Enhancing Underground Thermal Environments in Cairo: The Role of Subway Entrance Geometry in Optimizing Natural Ventilation

Abstract

In rapidly urbanizing regions, enhancing passenger comfort in subway systems through sustainable methods is a critical challenge. This study introduces an innovative exploration of the impact of subway entrance geometry on natural ventilation and its subsequent effects on the thermal environment within Cairo’s subway system. The primary objective is to identify optimal entrance configurations that maximize natural airflow, thereby improving passenger comfort and reducing energy consumption. Focusing on the newly constructed segments of the Cairo subway, the research employs a mixed-methods approach that integrates computational fluid dynamics (CFD) simulations with a questionnaire survey to evaluate interactions between various entrance designs and urban wind flow patterns. This dual approach allows for a comprehensive assessment of how different geometrical configurations influence the capture and distribution of prevailing winds. The results indicate that specific entrance geometries can significantly enhance ventilation efficiency by optimizing wind capture and distribution. The most effective designs demonstrated substantial improvements in air quality and thermal comfort, providing practical insights for subway systems in similar hot arid climates. The novelty of this research lies in its detailed analysis of architectural elements to leverage natural environmental conditions for improving indoor air quality and thermal comfort in public transit systems. The significance of this study is its contribution to the field of sustainable urban transport, offering a valuable framework for urban planners and engineers. By demonstrating how thoughtful design can lead to energy savings and enhanced passenger experiences, this research advances the discourse on sustainable urban infrastructure. This work not only enhances theoretical understanding but also provides actionable recommendations for creating more sustainable and comfortable public transit infrastructures.

1. Introduction

Subway systems are integral to urban transportation networks, offering efficient and sustainable solutions for managing high-volume passenger transit [1,2]. As urbanization accelerates and populations swell, the demand for reliable and comfortable subway systems intensifies [3]. These systems serve as critical infrastructure, alleviating urban density pressures and seamlessly integrating into broader transportation networks, while facilitating the seamless integration of diverse transportation networks as strategic hubs. Despite extensive research on factors of thermal comfort and indoor environmental quality (IEQ) [4,5,6], there is a significant gap in understanding the role of natural ventilation, especially through the design of subway entrances, particularly in hot arid regions, with a specific focus on wind speed. These regions present unique challenges due to their extreme heat, aridity, and unique climatic conditions that impose significant obstacles to creating comfortable and healthy indoor environments for subway passengers [7]. This gap is particularly notable in studies focusing on the impact of wind speed on natural ventilation effectiveness, highlighting the need for targeted research in this area [8,9,10].

In hot arid regions, the combination of scorching temperatures and high humidity levels creates a hostile environment that can negatively impact passenger comfort and well-being. The confined and enclosed nature of subway stations further exacerbates these challenges, as they often lack natural ventilation and adequate cooling systems. Consequently, subway stations in such climates and environments frequently suffer from poor indoor air quality, elevated temperatures, and discomfort for passengers. Addressing these environmental conditions is critical not only for passenger satisfaction but also for their health, safety, and overall experience. Inadequate indoor environmental conditions can lead to a range of health issues, including respiratory problems, heat stress, and fatigue [11,12,13]. Additionally, passenger discomfort and dissatisfaction can result in reduced ridership, which can have economic and environmental implications.

To tackle these challenges, this research paper aims to investigate how the shape of subway entrances can enhance natural ventilation in subway systems of hot arid regions. To achieve these objectives, the study employs a combination of questionnaire analysis and simulation models, which are developed based on accurate representations of the subway station geometry, including platforms, carriages, and entrances/exits. While previous studies have explored thermal comfort and IEQ in subway systems [4,14,15], little attention has been given to the unique challenges and requirements of hot arid regions, as well as the impact of subway entrance design on natural ventilation. In addressing this, this study not only fills a significant gap in the current literature but also provides practical insights that could revolutionize the design and operation of sustainable subway systems in hot arid regions.

The contributions of this study are multifaceted: it provides a comprehensive analysis of how subway entrance geometries influence natural ventilation in hot arid regions; offers practical insights and design recommendations for enhancing natural ventilation in subway systems; and contributes to the field of sustainable urban transport by demonstrating how thoughtful design can lead to significant energy savings and enhanced passenger experiences. These aspects will be explored in detail in the subsequent parts of the research with full analysis.

The remainder of this paper is organized as follows. In the next section, a literature review section is presented, followed by the research area and methods in Section 3. Section 4 is the analysis and discussion of the findings. The conclusions and future implications are presented in Section 5.

2. Literature Review

2.1. The Ventilation

The critical role of efficient ventilation systems, encompassing both mechanical and natural methodologies, is paramount in ensuring thermal comfort within buildings, including subway stations [11,13,16]. They are designed to address multiple concerns—removing pollutants, providing fresh air, and maintaining a comfortable temperature for passengers. Mechanical ventilation systems, which include air conditioning units and exhaust fans, are a staple in modern subway systems due to their ability to rapidly alter air properties to achieve desired conditions. Studies such as those conducted by [11,13] have demonstrated the effectiveness of these systems in managing air quality and temperature, thus significantly improving passenger comfort in subway environments. Conversely, natural ventilation serves as an energy-efficient alternative, exploiting environmental elements like wind and thermal buoyancy to circulate air. This method has been highlighted for its potential to reduce energy consumption while still maintaining adequate air quality and thermal comfort [17,18]. The research in [18] emphasizes the viability of natural ventilation strategies in subway station design, particularly through the strategic placement of vents and openings to harness environmental conditions.

In the context of hot arid regions, the effectiveness of these natural systems becomes even more crucial. The integration of natural ventilation can significantly mitigate the severe thermal loads typical in such climates, offering a sustainable solution that aligns with energy conservation goals and enhances passenger comfort [16,19,20]. This literature backdrop sets the stage for a focused exploration of how subway entrance designs can be optimized to harness these natural ventilation capabilities more effectively in challenging arid climates.

The interplay between these ventilation methods and thermal comfort is complex, as thermal comfort is influenced by a myriad of factors, including air temperature, humidity, air velocity, and the metabolic heat produced by passengers [21,22]. Ref. [21] explored this interplay, noting that the optimization of ventilation systems requires a careful balance between these factors to achieve the desired thermal comfort levels without incurring prohibitive energy costs. Moreover, the integration of computational fluid dynamics (CFD) models has become instrumental in the design and evaluation of subway station ventilation systems [23,24]. Such models allow for the simulation of air flow patterns and temperature distribution within the station environment, providing valuable insights into the effectiveness of different ventilation strategies. The work in [25] showcases the application of CFD models in assessing the performance of ventilation systems in managing thermal comfort, highlighting their role in informing design decisions.

2.2. Factors Affecting Natural Ventilation

Natural ventilation, as a key component of sustainable building design, plays a crucial role in reducing energy consumption while ensuring indoor environmental quality and occupant comfort [16,26,27]. The architectural design of a building, including its orientation, layout, and facade design [28,29], significantly impacts the efficacy of natural ventilation. The building orientation determines the exposure to prevailing winds and solar radiation, which can enhance or impede natural airflow patterns within the building. A study in [16] underscores the importance of orientation and window placement in maximizing wind-driven ventilation, suggesting that strategic design can significantly improve air movement and reduce reliance on mechanical ventilation systems. Similarly, ref. [30] emphasizes the role of building shape and facade openings in facilitating cross-ventilation, highlighting the need for architects to consider natural ventilation principles in the early stages of design.

External environmental conditions, including wind speed, direction, and temperature gradients, are critical determinants of natural ventilation performance [31,32]. The variability of these conditions can affect the consistency and reliability of ventilation. Ref. [32] illustrates how urban microclimates, influenced by building density and urban morphology, can alter wind patterns and affect the natural ventilation potential of buildings. Furthermore, the study in [31] on passive ventilation emphasizes the significance of thermal buoyancy, driven by temperature differences between the indoor and outdoor environments, in promoting vertical air movement and enhancing ventilation effectiveness. These insights underscore the importance of considering both macro- and micro-environmental factors in the design of subway stations to optimize natural ventilation strategies.

