land, transportation roads, railway lines, and a rural population of 31,000, along with numerous facilities in the county, are situated in these crisis zones. As presented in Figure 2, a railway line is located exactly in the active dunes of the region. The occurrence of sandstorms, particularly the onslaught of flowing sands, is a major hindrance to the railway development in this region. Fahraj County, adjacent to Sistan and the Baluchestan province and the Lut Desert, is consistently exposed to severe and lengthy sandstorms lasting for about 120 days [13], impacting the entire region. The visibility for locomotive drivers on the Kerman–Zahedan railway near the corridor from Fahraj to Shuregaz, one of the busiest railway corridors in Iran, is reduced to less than one meter during sandstorms. This poses significant challenges and safety issues for the passage and operation of trains on this route for hours.

2.2. Key Specifications

The simulation parameters of the numerical model must account for two crucial factors: the prevailing wind conditions in the Shuregaz region and the distribution of soil particle sizes. The maximum wind speed in the area is significant as it affects soil erosion and the construction of windbreaks. Statistical data from the Bam station have been used to analyze wind conditions across various months. As demonstrated in Figure 3, the maximum wind speed in the region ranges from 20 to 38 m/s throughout the year, with most months experiencing maximum wind speeds of 20 to 25 m/s.

Soil samples were collected from the proposed site within the Shuregaz railway area and subjected to laboratory analysis to determine their particle size distribution (PSD). According to the results, particles with diameters ranging from 150 to 250 µm have the highest frequency (52.62%), as depicted in Figure 4.

Table 1. Specifications of the project area.

Parameter	Quantity
Region	Shuregaz
Average wind speeds in the year	20–25 m/s
Sand density	2700 kg/m3
Average sand diameter	0.0002 m

provides some of the essential characteristics of the project area.

3. Modeling Procedure

3.1. Governing Equations

In this paper, the gas–solid two-phase flow modeling with the Discrete Phase Model (DDPM) is used based on the Kinetic Theory of Granular Flow (KTGF). This method calculates interactions between particles using the KTGF, which operates similarly to the Eulerian–Eulerian approach. Therefore, the DDPM-KTGF method blends both Eulerian–Lagrangian and Eulerian–Eulerian techniques for modeling multiphase flows [14]. This method allows us to accurately model interactions between particles, as well as between particles and walls, while considering the volume fraction of the solid phase, unlike the standard Lagrangian discrete phase method. The airflow is described by the Navier–Stokes equations and particle trajectories are calculated using Newton’s laws of motion within the Lagrangian framework [15].

In this research paper, we implement a bidirectional coupling between the sand phase and the gas phase. The gas phase exerts drag forces on the sand particles, influencing their trajectories and velocities. Conversely, the presence of sand particles affects the gas flow through momentum exchange, which is accounted for by adding source terms to the Navier–Stokes equations. This bidirectional interaction is crucial for accurately simulating the dynamics of wind–sand two-phase flow.

3.1.1. Air Phase

In the DDPM-KTGF approach, the gas is considered as a uniform fluid phase, and a set of Navier–Stokes equations is applied for a numerical solution, following Equations (1) and (2) [16].

$${\displaystyle \frac{\partial }{\partial t}}\left({\alpha }_f{\rho }_f\right)+\nabla \cdot \left({\alpha }_f{\rho }_f{\overrightarrow{v}}_f\right)=0$$

(1)

$$\begin{array}{c} \\ {\displaystyle \frac{\partial }{\partial t}}\left({\alpha }_f{\rho }_f{\overrightarrow{v}}_f\right)+\nabla \cdot \left({\alpha }_f{\rho }_f{\overrightarrow{v}}_f{\overrightarrow{v}}_f\right)\\ =-{\alpha }_f\nabla P+\nabla \cdot {\stackrel{̿}{\tau }}_f+{\alpha }_f{\rho }_f\overrightarrow{g}+k_{fs}\left({\overrightarrow{v}}_f-{\overrightarrow{v}}_s\right)+\left({\overrightarrow{F}}_f+{\overrightarrow{F}}_{lift,f}+{\overrightarrow{F}}_{vm,f}+{\overrightarrow{F}}_{O,f}\right)\end{array}$$

(2)

In this context, ${\alpha }_f$ represents the gas volume fraction, ${\rho }_f$ denotes the gas density, ${\overrightarrow{v}}_f$ is the gas velocity vector, ${\overrightarrow{v}}_s$ is the sand velocity vector, $P$ indicates the gas–solid shared pressure within the system, ${\stackrel{̿}{\tau }}_f$ represents the stress–strain tensor for the air phase, $\overrightarrow{g}$ is the gravity vector, $k_{fs}$ is the momentum exchange coefficient between the air and continuum solid phases, ${\overrightarrow{F}}_f$ denotes the external volumetric force, ${\overrightarrow{F}}_{lift,f}$ is the lifting force, ${\overrightarrow{F}}_{vm,f}$ is the virtual mass force, and ${\overrightarrow{F}}_{O,f}$ indicates other forces.