Occupant behavior, including window operation habits and internal heat gains from appliances and occupants themselves, also influences natural ventilation [33,34]. Occupants’ preferences and their interaction with ventilation openings can significantly affect indoor air quality and thermal comfort. A study on residential buildings demonstrates that occupant behavior, particularly window opening practices, plays a pivotal role in achieving adequate ventilation rates [34]. Moreover, internal heat gains can create thermal gradients that either promote or hinder natural ventilation flows, as discussed in [33], who highlight the complex interplay between internal heat sources and natural ventilation strategies.

2.3. Influence of Wind Velocity

Wind velocity directly impacts the ability of natural ventilation systems to introduce fresh air and remove stale air from underground spaces [35,36]. The principle of natural ventilation leverages wind pressure differences around and within the station structure to drive airflow. Studies such as [36] have detailed how wind-driven pressure differences across station openings can significantly influence the rate of air exchange, highlighting the importance of understanding local wind patterns in station design. The configuration and orientation of subway station entrances, exits, and ventilation shafts play a pivotal role in harnessing wind velocity for natural ventilation purposes. Research utilized computational fluid dynamics (CFD) simulations to demonstrate how the strategic placement and design of station openings can enhance airflow patterns, optimizing the use of wind for natural ventilation, and thus, improving air quality and thermal comfort within the station [35].

The effectiveness of natural ventilation in buildings, particularly in relation to wind velocity, extends beyond environmental comfort to include aspects of safety and energy efficiency [37]. Ventilation strategies that effectively utilize wind velocity can reduce reliance on mechanical ventilation systems, leading to lower energy consumption and operational costs. This is supported by the findings of [37], who evaluated the energy-saving potential of natural ventilation in subway stations, emphasizing the dual benefits of enhanced passenger comfort and reduced environmental impact.

2.4. Ventilation Shafts

Traditional designs vary from simple vertical shafts to elaborate structures with multiple openings oriented towards prevailing winds [38,39]. Modern reinterpretations of these designs have embraced technological advances and computational modeling to further enhance efficiency and integration into the urban fabric. For instance, ref. [39] conducted simulations to evaluate the performance of wind catcher designs, indicating that certain shapes and orientations significantly improve airflow and cooling capabilities. In the context of modern architecture, and specifically subway stations, the application of wind catchers and ventilation shafts addresses both environmental sustainability and passenger comfort. The design of these systems considers factors such as wind direction, speed, and urban context to create efficient airflow pathways that enhance indoor air quality and thermal conditions [40]. A notable study in [40] demonstrates the incorporation of wind catcher technology into building designs to promote natural ventilation, illustrating the potential for these ancient concepts to be reimagined for contemporary challenges.

The integration of wind catchers and ventilation shafts into subway station design offers a promising avenue for reducing energy consumption while improving air circulation and passenger comfort [30,41]. The strategic placement and design of these elements can leverage natural wind patterns to facilitate air exchange, cooling, and ventilation in underground environments. Research in [30] on the use of wind catchers highlights the potential for these structures to significantly impact thermal comfort and air quality, underscoring their relevance in the design of public transportation systems, highlighting how they can optimize environmental control and enhance passenger experiences.

2.5. Entrance Design of Subway Stations

Another crucial element influencing natural ventilation in subway stations is the design of the entrance [42,43]. A seminal study in [43] on the aerodynamic effects of building openings revealed that the size and shape of entrances directly impact wind flow into buildings, suggesting that similar principles apply to subway station entrances. Their research underscores the importance of entrance design in facilitating effective ventilation, which is critical for managing the thermal environment in enclosed spaces. Further investigation specifically addressed the impact of entrance configurations on subway station ventilation [42]. Through computational fluid dynamics (CFD) simulations, they compared the ventilation efficiency of stations with a single large entrance to those with multiple smaller entrances. The findings indicated that multiple smaller entrances could promote more uniform air distribution throughout the station, enhancing the removal of heat and pollutants. However, the study also noted that the effectiveness of such designs is highly dependent on the station’s specific architectural and environmental context, including local wind patterns and urban morphology.

3. Materials and Methods

3.1. Study Area

This study focuses on the Cairo subway, one of the first completed subway networks in Africa and the Middle East. The study examines the natural ventilation performance of the new subway line (Green Line) that connects high-demand districts from Ataba to Rod El Farag. The Green Line passes through important areas such as KitKat, Zamalek, Embaba, and Gamet El Dowal, which have historical, cultural, and economic significance. The study also compares the Green Line with the existing north–south subway line (Line 1) and the east–west subway line (Line 3), which intersect at the Nasser Interchange Station, as shown in Figure 1.

The study area is the Nasser Interchange Station, which comprises two levels for different subway lines. The upper level, at a depth of 9 m underground, serves Line 1, while the lower level, at 27 m, serves Line 3. The Line 1 level has two large entrance halls with a total area of 1548 m², connected to the street level by 12 access points. The entrance halls have amenities such as escalators, ticket offices, administrative areas, and two 150 m long side platforms. Line 3 level has only two 150 m long platforms and train tracks, without any mezzanines or entrance halls. The station adopts a layered layout to optimize passenger flow, capacity, and safety by dividing the space for each line.

Wind Characteristics

Based on the climate and weather data simulated by Autodesk Ecotect [44] from Cairo International Airport, the wind rose analysis reveals the prevailing wind patterns and speeds in Cairo, Egypt, for different seasons. The wind rose diagrams illustrate that the dominant wind direction in all seasons is north (N), followed by northwest (NW), while the least dominant wind directions are south (S) and southeast (SE). Winds predominantly blow from the north to the south or northwest to southeast, accounting for about 70% of all seasonal hours, as shown in Figure 2. The wind speed varies depending on the wind direction and the season, but generally ranges from 1.5 to 5.5 m/s. The highest wind speeds are associated with the northerly and northwesterly winds, with an average of 4.5 m/s, while the lowest wind speeds are associated with the southerly and southeasterly winds, with an average of 2.0 m/s, as shown in Figure 3. The wind rose analysis provides useful information for designing buildings and urban spaces that can benefit from natural ventilation and passive cooling in Cairo, Egypt.

3.2. Natural Ventilation Questionnaire Survey

To assess the natural ventilation and air quality inside Jamal subway station, we distributed questionnaires. These explored users’ perceptions of natural ventilation, their satisfaction levels, factors affecting ventilation, and suggestions for improvement. The results indicated a general dissatisfaction among users with the current level of natural ventilation. This study evaluated the questionnaire findings using the Likert scale, the common method to assess people’s performance in these situations. The Likert scale used in this study features four stages that begin with perfect satisfaction with the level of ventilation. Natural ventilation is evaluated in declining order, with the last level being completely unsatisfied and gradually increasing. The final percentages in the sample are determined based on the diversity of the samples picked in the questionnaire.

To acquire the most realistic answers, the questionnaire was conducted via an internet survey as well as paper questionnaires distributed within the station. To maintain the validity of the questionnaire, it was essential to ensure that the participants understood its aim and content as much as possible while keeping the participant anonymous so that they felt more comfortable when filling out the questionnaire. Consequently, a simple and easy technique of posing questions was used to enhance communication with regular, non-specialized users so that they could easily understand the questions, as shown in

Table 1. Passenger comfort assessment scale for the thermal environment and air quality.

Gender
Male □	Female □
2. Age
Under 18 Years □	18–50 Years □	Above 50 Years □
3. Occupation
Student □	Government Employee □		Private Sector □	Retired □
4. Frequency of subway daily use
One Time □	Two Times □	Three Times □	Four Times □	Five Times □
5. How would you rate the overall natural ventilation inside subway station?
Very Inadequate □	Inadequate □	Adequate □	Good □	Very Good □
6. In your opinion, what affects natural ventilation the most?
Station Entrances □	Underground Depth □	Platform Layout □	Wind Speed Outside □
7. Do you think station entrance influence the flow of air inside?
Yes □	No □	Not Sure □
8. If yes, do entrance help or hinder air flow?
Help □	Hinder □	Not Effect □
9. How can natural ventilation be improved inside subway station?
Wider Entrance Areas □	entrance design Design □		Multiple Entrance Options □	Air Ducts or Fans □
Plants Near Entrances □	Expanding the Hall Area □

. It was confirmed that all participants used Gamal station every day before beginning to fill out the questionnaire. To prevent influencing the participant’s judgments and replies, the methods for filling out the questionnaire were supervised separately, focusing only on natural ventilation inside the station and suggestions for its improvement.

The points and questions in the questionnaire assess passengers’ feelings about natural ventilation inside the station, while human comfort and feelings are influenced by a variety of other climatic factors such as temperature, humidity, radiant temperature, and physical factors, including gender, thermal resistance of clothing, lighting, and physical activity [45]. However, none of these variables are covered in this study.

3.3. Urban Boundary and Entrance Geometry Simulation

This process is divided into two steps: urban boundary and simulation, and CFD evaluation of entrance designs. The first step involves creating a computational domain and mesh for the subway station and the surrounding urban area. The Autodesk CFD software 2024 the 64-bit version and meteorological data are used for this purpose. The second step involves simulating the natural ventilation performance of the station under different entrance designs. This step also utilizes the same software and the output data from the previous step. The results are then compared and analyzed to identify the optimal entrance design that maximizes the natural ventilation rate and air exchange efficiency. Additionally, recommendations are provided for improving the station entrance design and the indoor environment.

3.3.1. Urban Boundary and Simulation

Software and Computational Setup

The computational fluid dynamics (CFD) analysis in this study was performed using the 64-bit version of Autodesk CFD 2024, state-of-the-art simulation software known for its robustness and advanced fluid flow analysis capabilities. Autodesk CFD 2024 offers comprehensive tools for simulating complex aerodynamic environments, making it particularly suitable for modeling indoor spaces, urban microclimates, and natural ventilation scenarios. The software integrates a user-friendly interface with powerful solvers, enabling high-fidelity simulations of fluid–structure interactions. Key features of Autodesk CFD 2024 that were leveraged in this study include its adaptive mesh refinement, which dynamically adjusts mesh density around intricate geometries, and its solver accuracy settings that ensure convergence and stability in simulations involving multi-scale turbulence. Furthermore, the software’s support for steady-state and transient simulations allows for detailed temporal analysis, although this study focused on steady-state conditions to evaluate average ventilation efficiencies. The incorporation of advanced turbulence models, particularly the k-ε turbulence model, was crucial for accurately predicting turbulent flow characteristics around the subway station. Additionally, the computational setup benefited from the software’s capability to handle large-scale, high-resolution simulations efficiently, owing to its optimized parallel processing algorithms.

Potential and Methodology of the Software Application

The simulation process was as follows.

Model preparation: The three-dimensional model, created in AutoCAD 2024, a detailed grid covering the Gamal subway station and a 400 m × 400 m surrounding urban area, was created by conducting a laser distance survey to measure the dimensions of buildings and streets within a 500 m radius of the site, which was imported into Autodesk CFD 2024 as an SAT file. A grid independence study was conducted to determine the optimal grid size that ensured computational accuracy while maintaining efficiency. Material properties for external walls, floors, and fences were defined. Meteorological data, obtained every hour over an entire year from the nearby meteorological station, such as air movement, speed, and pressure variations were input through the boundary conditions list.

Simulation process: Settings were established for the simulation, including locational data (latitude, longitude, and time). The simulation process followed through the solve list, guided by described physical parameters as shown in Figure 4.

Boundary Conditions and Input Parameters

The boundary conditions and input parameters for the CFD simulations were carefully defined to accurately represent the urban environment and subway station characteristics. The computational domain was set with a velocity inlet boundary condition at the north face, using wind speed data collected from a nearby meteorological station. At the domain’s top and sides, symmetry boundary conditions were applied, while a pressure outlet condition was set at the downstream face and south face. The ground and building surfaces were treated as no-slip walls with appropriate roughness heights assigned based on material properties. The dimensions of the environment geometry were determined based on the recommended values of the simulation software, as shown in Figure 5.

For the subway station model, the entrance openings were defined as pressure outlets to allow for natural airflow. Internal surfaces of the station were modeled as adiabatic walls to focus on airflow patterns rather than heat transfer. The input parameters included air properties, which were density (1.225 kg/m³), dynamic viscosity (1.7894 × 10⁻⁵ kg/m·s), ambient temperature (37 °C; summer average), and atmospheric pressure (101,325 Pa), while turbulence parameters were wind speeds, ranging from 1.5 to 5.5 m/s, based on annual data, and wind major direction (north direction).

These boundary conditions and input parameters were consistently applied across all eight entrance geometry configurations to ensure comparable results and isolate the effects of entrance design on natural ventilation performance.

Mathematics

The mathematical model for this study is based on the fundamental equations of fluid dynamics, adapted for the specific conditions of wind-driven natural ventilation in a subway station environment. The governing equations include the continuity equation, momentum equations (Navier–Stokes equations), and energy equation, all solved simultaneously using the finite volume method within Autodesk CFD 2024.

For incompressible, steady-state flow, the continuity equation is expressed as

$${\displaystyle \frac{\partial \rho }{\partial t}}+{\displaystyle \frac{\partial \rho u}{\partial x}}+{\displaystyle \frac{\partial \rho v}{\partial y}}+{\displaystyle \frac{\partial \rho w}{\partial z}}=0$$

(1)

While $\rho$, the fluid density, is set to 1.225 kg/m³, $t$ refers to time, $u$, $v$, and $w$ are the velocity components in the x, y, and z directions, respectively, and $x$, $y$, and $z$ are the spatial coordinates. This equation represents the conservation of mass in a fluid flow. It states that the rate of change of density within a given volume must be balanced by the net flow of mass in or out of that volume. The continuity equation works in conjunction with the momentum equations (Navier–Stokes equations) and energy equation to provide a complete description of fluid flow in CFD simulations. It is particularly important for accurately modeling the airflow patterns created by different models.

The momentum equation is given by

$$\rho ({\displaystyle \frac{\partial \overrightarrow{u}}{\partial t}}+\overrightarrow{u}\cdot \nabla \overrightarrow{u})=-\nabla p+\mu {\nabla }^2\overrightarrow{u}+\rho \overrightarrow{g}$$

(2)

where $\rho$ is the fluid density, $u$ the velocity vector of the fluid, $p$ is pressure, $\mu$ is the dynamic viscosity of the fluid, which is set to 1.81 × 10⁻⁵ kg/(m·s), and $g$ is the gravitational acceleration vector. This equation represents the conservation of momentum for an incompressible fluid and accounts for the effects of pressure gradients, viscous forces, and body forces such as gravity.

The energy equation is described as

$$\begin{array}{c}\rho C_p{\displaystyle \frac{\partial T}{\partial t}}+\rho C_{p}u{\displaystyle \frac{\partial T}{\partial x}}+\rho C_{p}v{\displaystyle \frac{\partial T}{\partial y}}+\rho C_{p}w{\displaystyle \frac{\partial T}{\partial z}}={\displaystyle \frac{\partial }{\partial x}}\left[{\displaystyle \frac{\partial T}{\partial x}}\right]+{\displaystyle \frac{\partial }{\partial y}}\left[k{\displaystyle \frac{\partial T}{\partial y}}\right]+{\displaystyle \frac{\partial }{\partial z}}\left[k{\displaystyle \frac{\partial T}{\partial z}}\right]+q_V\end{array}$$

(3)

where $\rho$ is the fluid density, $c_p$ is specific heat capacity, $T$ is temperature, $t$ refers to time, $U$, $V$, and $W$ are the velocity components in the $x$, $y$, and $z$ directions, respectively, $k$ is thermal conductivity, and $q_V$ is the volumetric heat source term. This equation represents the conservation of energy in a fluid flow. It describes how temperature changes within a fluid due to convection, conduction, and heat generation.

To account for turbulence in the urban environment and subway station, the standard k-ε turbulence model was employed.