In this paper, a significant aspect is the implementation of a distinct relationship governing the effective dynamic viscosity within the model framework. This adjustment is crucial for enhancing the efficacy of the conventional k−ε model, especially in scenarios characterized by low Reynolds numbers near the surface. Consequently, the adoption of the realizable k−ε model in our gas–solid modeling can expedite convergence and yield superior outcomes, thereby enhancing the accuracy of the modeling process [17,18,19]. The constants associated with the realizable k−ε model have been rigorously determined to ensure optimal model performance under specific standard flow conditions. These constants include $C_2=1.90-C_{1\epsilon }=1.44-{\sigma }_k=1.00-{\sigma }_{\mathrm{\varepsilon }}=1.20$ [16].

3.1.2. Sand Phase

In the DDPM-KTGF method, particles move within parcels using a Lagrangian description within a Eulerian framework (Equation (3)). The equation of motion is derived from Newton’s second law, incorporating particle-to-particle interactions addressed by the KTGF [16,20].

$${\displaystyle \frac{d}{dt}}{\overrightarrow{v}}_s=K_{fs}\left({\overrightarrow{v}}_f-{\overrightarrow{v}}_s\right)-{\displaystyle \frac{g\left({\rho }_s-{\rho }_f\right)}{{\rho }_s}}+{\displaystyle \frac{\nabla {\stackrel{̿}{\tau }}_s}{{\alpha }_s{\rho }_s}}-{\displaystyle \frac{\nabla P_s}{{\alpha }_s{\rho }_s}}$$

(3)

The described terms are as follows: $K_{fs}$ denotes the coefficient for inter-phase momentum exchange between solid and gas phases, ${\rho }_s$ represents the density of the solid phase, ${\alpha }_s$ signifies the averaging of the volume of particle parcels within a cell divided by the cell volume, ${\stackrel{̿}{\tau }}_s$ indicates the stress–strain tensor for the solid phase, and ${\rho }_s$ represents the pressure of the solid phase.

In our paper, the dynamics of sand particles in the wind–sand two-phase flow are governed by several forces, which are crucial for accurately capturing the behavior and movement of the sand particles within the flow. The forces considered in the model include Drag Force, Shear Lift Force, Pressure Gradient Force, Gravity Force, and Basset Force.

3.2. Hump Slab Geometry

For simulation purposes, the hump slab geometry considered has a width of 2.4 m and a length of 1.8 m. The design of the humps focuses on three main aspects. Firstly, create an adequate space between the supports to allow the passage of flowing sand. Secondly, maintain the aerodynamic profile of the humps to better withstand the flowing sand. Lastly, ensure that the humps establish a flat and smooth surface for the placement of rail fastening systems. Consequently, the CR geometric shapes, denoted as CR15, CR20, and CR25, with heights of 15, 20, and 25 cm, respectively, are selected, as illustrated in Figure 5.

3.3. Flow Type and Meshing

ICEM CFD is used for generating computational elements [21]. An example of the three-dimensional meshed space, incorporating the geometric model of the hump slab track, is illustrated in Figure 6. In this particular model, four distinct types of boundary conditions are established. These conditions encompass a bottom wall, symmetry on the ceiling, and perpendicular sides to the hump line. Additionally, inlet conditions are specified as velocity inlets while outlet conditions are set as pressure outlets (Figure 7). The rationale behind employing symmetry conditions is the presence of symmetrical air phase conditions in the mentioned directions. Moreover, wind profiles and sand particles enter the model through the inlet.