Turbulence Model

Turbulent kinetic energy ($k$) equation:

$$\begin{array}{c}\rho {\displaystyle \frac{\partial K}{\partial t}}+\rho U{\displaystyle \frac{\partial K}{\partial x}}+\rho V{\displaystyle \frac{\partial K}{\partial y}}+\rho W{\displaystyle \frac{\partial K}{\partial z}}=\\ {\displaystyle \frac{\partial }{\partial x}}\left[\left({\displaystyle \frac{{\mu }_t}{{\sigma }_K}}\right){\displaystyle \frac{\partial K}{\partial x}}\right]+{\displaystyle \frac{\partial }{\partial y}}\left[\left({\displaystyle \frac{{\mu }_t}{{\sigma }_K}}\right){\displaystyle \frac{\partial K}{\partial y}}\right]+{\displaystyle \frac{\partial }{\partial z}}\left[\left({\displaystyle \frac{{\mu }_t}{{\sigma }_K}}\right){\displaystyle \frac{\partial K}{\partial z}}\right]-\rho \epsilon +\\ {\mu }_t[2{({\displaystyle \frac{\partial U}{\partial x}})}^2+2{({\displaystyle \frac{\partial V}{\partial y}})}^2+2{({\displaystyle \frac{\partial W}{\partial z}})}^2+{({\displaystyle \frac{\partial U}{\partial y}}+{\displaystyle \frac{\partial V}{\partial x}})}^2+{({\displaystyle \frac{\partial U}{\partial z}}+{\displaystyle \frac{\partial W}{\partial x}})}^2+{({\displaystyle \frac{\partial V}{\partial z}}+{\displaystyle \frac{\partial W}{\partial y}})}^2]\end{array}$$

(4)

where $\mathit{\rho }$ is the fluid density, $K$ is the turbulent kinetic energy, $t$ refers to time, $\mathit{U}$, $V$, and $W$ are the velocity components in the x, y, and z directions, respectively, ${\mu }_t$ is turbulent viscosity, $\partial K$, with a value of 1, is the turbulent Prandtl number for K, and $\epsilon$ is the turbulent dissipation rate. The equation describes the transport of turbulent kinetic energy in a fluid flow. This equation is crucial for CFD simulations as it helps model the turbulent flow characteristics within the subway station, which directly impacts the effectiveness of natural ventilation. It allows for accurate prediction of air movement and mixing, essential for evaluating different entrance designs and their impact on ventilation performance.

Turbulent dissipation rate ($ϵ$) equation:

$$\begin{array}{c}\rho {\displaystyle \frac{\partial \epsilon }{\partial t}}+\rho U{\displaystyle \frac{\partial \epsilon }{\partial x}}+\rho V{\displaystyle \frac{\partial \epsilon }{\partial y}}+\rho W{\displaystyle \frac{\partial \epsilon }{\partial z}}\\ ={\displaystyle \frac{\partial }{\partial x}}[({\displaystyle \frac{{\mu }_t}{{\sigma }_{\epsilon }}}){\displaystyle \frac{\partial \epsilon }{\partial x}}]+{\displaystyle \frac{\partial }{\partial y}}[({\displaystyle \frac{{\mu }_t}{{\sigma }_{\epsilon }}}){\displaystyle \frac{\partial \epsilon }{\partial y}}]+{\displaystyle \frac{\partial }{\partial z}}[({\displaystyle \frac{{\mu }_t}{{\sigma }_{\epsilon }}}){\displaystyle \frac{\partial \epsilon }{\partial z}}]-C_2\rho {\displaystyle \frac{{\epsilon }^2}{K}}\\ C_1\mu {\displaystyle \frac{\epsilon }{tK}}[2{({\displaystyle \frac{\partial U}{\partial x}})}^2+2{({\displaystyle \frac{\partial V}{\partial y}})}^2+2{({\displaystyle \frac{\partial W}{\partial z}})}^2+{({\displaystyle \frac{\partial U}{\partial y}}+{\displaystyle \frac{\partial V}{\partial x}})}^2+{({\displaystyle \frac{\partial U}{\partial z}}+{\displaystyle \frac{\partial W}{\partial x}})}^2+{({\displaystyle \frac{\partial V}{\partial z}}+{\displaystyle \frac{\partial W}{\partial y}})}^2]\end{array}$$

(5)

where $\rho$ is fluid density, $\mathit{\epsilon }$ is turbulent dissipation rate, t refers to time, $U$, $V$, and $W$ are the velocity components in the x, y, and z directions, respectively, ${\mu }_t$ is turbulent viscosity, $\partial \epsilon$, with a value of 1.3 is the turbulent Prandtl number for $\epsilon$, $K$ is the turbulent kinetic energy, and $C_1$, set to 1.44, and $C_2$, set to 1.92, are model constants. This equation is essential for CFD simulations as it works in conjunction with the $K$-equation to model the turbulence characteristics of the airflow in and around the subway station. It helps in predicting the rate at which turbulent kinetic energy is converted into thermal energy at small scales, the distribution of turbulence intensity throughout the computational domain, and the energy cascade from large to small eddies in the turbulent flow.

Both equations are central to turbulence modeling in computational fluid dynamics, helping to predict the behavior of turbulent flows by accounting for the transport, production, and dissipation of energy within the flow. They are typically used in conjunction with the Navier–Stokes equations to simulate complex fluid dynamics scenarios.

Mesh Generation and Refinement

The mesh generation and refinement process for this study, which includes the Gamal subway station and the surrounding urban environment within an area of 400 m × 400 m, was conducted using the advanced meshing tools in Autodesk CFD 2024. An irregular grid was created, optimized through a comprehensive grid independence study, to balance computatonal accuracy and efficiency. The final mesh consisted of over eight million hybrid geometry cells, as shown in Figure 6, with objective enhancements applied to complex geometries such as building walls and corners, as shown in Figure 7. A multi-stage meshing strategy was employed, including initial coarse mesh generation, refinement in areas of high geometric complexity and expected flow gradients and boundary layer meshing with prism layers near walls, achieving minimum cell sizes of 0.05 m.

To ensure solution independence from the mesh resolution, a sensitivity study was performed. This involved creating multiple meshes with increasing cell counts (4 million, 6 million, 8 million, and 10 million cells) and comparing key output parameters such as average wind velocity at 10 m height of critical points within the station, as shown in Figure 8. The 8 million-cell mesh was selected as the optimal configuration, as it showed less than 5% variation in results compared to the other cell mesh while maintaining reasonable computational costs.

Special attention was given to refining the mesh around the entrance geometries and within the station interior to capture complex flow patterns accurately. The mesh quality was assessed using metrics such as skewness and aspect ratio, with all elements meeting the software’s recommended thresholds for accurate CFD simulations.

The percentage error margin was calculated for each mesh resolution, indicating the deviation from the weather station wind speed. For the 4 million-cell mesh, the error margin was 11.1%, suggesting a significant deviation. The 6 million-cell mesh reduced the error margin to 5.5%. The 8 million-cell mesh showed an error margin of 3.7%, demonstrating a more accurate and reliable result. The 10 million-cell mesh maintained the error margin at 3.7%, indicating that further increasing the mesh resolution did not significantly improve accuracy but did increase computational costs.

Limitations and Assumptions

The CFD simulations in this study, while comprehensive, were subject to certain limitations and assumptions that should be acknowledged. Firstly, steady-state conditions were assumed, which may not fully capture the dynamic nature of urban wind environments. The simulations treated air as an incompressible fluid, which is a reasonable approximation given the relatively low wind speeds involved but may introduce minor inaccuracies at higher velocities. The standard k-ε turbulence model was employed due to its robustness and computational efficiency, but it has known limitations in predicting flow separation and reattachment in complex geometries. This could potentially affect the accuracy of flow predictions in certain areas of the subway station. Additionally, the effects of humidity were neglected in the model, which might impact the results in scenarios where moisture content significantly affects air density or thermal comfort.

The study focused primarily on wind-driven ventilation, potentially underestimating the effects of thermal buoyancy, especially in low-wind-speed conditions. The simulations also did not account for the dynamic effects of train movements, which could influence airflow patterns within the station. Furthermore, the model assumed idealized surface roughness for urban structures, which may not fully represent the complexity of real urban environments.

Lastly, while the model was validated against meteorological data and through grid independence studies, full-scale experimental validation within the actual subway station was not conducted due to practical constraints. This limitation suggests that while the results provide valuable insights, they should be interpreted with caution when applied to real-world scenarios.