The Courant number corresponding to the minimum element size at the maximum wind speed is 0.6. Additionally, the Reynolds numbers at wind speeds of 10, 15, 20, 25, and 30 m/s are calculated as 82,150, 123,225, 164,300, 205,376, and 246,451, respectively [22]. The software is configured with the default “standard wall functions”. These functions impose wall boundary conditions on all turbulence model solution variables (k–ɛ), aligning with the logarithmic law for wind speed along the entire bottom wall of the domain, following the principles outlined by Launder and Spalding [23]. Employing a pressure-based solver, the method of solution is SIMPLE, utilizing implicit discretization. A critical factor in improving the mesh near the wall is the dimensionless wall distance (y+). For obtaining reasonably reliable results, the initial mesh point close to the wall should be within the log-law region, where the dimensionless wall distance needs to fall within an acceptable range (11.5–30 < y+ < 200–400) [24]. In this case, a wall distance of 150 mm was selected, which was within the acceptable range. This distance corresponds to a first-mesh height of 1.6 mm adjacent to the wall. For each analysis, particles with diameters of 150, 175, 200, 225, and 250 µm are introduced individually into the computational domain. Each time step is limited to a maximum of 400 iterations, with the convergence accuracy of the equations set to 0.00001.

3.4. Model Settings

The solver used for this problem was the pressure-based solver, operating in transient mode, taking into account a gravitational acceleration of 10 m/s². In the ANSYS Fluent database, the density of air is assumed to be 1.225 kg/m³, with a dynamic viscosity of 0.000017894 kg/m·s The sand particle phase can be specified as a separate entity within the software, with its density set to match that of the local sand, which is 2700 kg/m³.

The benchmark for measurement and comparison entails the sand flow rate (sand flux), quantified in kg/m²·s. The outlet point on the hump slab line at a height of 0.8 m is chosen for controlling the sand flux (Figure 8).

The results indicate that a simulation time of 2.5 s, considering the measured sand flux values at the outlet point along the hump slab line, is appropriate for analysis at wind speeds ranging from 10 to 30 m/s based on the reference point described in Figure 9.

The minimum and maximum element sizes of 0.0005 and 0.06 are taken into account, respectively, with an element ratio of 1.05, as shown in Figure 9. This is done to ensure finer elements are present near the bed, enhancing computational accuracy. Additionally, to assess sensitivity to the number of elements [25], simulations are conducted with three different sizes, namely coarse, medium, and fine elements, as specified in

Table 2. Various types of elements to measure sensitivity.

Coarse	Medium	Fine	Types of Elements
120,240	435,247	1,138,393	Number of elements

.

Subsequently, a comparison of mass flux changes is conducted concerning the heights of various elements, as illustrated in Figure 10. The findings indicate that there are no discernible changes when increasing the number of elements to 1,138,393. Therefore, it is determined that 435,247 elements are sufficient for the simulation.

Additionally, it is essential to explore the impact of the number of parcels in each injection (NOP) on the analysis results. The variations in mass flux over the computational domain are investigated for different numbers of injected parcels, specifically 100, 300, 500, 700, and 1000, at each time step, as shown in Figure 11. The range of sand flux oscillation within the 2.5 s time interval at the specified point in Figure 8 increases with the rise in the NOP. In the case of an NOP = 1000, the range of sand flux oscillation increases up to 0.22, while the minimum sand flux, equivalent to an NOP = 500, is recorded at 0.08.

The analysis involves the utilization of fifty-eight-core cluster systems, each equipped with 48 GB of memory. It takes approximately 1 to 2 days for each analysis using the available computers. The relevant geometric details and properties considered for the simulation are presented in

Table 3. Specifications used in this study.

Parameter	Quantity
Geometry Specification
Depth of the domain	1.8 m
Length of the domain	8.8 m
Height of the domain	1.6 m
Length of the hump slab track	1.8 m
Width of the hump slab track	2.4 m
Constants
Air density	1.225 kg/m3
Dynamic viscosity of air	0.000017894 kg/m s
Sand velocity at the inlet	0 m/s
Air temperature	298.15 K
Atmospheric pressure	101,325 Pa
Other specifications
Sand density	2700 kg/m3
Average diameter of the sand	0.0002 m
Sand injection rate	0.744 kg/s
Granular viscosity	Syamlal–Obrien
Granular bulk viscosity	Lun-et-al
Frictional viscosity	Schaeffer
The angle of internal friction	30°
Frictional Pressure	Based—ktg
Friction packing limit	0.61
Solid pressure	Syamlal–Obrien
Radial distribution	Syamlal–Obrien
Packing limit	0.63
Drag law	Gidaspow
Total analysis time	2.5 s

[25,26].

4. Validation of the Numerical Model

4.1. Wind Tunnel Test

To validate the software model simulating sandstorm passage over a hump slab track and assess its impact on the movement of solid particles in fluid mediums like air, controlled experimental testing was conducted at the Environmental Wind Tunnel of the Research Institute of Forests and Rangelands. This involved constructing a hump slab prototype (Figure 12a), implementing roughness elements for simulating wind flow (Figure 12b), creating a sand sampling structure (Figure 12c), and implementing a sand injection system (Figure 12d). The hump slab prototype consisted of humps and rails, with dimensions adjusted to fit the wind tunnel. Roughness elements, such as cubic and pyramidal structures, were used to establish a turbulent boundary layer. A sand sampler structure with 21 containers and glass tubes was designed for data collection at various heights. Additionally, a sand injection system was developed to introduce sand particles uniformly into the wind tunnel, simulating a sandstorm over the hump slab track. The outlined steps for experimenting are as follows [27]:

• The preparation of the laboratory and wind tunnel conditions for testing.
• The calibration of the wind tunnel velocity using equipment such as Pitot tubes, hot wires, etc.
• The placement of roughness elements inside the tunnel to create a logarithmic wind flow profile.
• The installation of an injection system at the top of the tunnel for introducing particles into the wind tunnel test chamber.
• The placement of the hump slab track at a specified distance inside the wind tunnel test chamber and, additionally, the preparation of the sample collector for gathering samples at various heights within the wind tunnel and deployment of the collector at a predetermined distance of 2 m from the obstacle (Figure 13).
• The installation of a hump slab track 1.6 m away from the sand injection point (Figure 12a).
• The arrangement, to enhance the accuracy of the sand flux curve at different heights in the wind tunnel, of the containers in parallel rows with a 3.5 cm gap between each row. This setup creates 21 data collection points along the height of the wind tunnel, reducing potential errors (Figure 12c).
• The continuous feeding of sand particles by a particle injection system into a 100 cm wide section of the wind tunnel (Figure 12d).

For each test, approximately 40 kg of sand particles, with a specific mass flow rate of 44,640 gr/min [28], was utilized under an optimized wind speed of 20 m/s for the hump slab track. The experiments were carried out continuously by introducing the tray of the sand and gravel particle injection system into the wind tunnel test chamber as illustrated in Figure 13. Subsequently, the collected sand particles were weighed using a precision scale with an accuracy of 0.1 g and their mass flow rates were determined along the vertical direction of the wind tunnel.

4.2. Numerical Simulation

For uniformity in particle size conditions used in the wind tunnel experiment and numerical simulation, the Rosin–Rammler distribution was employed. The Rosin–Rammler expression serves as a convenient means to represent the droplet size distribution. The entire spectrum of droplet sizes is discretized into a suitable number of intervals, with each interval characterized by a mean diameter for conducting trajectory calculations.

The default approach involves defining the particle size distribution by specifying diameters for the initial and final points and utilizing the linear equation (Figure 14) to adjust the diameter for each particle stream within the group. However, in cases where a distinct mass flow rate is desired for each particle or droplet size, the linear variation might not produce the desired distribution. In such instances, the particle size distribution can be more conveniently defined by fitting the size distribution data to the Rosin–Rammler equation [25]. The parameters related to fitting the particle size distribution of the study area based on the Rosin–Rammler equation are expressed as follows: $\overline{d}$ (mean diameter) = 236 µm, n = 18.26, maximum diameter = 4000 µm, minimum diameter = 64 µm, and number of diameters = 10.

To validate the numerical simulation results [29], the Eulerian model with a one-fluid phase for air properties and sand is employed. Additionally, the k–ɛ turbulence model in a mixture state is utilized for solving. The air density is considered as 1.225 kg/m³, and its dynamic viscosity is 0.000017894 kg/m·s⁻¹, in the wind tunnel test conditions. Moreover, the density of sand particles is assumed to be 2700 kg/m³. Based on the information available from the Shuregaz sand trap area, the sand grading of the region is applied in the software with an inlet sand rate of 0.744 kg/s, consistent with the conducted experiment. It is noteworthy that the sand rate is 0.744 kg/s, which is considered as the input for the software. At this rate, sand particles enter the model and are then carried by the wind until they reach their steady-state velocity approximately 1 m after the inlet.

In this simulation, considering a wind speed of 20 m/s as the reference in the wind tunnel and the applicable wind tunnel conditions, a boundary layer height of 12 cm is predicted. The comparison between the output of the developed model and the wind tunnel experiment reveals an approximate 10% disparity in the numerical results compared to the experimental findings. This validates the accuracy of the numerical simulation results (Figure 15).