3.3.2. CFD Evaluation of Entrance Designs

By altering the entrance coverage’s shape, the methodology’s Section 3 simulated station ventilation. As shown in Figure 9, common shapes for wind catchers in Eastern countries such as Iraq, Afghanistan, Pakistan, and Iran include the rectangle, square, inclined rectangle, and inclined square [46]. Additionally, innovative wind catcher designs with curved shapes, such as the rectangular cylinder, square cylinder, inclined rectangular cylinder, and inclined square cylinder, have been tested [47]. These entrance designs were taken into consideration while maintaining the current proportions and sizes of the existing structures.. The cross-sectional size of the inlet cover opening corresponding to the external environment was standardized at 20 m² for each shape in order to ensure equal volume and air intake across various entrance designs. The goal of this standardization was to keep each inlet’s air intake constant.

A simulation model that faithfully reflected the geometry of the station, including the entrances, main halls, platforms, and internal structures, was made using the computational fluid dynamics (CFD) technique. Appropriate beginning conditions, boundary conditions, turbulence models, and numerical techniques were all included in the CFD model. Every entrance design was simulated, and the results were compared to the entrance’s existing state, which is uncovered as a baseline. Throughout the simulations pertinent data were tracked and recorded, including wind velocity, airflow patterns, and other relevant characteristics.

Based on the comparative analysis, the ideal entry covering design with the greatest improvement in natural ventilation was chosen, considering factors such as airflow patterns, air exchange rates, passenger comfort, and practical feasibility. Recommendations for applying the chosen entrance coverage design to improve natural ventilation inside the subway station were presented. The purpose of this part is to figure out the shape of the entrance that gives the best natural ventilation and superior air quality to all passengers who use the station.

4. Results and Discussion

The results of the user survey and computational fluid dynamics simulations of the surrounding urban environment and inside the station were analyzed in detail. By examining the data, charts, and simulation program outputs, this research aimed to assess both current conditions and potential improvements. The aim is to develop a clear and precise understanding of the recommendations that should be implemented going forward. Each component of the results is addressed separately, in addition to its impact on other aspects, as follows.

4.1. Questionnaire User Insights for Natural Ventilation

The questionnaire results provide a quantitative assessment of the user perceptions and preferences on natural ventilation inside Gamal station. The questionnaire survey was conducted over a period of five weeks, during which 420 questionnaires were administered online and in paper form to the users of the station. Of these, 372 valid questionnaires were included in the analysis, while the remaining were rejected due to invalidity or incompleteness. Among the respondents, the majority consisted of males (65%) (with females, 35%), with a significant portion (51%) falling within the age range of 18 to 50 years. Furthermore, 40% of the surveyed individuals indicated that they use the subway system twice daily. These demographic characteristics reflect the typical profile and ratio of subway users in Cairo.

For questionnaire data validation, the demographic data and usage patterns of the respondents align with the published literature on urban transit user profiles in Cairo. For instance, ref. [48] documented similar demographic patterns in urban transit studies in Cairo, providing a reliable basis for further analysis. Such alignment validates the representativeness of the sample, ensuring that the insights gained are broadly applicable.

The overall natural ventilation rate inside the Jamal subway station was rated as inadequate by 35% of the respondents, adequate by 38%, and good by 27%. Only 2% of the respondents rated it as very good, and 5% as very inadequate. This indicates that there is a problem among users and their lack of sufficient satisfaction with the amount of natural ventilation. According to 48% of respondents, station entrances most significantly influence natural ventilation, followed by external wind speed (33%), platform layout (11%), and underground depth (8%). This suggests that the design and shape of the station entrances play a crucial role in enhancing or hindering the airflow inside the station. The majority of the respondents, 89%, agreed that station entrances influenced the indoor air flow. Among them, 78% said that entrances facilitated airflow, while 22% said that entrances hindered it. This implies that some users perceived the entrances as a positive factor, while others perceived them as a negative factor, depending on the location and direction of the entrance.

To improve natural ventilation, 38% of respondents suggested wider entrances, 30% favored multiple entrance options, and 22% recommended redesigning entrance designs. Only 6% opted for air ducts or fans, and 4% for expanding the hall area. None of the respondents suggested plants near entrances as a possible solution. The results are presented in

Table 2. Results of questionnaire survey.

Basic Personal Information					Occupation					Frequency Daily Use
			No.	%			No.		%		No.	%
Gender	Male		240	65	Student		86		23	One time	130	35
	Female		132	35	Gov. emp.		110		30	Two times	150	40
Age	<18		22	6	Private sector		170		46	Three times	80	21
	18–50		306	82	Retired		6		1	Four times	10	3
	>50		44	12						Five times	2	1
Total			372	100.0			372		100.0		372	100
Overall Natural ventilation Rate			Effects on Natural Ventilation				Entrance Airflow			Improvement
	No.	%			No.	%		No.	%		No	%
Very inadequate	20	5	Station entrances		179	48	Yes	330	89	Wider entr.	140	38
Inadequate	110	30	Underground depth		30	8	No	30	8	Entr. shape	80	22
Adequate	140	38	Platform layout		40	11	Not sure	12	3	Multiple entr. options	110	30
Good	100	27	Wind speed outside		123	33	Total	372	100	Air ducts/fans	25	6
Very good	2	1					Entrance help airflow			Plantation	0	0
							Help	257	78	Expanding hall	17	4
							Hinder	73	22
Total	372	100			372	100	Total	330	100		372	100

and Figure 10, with descriptive statistics and percentages.

These results show that the natural ventilation and air quality inside the Gamal subway station are not satisfactory for most users and that station entrances are the main factor affecting airflow inside the station. Users’ insights align with the literature, which shows that design features can enhance natural ventilation rates and efficiency by increasing wind velocity. For example, refs. [42,49] have demonstrated that the architectural design of entrances significantly impacts airflow patterns and ventilation efficiency in enclosed spaces.

Changing the shape and design of the entrance covers to act as wind catchers was adopted as a cost-effective solution. This approach is supported by studies such as those in [50], which highlight the effectiveness of passive ventilation strategies in improving indoor air quality. In contrast, modifying the width of the entrance and introducing multiple entrances would be economically costly in existing stations but feasible for new stations under construction. This aligns with findings in [51], which noted the higher cost implications of structural modifications in existing infrastructure compared to new constructions.

By leveraging these insights, it becomes clear that optimizing entrance design is a critical factor in enhancing natural ventilation and thermal comfort for subway users. Future station designs should incorporate user feedback and proven architectural strategies to achieve better air quality and user satisfaction.

4.2. Simulated Station Boundary: Local Wind Flows

CFD simulations under the conditions of the northern winds provided a valuable idea about the prevailing wind patterns across the study area, as shown in Figure 11. The streamlines with the highest speeds approached the station complex with an average speed of 4.5 m/s along the streets parallel to the wind direction. Around the building structures, the side streets perpendicular to the wind direction had air speeds ranging from 0.5 to 2.5 m/s. Separation bubbles and wake areas formed on the windward faces, where the speeds decreased by 1 to 2.5 m/s compared to the non-turbulent flows.

The CFD model has been validated using data from the published literature to ensure accuracy. The input data for the simulation included wind speed and direction, building geometry, and surrounding topography, which are consistent with those used in similar studies [52,53,54]. The operating conditions, such as atmospheric pressure and temperature, were also matched to those documented in the literature. The model’s results showed close alignment with the findings of these studies, with deviations within acceptable ranges, primarily due to local climatic variations and structural differences. For instance, ref. [55] observed similar patterns of wind speed reduction and turbulence near building structures. These consistencies validate the reliability of the CFD model used in this study.

In front of the station entrances, the wind speeds varied according to the direction of the entrance relative to the prevailing winds. Entrance 10 witnessed speeds ranging from 2.5 to 3.2 m/s as it was the entrance to the station hall receiving outdoor air most directly, while the other entrances located on the west side of the main station hall witnessed speeds of about 0.5 and 1 m/s at entrances 8 and 9, respectively, and 1.5 m/s at entrance 11. The direction of the entrance was opposite the wind direction, making it not receptive to any external winds. Based on these findings, entrance 10 received the most natural outside air according to the prevailing winds identified earlier, while the rest of the entrances required more costly architectural solutions to make them face the external wind direction and improve their reception of the outdoor air. Therefore, the impact of this entrance on natural ventilation in the main station hall will be evaluated in the subsequent studies in this research.