5. Results and Discussion

To evaluate sand passage through channels, two planar surfaces aligned with the axis of the hump slab track are used. These surfaces pass through the line of symmetry of the rail. The surface plotted in the direction of the incoming sandstorm is designated as Plane 1 while the surface plotted in the direction of the outgoing sandstorm is labeled as Plane 2 (Figure 16).

Firstly, the simulation model including the project area’s parameters extracted from Table 1 is analyzed. In the next steps, simulations of sand particle movement around the hump slab track to determine the impact of various parameters on the railway track are evaluated. These evaluations include comparisons of DPM and SMV values at the centers of sand passage channels. Inlet channels are defined as pathways for sand to pass through gaps between humps in the direction of a storm. Similarly, outlet channels are defined on the opposite side. Therefore, a sensitivity analysis was conducted on sandstorm-related parameters such as the storm speed, sand diameter, and sand flow rate based on three hump heights.

5.1. The Performance of Different Humps

This investigation aims to analyze the variations in DPM and SMV values by considering the actual sand PSD of the study area. The objective is to understand how varying heights of humps affect the changes in DPM and SMV (as shown in Figure 17) at a wind speed of 25 m/s in the study area.

With an increase in the height of humps, the DPM values in the inlet and outlet channels decrease. Specifically, as the hump height increases from 15 cm to 25 cm, the DPM value decreases by approximately 40%. This results in a lower likelihood of sand deposition. On the other hand, the SMV values increase with the increase in hump height from 15 to 25 cm by approximately 57% in the inlet channels and 37% in the outlet channels. Therefore, based on the project area conditions, humps with a height of 25 cm provide better conditions for sediment accumulation and discharge compared to those of other heights.

5.2. Effect of Particle Diameter

The investigation initially involves analyzing the variations in DPM and SMV values for mean particle diameters of 150, 175, 200, 225, and 250 µm. The objective is to understand how different particle sizes influence the changes in the DPM and SMV in CR15, CR20, and CR25 models with wind speeds of 25 m/s. As presented in Figure 18, with an increase in particle diameter from 150 µm to 250 µm, the DPM values for both inlet and outlet channels show increases (i.e., 96% for CR15, 0.87% for CR20, and 49% for CR25). The results indicate that with an increase in the height of the humps, the DPM values for all diameters decrease. For example, in CR15 compared to CR25, the input and output DPM values for particles with a diameter of 200 µm decrease by approximately 15%.

Figure 19 also shows that increasing the diameter of the particle results in a reduction in the SMV value. For instance, in the CR25 model, as the particle diameter increases from 150 to 250 µm, the SMV values decrease by about 35% and 21% for the inlet and outlet channels, respectively. Similarly, the SMV values increase as the height of the hump increases from 15 cm to 25 cm. It is worth noting that with an increase in sand particle size, the difference between the SMV values in the inlet and outlet channels decreases. For a hump with a height of 25 cm, the difference between the outlet and inlet SMV values decreases from 21% at 150 μm to 4% at 250 µm.

5.3. Effects of Mass Flow Rate

This section analyzes the variations in DPM and SMV values for a mean particle diameter of 200 µm and mass flow rates ranging from 0.0066 to 10 kg/s. The focus is on comprehending the effects of different mass flow rates on CR15, CR20, and CR25 models at a wind speed of 25 m/s. As shown in Figure 20, there is a nonlinear relationship between flow rate and DPM values. For example, in the CR25 model, the sand flow rate increases by approximately 1500 times, but the DPM values show an increment of about 1700 times for the inlet channel and around 2400 times for the outlet channel. As the mass flow rate increases, there is a more pronounced sand accumulation in both the inlet and outlet channels. Among the different models, the CR15 model has the highest DPM values in the inlet and outlet channels while the CR25 model has the lowest values. On average, the differences in DPM values for outlet channels are about 65% between the CR20 and CR15 models and about 50% between the CR25 and CR20 models.

As depicted in Figure 21, the SMV values in both inlet and outlet channels exhibit a decreasing trend with the augmentation of the mass flow rate. For example, in the CR25 model, the sand flow rate increases by approximately 1500 times, but the SMV values show a reduction of about 40% for the inlet channel and around 32% for the outlet channel. Among the different models, the CR15 model has the lowest SMV values in the inlet and outlet channels while the CR25 model has the highest values. On average, the differences in SMV values for outlet channels are about 19% between the CR20 and CR15 models and about 10% between the CR25 and CR20 models.