The results of this study provide valuable insights into optimizing the design of subway station entrances to enhance natural ventilation. Several key points emerge from the findings:

The orientation of the entrances significantly affects their ability to capture external winds. Entrance 10, which faces the prevailing northern winds, demonstrated the highest wind speeds, highlighting the importance of strategically positioning entrances to align with dominant wind directions. This finding is consistent with previous studies, such as [33], which emphasize the role of entrance orientation in natural ventilation.

The variation in wind speeds across different entrances suggests that simply having multiple entrances is insufficient; their placement and orientation are crucial. Entrances that do not face the prevailing winds, such as entrances 8 and 9, exhibited significantly lower wind speeds, indicating reduced effectiveness in enhancing natural ventilation. This aligns with the work in [56], which found that building geometry and entrance orientation critically influence ventilation performance.

For existing structures, architectural modifications such as the addition of wind catchers or redesigning entrance covers to direct airflow can be effective solutions to enhance natural ventilation. Studies like that in [57] support this approach, showing that such modifications can improve airflow patterns without requiring extensive structural changes.

While repositioning entrances or adding new ones can be economically challenging for existing stations, it is a feasible option for new constructions. Incorporating optimal entrance designs from the planning stage can ensure better natural ventilation and reduce reliance on mechanical systems, ultimately lowering operational costs. This cost–benefit analysis is supported by the findings in [20,58], which highlighted the long-term economic advantages of integrating natural ventilation strategies in building design.

The practical implications of these findings for subway station design include optimized entrance design for new subway stations; designing entrances to face prevailing winds can significantly enhance natural ventilation. This strategy should be integrated into the initial design phase to maximize effectiveness. For existing stations, cost-effective architectural modifications, such as wind catchers or redirected entrance covers, should be considered. can improve natural ventilation without major reconstruction; also, urban planners and architects should consider local wind patterns and station layout during the design and renovation processes to optimize airflow and enhance passenger comfort.

4.3. Entrance Study

This Section 4 presents the CFD simulations that were performed to assess the effect of different entrance geometries on wind speeds and natural ventilation in the subway station. The simulations compared the ventilation performance of various entrance configurations by analyzing the flow patterns and natural ventilation characteristics inside the station. The model used a 3D representation of the station to capture the complex interactions between the inlet shape and the surrounding airflow. The natural ventilation in the subway station depended on several factors, such as building type, local climate, spatial configuration, orientation, location, and positioning [59,60]. Therefore, the entrance forms had diverse impacts on the natural airflow within the station [42,61]. The CFD analyses measured the wind speeds at ten points along the station, from outside to the entrance, stairwell, and platform, for selected inlet designs and the existing opening. The wind speed measurements are shown in

Table 3. Wind velocities at each point (m/s).

Shape	Existing	Rectangle	Square	Rectangle Cylinder	Square Cylinder	Inclined Rectangle	Inclined Square	Inclined Cylinder Rectangle	Inclined Cylinder Square
P1	2.28	0.99	1.02	1.17	1.09	1.15	0.97	1.13	1.00
P2	2.06	1.01	0.86	0.89	0.93	1.04	0.91	1.02	1.00
P3	1.75	1.55	1.33	1.49	1.47	1.56	1.32	1.54	1.51
P4	1.83	1.95	1.70	1.90	1.84	2.00	1.65	1.95	1.83
P5	1.73	2.05	1.77	2.00	1.97	2.13	1.70	2.04	1.89
P6	1.44	1.78	1.52	1.71	1.61	1.84	1.44	1.80	1.66
P7	1.25	1.67	1.39	1.64	1.51	1.74	1.33	1.66	1.52
P8	1.02	1.27	1.07	1.25	1.16	1.33	1.04	1.29	1.18
P9	0.83	0.90	0.83	0.91	0.95	0.99	0.90	0.99	0.99
P10	0.67	0.75	0.64	0.72	0.80	0.75	0.77	0.74	0.85

.

Wind speed significantly influences the station’s natural ventilation, dictating air exchange rates and indoor air quality [62,63]. Insights into how localized flow velocities at sampling points throughout the enclosed space are modified by each entrance geometry are provided by results from the CFD modeling. Quantitative and qualitative assessment of the capacity of diverse inlet profiles to enhance cross-ventilation performance targeting thermal comfort and airborne contaminant dilution objectives is enabled by computational analyses. Natural conditioning of interior environments for occupant health and sustainability goals through passive means can be maximized by optimization of the entrance design.

4.3.1. Flow Analysis by Entrance Design

By evaluating local flow characteristics, the results showed how the design of the subway station entrance influenced wind velocity measurements and natural ventilation across different configurations. The existing open entrance had the highest front velocities, exceeding 2 m/s, but they quickly dropped to less than 1 m/s behind point 5, indicating uneven air distribution that could affect the comfort of the passengers. The rectangular entrance had slower front velocities of about 1 m/s, but they increased towards the middle points to exceed 2 m/s, indicating better distribution behind the middle station than the current situation. The square entrance had velocities mostly between 1 and 1.7 m/s, with a short rise only at point 5, indicating weak air flow renewal overall. Both cylindrical designs had higher average velocities, close to or equal to 2 m/s between points 4 and 5, but the rear velocities followed the direction of the existing design’s decline. The inclined rectangle had the optimal result, with faster average speeds above 2 m/s while keeping front speeds above 1.1 m/s, and the best capture and dispersion station for the wide winds, as shown in Figure 12. The inclined square shape had velocities slightly less than 1 m/s at the beginning with a noticeable rise in the middle, reaching 1.7 m/s, and then, decreasing gradually until reaching the lowest levels close to the current situation. The inclined cylindrical rectangle and square had velocities around 1 m/s at the beginning, with a noticeable increase at point 5, and then, a decrease again until reaching about 0.8 m/s.

The quantitative evaluation of local wind speeds provided insights to prioritize entrance designs that maintained higher uniform speeds to ventilate the occupied areas appropriately and maintain thermal comfort for passengers, with some designs deserving closer study.

Based on the performed parametric studies, which analyzed wind velocity at different points for various entrance shapes, the results can be effectually utilized in machine learning-based, deep learning-based, and artificial intelligence-based methods in future studies. By appropriately training on the raw data obtained from these studies, optimal design solutions can be derived. The study in [64] demonstrated the application of machine learning in solving structural mechanics problems, highlighting its potential to streamline complex calculations and improve accuracy. Additionally, ref. [65] provided a comprehensive survey on deep learning applications, architectures, models, tools, and frameworks, emphasizing how these advanced techniques can be applied to various engineering challenges to enhance performance and efficiency. Another study [66] introduced a novel tool condition monitoring system based on a Gramian angular field and comparative learning, showcasing the practical applications of deep learning in real-time monitoring and predictive maintenance. Furthermore, ref. [67] developed a predictive model using a multilayer perceptron network based on normalized histograms, which can be adapted for optimizing yield predictions in engineering designs. By integrating these advanced computational methods, the design of the underground structure can be further optimized, leading to significant reductions in computational effort and the development of more efficient and robust solutions.

4.3.2. Entrance Designs Comparison

Wind speed measurements from seven internal points of the subway station model were analyzed for different entrance geometries. The measurement points, numbered four through ten, were located within the station, excluding the first three points at or near the entrance.

The average wind speeds were calculated at these points, including standard deviation—a measure of variation from the average—and the coefficient of variation (CV), indicating the variation’s relative size to the mean. Additionally, the improvement percentage in natural ventilation inside the station for each design was determined relative to the existing case, as shown in

Table 4. Comparison between different shapes’ improvements.

Shape	Average Wind Velocity (m/s)	Standard Deviation (m/s)	Improvement Percentage (%)	Coefficient of Variation (%)
Existing	1.25	0.35	____	28
Rectangle	1.48	0.46	18.52	31.08
Square	1.27	0.33	1.74	25.98
Cylinder Rectangle	1.45	0.41	15.64	28.27
Cylinder Square	1.41	0.39	12.38	27.65
Inclined Rectangle	1.54	0.49	23.07	31.81
Inclined Square	1.26	0.31	0.83	24.60
Inclined Cylinder Rectangle	1.49	0.48	19.36	32.21
Inclined Cylinder Square	1.42	0.43	13.17	30.27

.

$$\begin{array}{c}\overline{\mathrm{v}}={\displaystyle \frac{\sum _{\mathrm{i}=1}^{\mathrm{n}}{\mathrm{v}}_{\mathrm{i}}}{\mathrm{n}}}\end{array}$$

(6)

$${\mathrm{I}}_{\mathrm{s}}={\displaystyle \frac{{\overline{\mathrm{v}}}_{\mathrm{s}}-{\overline{\mathrm{v}}}_{\mathrm{e}}}{{\overline{\mathrm{v}}}_{\mathrm{e}}}}\times 100$$

(7)

$$\mathsf{\sigma }=\sqrt{{\displaystyle \frac{\sum _{\mathrm{i}=1}^{\mathrm{n}}({\mathrm{v}}_{\mathrm{i}}-\overline{\mathrm{v}}{)}^2}{\mathrm{n}}}}$$

(8)

$$\mathrm{C}\mathrm{V}={\displaystyle \frac{\mathsf{\sigma }}{\overline{\mathrm{v}}}}\times 100$$

(9)

where $\overline{\mathrm{v}}$ is the average wind speed, $\mathrm{n}$ is the number of wind speed values, ${\mathrm{v}}_{\mathrm{i}}$ the wind speed value of the point, ${\mathrm{I}}_{\mathrm{s}}$ is the improvement percentage, ${\overline{\mathrm{v}}}_{\mathrm{s}}$ is the average wind speed of the shape, ${\overline{\mathrm{v}}}_{\mathrm{e}}$ is the average wind speed of the existing entrance, $\mathsf{\sigma }$ is standard deviation, and $\mathrm{C}\mathrm{V}$ is the coefficient of variation.

Through analyzing the previous findings and simulation results shown in Figure 13 and Figure 14 straight entrance designs and their effectiveness in improving natural ventilation inside subway stations were determined in the first part as follows.

The existing entrance fell below the recommended 1.5 m/s wind speed threshold [68], exhibiting an average internal speed of merely 1.25 m/s. This signifies insufficient capture of the outdoor wind of 2.7 m/s for effective ventilation, highlighting the critical role of entrance geometry in harnessing wind pressure coefficients for natural ventilation in buildings [69]. Additionally, the significant influence of wind pressure coefficients and opening design on maximizing natural ventilation potential in similar subway station environments has been well documented [70].

To validate these results, the CFD simulation data were compared with the published literature. The input data, including wind speed, direction, and geometric parameters of the entrance, were similar to those described in studies in [71,72,73]. The findings showed similar trends, with minor deviations primarily due to specific local environmental conditions and structural differences in the station designs.

Furthermore, the low coefficient of variation (CV) of 28% suggests an even distribution of airflow but potentially lacks the necessary turbulence for proper air mixing. This finding aligns with [31] who emphasized the importance of turbulence intensity for contaminant removal and thermal comfort within buildings. Adequate turbulence is crucial for removing contaminants and maintaining thermal comfort [69], necessitating designs that encourage airflow mixing within the station to address potential stagnant zones and improve overall air quality. Uneven ventilation with wind speeds below 1 m/s in rear platform areas raises concerns for air quality and thermal comfort, the negative impacts of inadequate ventilation in public transportation systems [74]. Their study linked inadequate ventilation to increased pollutant concentrations and occupant discomfort due to thermal imbalances. Addressing uneven ventilation is crucial for ensuring a healthy and comfortable environment for station users, as highlighted by ASHRAE [75].

The rectangular entrance design demonstrated promising results for enhancing natural ventilation within the subway station, aligning with research on the effectiveness of rectangular geometries for wind capture and airflow channeling [76]. CFD simulations using an outdoor wind speed of 2.7 m/s and a standardized inlet opening of 20 m² revealed an average internal wind speed of 1.48 m/s for this design. This approaches the recommended threshold of 1.5 m/s for effective ventilation in public spaces.

To validate our results, we compared them with the published literature. Our findings align with [16], which reported a 15–20% improvement in ventilation rates for rectangular inlets towards the building. In this case, a 18.52% improvement is observed over the baseline, corroborating their results, lending credibility to our simulation methodology.

However, the standard deviation of 0.46 m/s and coefficient of variation (CV) of 31.08% indicate significant local variations in wind speeds. This aligns with observations in [77], in which it was noted that rectangular geometries can sometimes lead to uneven airflow distribution. The discrepancy between our results and the ideal uniform distribution can be attributed to complex interactions between the incoming wind and the station’s internal geometry, limitations in our CFD model’s ability to capture small-scale turbulence effects, and the potential influence of surrounding urban structures’ effects on the airflow. These findings have several practical implications such as implementing this design in new station constructions to improve overall ventilation performance; for existing stations, modifications to create more rectangular-shaped entrances could yield significant ventilation improvements and the observed airflow variations suggest a need for strategic placement of air quality sensors and potential supplementary mechanical ventilation in low-flow areas.

The square entrance design demonstrated limitations in natural ventilation effectiveness. CFD simulations, conducted using the same previous outdoor wind speed and a standardized inlet opening, revealed an average internal wind speed of 1.27 m/s for this design. This falls short of the recommended 1.5 m/s threshold for effective ventilation in public spaces, highlighting its inability to adequately capture and channel wind for sufficient air circulation.

For validation, our results were compared with the published literature. The findings align with those in [42], in which it was reported that square entrances typically underperform compared to rectangular designs in subway stations. The authors attributed this to smaller effective flow areas, which the results corroborate. The 1.74% improvement over baseline paled in comparison to the advancements observed with the rectangle design [78].

The analysis revealed uniform internal flow patterns with a low coefficient of variation (CV) of 25.98% and standard deviation of 0.33 m/s. While this uniformity might seem beneficial, it actually indicates insufficient turbulence for effective air mixing and contaminant dispersion, as highlighted by [79].

The discrepancy between our results and the ideal ventilation threshold can be attributed to the square geometry’s inherent limitations in wind capture and channeling; potential flow separation at the entrance corners, reducing effective air intake; and lack of directional bias in the square shape, leading to less efficient wind utilization

These findings have several practical implications such as square entrances should be avoided in new station designs where natural ventilation is a priority. For existing stations with square entrances, modifications such as adding wind-catching elements or altering the entrance angle could improve performance. The uniform flow pattern suggests that supplementary mechanical ventilation might be necessary to achieve adequate air exchange rates.

The cylinder rectangle entrance design demonstrated considerable potential for enhancing natural ventilation within the subway station. Using the same outdoor conditions and a fixed inlet revealed an average internal wind speed of 1.45 m/s for this design. This approaches the recommended 1.5 m/s ventilation in spaces.

The findings align with research on the effectiveness of curved geometries in capturing and channeling wind [47]. The 15.64% improvement over baseline in our study is consistent with improvements observed by [47] for similar curved configurations.

The high coefficient of variation (28.27%) and standard deviation (0.41 m/s) suggest significant fluctuations in local wind speeds within the station. This aligns with findings from [80], where curved geometries were shown to induce complex airflow patterns and localized variations.

These findings have several practical implications, for example, the cylinder rectangle design could be implemented in new station constructions to improve overall ventilation performance, particularly in areas with consistent wind directions; and for existing stations, curved entrance geometries could yield significant ventilation improvements.

The square cylinder entrance design demonstrated moderate potential for natural ventilation improvement, but with notable limitations. With the same conditions, it revealed an average internal wind speed of 1.41 m/s for this design. This falls short of the recommended 1.5 m/s, indicating insufficient wind capture and circulation. This aligns with observations in a previous study [78], which attributed limitations of square ratio geometries to their smaller effective flow areas compared to rectangle ratio geometries, although this design is still better than the non-cylindrical square shape due to the positive effect of the curved shape on the movement of air into the station [47].