5.4. Effects of Sandstorm Speed

The variations in DPM and SMV values for a mean particle diameter of 200 µm and sandstorm speeds ranging from 10 to 30 m/s are evaluated in this section. The focus is on understanding the impacts of different sandstorm speeds within the CR15, CR20, and CR25 models at an inlet mass flow rate of 0.744 kg/s. Figure 22 shows that an augmentation in the sandstorm speed leads to a reduction in DPM values. The findings demonstrate that as the sandstorm speed triples, the DPM values in the inlet channel diminish to around a quarter in the CR25 model while the DPM in the outlet channels undergoes an even more substantial decrease. However, at sandstorm speeds of 25 m/s and beyond, sand accumulation in the input and outlet channels stabilizes. Consequently, the speed of the sandstorm can be regarded as a contributory factor in facilitating the discharge of windblown sand from the inlet and outlet channels. With an increase in the height of the hump from 15 to 25 cm, the accumulated sand values in the inlet and outlet channels decrease, reducing the risk of sand deposition.

Figure 23 also indicates that the SMV values in both input and outlet channels increase in tandem with the elevation of the sandstorm speed. Nevertheless, at speeds of 15 m/s and below, the SMV values converge towards a uniform state. Moreover, with an upsurge in the sandstorm speed from 15 m/s to 30 m/s, the SMV values experience a notable surge in the inlet and outlet channels. In terms of height, as the height of the hump increases, the sand passage speed increases and the risk of sedimentation decreases significantly.

6. Conclusions

Sandstorms present a considerable challenge to rail transport in sandy regions, posing both financial and life-threatening risks. To address this issue, our study developed and evaluated a new method known as the hump slab track superstructure. This innovative approach involves raising the rail on concrete bases, referred to as humps, and eliminating the traditional ballast layer. Additionally, sand channels beneath the rail allow for the controlled passage of incoming sand away from the railway line.

Shuregaz in Iran is one of the regions where sandstorm speeds can reach up to 30 m/s. During the 120-day sandstorm season, thousands of tons of sand are displaced, causing significant damage to Iran’s railway infrastructure. However, the implementation of the hump slab track in this region appears to have the potential to significantly reduce the impacts of these sandstorms.

This paper aimed to validate a solid–fluid aerodynamic model using ANSYS Fluent. The model simulated a railway sandstorm in the Shuregaz region with computational fluid dynamics, employing the DDPM-KTGF method. The research focused on analyzing the effects of various sandstorm parameters as they interact with the hump slab track. To validate the model, wind tunnel tests were conducted. Additionally, a parametric study was carried out to assess different model specifications, including sand particle diameter, mass flow rate, and sandstorm velocity, while considering the presence of CR humps at different heights. The study analyzed DPM and SMV values within the channels designed for sand transit between the humps. The key findings of the research are as follows:

• Increasing the height of the humps up to 0.25 m decreases the risk of sand accumulation in the inlet channels, given the characteristics of sandstorms in the Shuregaz region. A higher hump height has a direct relationship with the improvement of the performance of the hump slab track system.
• Sand accumulation is minimal in sand inlet and outlet channels for particles with a diameter of 150 µm while sand discharge velocity is maximal. Therefore, the probability of particle sedimentation with this diameter in the inlet and outlet channels is lower compared to those with other particle diameters.
• Increasing the sand flow rate has a nonlinear and increasing impact on sand accumulation in the inlet and outlet channels. However, at a sand flow rate of 0.0066 kg/s, the effective discharge of the channels occurs more efficiently.
• Sand accumulation in the inlet and outlet channels significantly decreases with an increase in sandstorm speed from 10 to 30 m/s (on average 80% in CR25-15 models). However, the DPM values stabilize relatively at sandstorm speeds of more than 25 m/s. Moreover, the SMV values in the inlet and outlet channels noticeably increase with an increase in wind speed from 10 to 30 m/s.
• The consolidated results indicate that the implementation of hump slab track with a hump height of 25 cm can be considered a practical solution for critical desert rail areas. The application of this innovative slab track system in the Shuregaz region has demonstrated effective and sustainable performance against sandstorm passage through the superstructure section. Naturally, the use of this type of superstructure in other regions should be carefully examined and evaluated based on the prevailing conditions of the respective desert area.

7. Future Work

It is recommended for future research to consider investigating the impact of the wind angle on the hump slab track. It is also recommended to add corresponding wind and sand concentration distribution maps, pressure cloud maps, velocity cloud maps, etc. under different working conditions for qualitative analysis.