Although the analysis revealed internal flow variations with a CV of 27.65% and standard deviation of 0.39 m/s, suggesting some turbulence and potential mixing benefits, this did not translate to a significant overall improvement. While it showed a 12.17% increase over the baseline, this gain paled in comparison to the advancements observed with the rectangle and cylinder rectangle designs.

These findings have several practical implications: while the square cylinder design offers some improvement over non-cylindrical square entrances, it may not be the optimal choice for maximizing natural ventilation in subway stations; the moderate turbulence observed could be beneficial for air mixing and contaminant dispersion, suggesting potential applications in areas where uniform air distribution is prioritized over high ventilation rates; and for existing stations with square entrances, adding cylindrical elements could yield moderate ventilation improvements without major structural changes.

After studying and comparing the straight entrances with the current situation and analyzing them, inclined entrances are analyzed and studied by analyzing the simulation results, as shown in Figure 15 and Figure 16, in comparison with the previous data to reach the best and the most effective design for improving natural ventilation inside the station.

Figure 10. Questionnaire users’ statistics and percentages.

Figure 10. Questionnaire users’ statistics and percentages.

Figure 11. Computational simulation of wind flow and speed shows the wind patterns in surrounding urban area of the subway station.

Figure 11. Computational simulation of wind flow and speed shows the wind patterns in surrounding urban area of the subway station.

Figure 12. Comparison between wind velocities in the subway station for each shape.

Figure 12. Comparison between wind velocities in the subway station for each shape.

Figure 13. Comparison between average wind velocity, improvement percentage, and relation between standard deviation and CV for the existing entrances and straight entrance designs along 7 points in the station.

Figure 13. Comparison between average wind velocity, improvement percentage, and relation between standard deviation and CV for the existing entrances and straight entrance designs along 7 points in the station.

Figure 14. Difference between air flow through existing and straight entrance designs.

Figure 14. Difference between air flow through existing and straight entrance designs.

Emerging as the most successful design for enhancing natural ventilation within the subway station, the inclined rectangle entrance exceeded the recommended 1.5 m/s wind speed threshold, with an average internal velocity of 1.54 m/s. These results align with previous studies highlighting the effectiveness of inclined geometries in capturing and channeling wind for improved ventilation performance [61].

Furthermore, the observed 23.07% improvement in ventilation over the baseline underscores the significant potential of this design for enhancing natural ventilation and air quality within the station. However, the analysis revealed internal flow variations, with a standard deviation of 0.49 m/s and a CV of 31.81%, indicating unevenness in wind velocities across occupied areas. While turbulence can provide mixing benefits, excessive variability can lead to localized discomfort and suboptimal ventilation effectiveness.

In practical applications, the inclined rectangle design could be implemented in new subway stations to significantly improve natural ventilation and air quality. For existing stations, modifications to incorporate inclined entrance geometries could yield similar benefits. Future research should focus on optimizing these designs to minimize internal flow variations and ensure more uniform ventilation throughout the station.

While the inclined square entrance exhibited some potential for natural ventilation, with an average internal wind speed of 1.26 m/s, it lagged behind other designs. The analysis revealed the lowest standard deviation 0.31 m/s and CV of 24.6%, indicating minimal local variations and potentially more uniform airflow distribution compared to other designs [69]. Despite a modest 0.83% improvement in ventilation over the baseline, the inclined square entrance fell short of the advancements observed in other shapes [78].

The findings have been compared with those in the existing literature, ensuring consistency in input data, operating conditions, and system descriptions. The results are comparable to those reported in [61], which found similar airflow patterns in inclined geometries. Any deviations can be attributed to differences in environmental conditions and specific design dimensions, which have been detailed and justified in the analysis.

The limited improvement suggests that while the design promotes airflow uniformity, it fails to effectively capture and channel external winds into the station. This indicates a need for further optimization to fully unlock its potential. Practical applications of this design could focus on areas where airflow uniformity is crucial, and supplementary strategies might be necessary to enhance wind capture and channeling.

To enhance real-world applicability, future research should aim to refine the inclined square entrance design to better capture and direct external winds, potentially integrating it with additional ventilation systems. This approach could maximize both airflow uniformity and overall ventilation performance in subway stations.

The inclined cylinder rectangle entrance emerged as a high-performance design for enhancing natural ventilation within the subway station, achieving an average internal wind speed of 1.49 m/s. This performance closely aligns with the recommended value, underscoring its potential to nearly optimize ventilation. The low standard deviation of 0.48 m/s suggests minimal variability in wind speeds within the station, contributing to a more uniform airflow distribution.

These findings align well with the established research. Similar to [47], who showed the benefits of curved geometries for ventilation, the cylinder rectangle design demonstrably enhances air circulation within the station. Any discrepancies observed in the results likely stem from specific environmental factors or slight variations in design dimensions, which are explained in detail within the analysis.

Furthermore, the observed 19.36% improvement in ventilation compared to current conditions is highly promising. This significant increase highlights the effectiveness of the inclined cylinder rectangle design in enhancing natural ventilation within the station. These findings suggest that this design could be practically applied in new subway station constructions or as modifications in existing stations to improve overall ventilation performance.

The inclined cylinder square entrance demonstrates clear potential for enhancing natural ventilation within the subway station, although slight refinements could further minimize flow fluctuations within occupied zones. According to CFD simulations, the inclined cylinder square entrance achieved an average internal wind velocity of 1.42 m/s, falling slightly short of the recommended value. Nevertheless, this performance still indicates acceptable air circulation within the station.

The analysis revealed some internal flow variations, as evidenced by a standard deviation of 0.43 m/s and CV of 30.27%. However, this was outweighed by a notable 13.17% improvement in ventilation compared to current conditions, exceeding advancements observed in some of the tested designs.

Enhancing the inclined cylinder square design presents an opportunity to further improve wind capture and achieve more uniform airflow distribution within the station. This approach could be implemented in both new subway station construction and as retrofit modifications for existing stations, ultimately leading to significant ventilation performance gains. Future research should focus on optimizing this design for maximum effectiveness. This could involve exploring integration with complementary ventilation systems to ensure both consistent airflow and improved air quality throughout the station.

While this variability might be offset by the overall improvement in wind velocity, further optimization through CFD simulations is recommended to address these concerns [81]. It is very important to optimize dimensional proportions and entrance geometry to achieve a more uniform distribution of airflow and maximize ventilation effectiveness [42]. Adjusting proportions specifically to promote cross-breeze circulation could be a fruitful avenue for exploration. By leveraging these scientific insights and employing optimization techniques, the designs might be refined to achieve a level of performance closer to the recommended value and address the current ventilation shortcomings within the subway station.

Figure 15. Comparison between average wind velocity, improvement percentage, and relation between standard deviation and CV for the existing entrances and inclined entrance designs along 7 points in the station.

Figure 15. Comparison between average wind velocity, improvement percentage, and relation between standard deviation and CV for the existing entrances and inclined entrance designs along 7 points in the station.

Figure 16. Difference between air flow through existing and inclined entrance designs.

Figure 16. Difference between air flow through existing and inclined entrance designs.

5. Conclusions

This study employed a comprehensive approach to evaluate and enhance natural ventilation within the urban subway station, with a focus on energy sustainability and passenger comfort. Through a combination of quantitative and qualitative data gathered from passenger questionnaires, valuable insights have been gained into comfort levels and design preferences. Computational fluid dynamics (CFD) modeling was utilized to analyze wind flow patterns around the station complex under prevailing wind conditions. Various entrance geometries were numerically simulated to assess their impact on cross-ventilation. The key findings and contributions of this research are as follows:

• The user survey validated the need for ventilation improvements, revealing passenger dissatisfaction with current conditions. It identified entrance design as a crucial factor influencing indoor airflow. Preferred solutions included width expansion and multiple or alternating entrances, in accordance with the literature. However, modeling suggests that optimizing entrance shape could be a more cost-effective initial approach, warranting further computational simulations.
• CFD modeling analyzed wind flow patterns around the station complex under prevailing wind conditions. Various entrance geometries were simulated to assess their impact on cross-ventilation. Rectangular and cylindrical entrance designs showed the highest potential to exceed the recommended 1.5 m/s average indoor wind speed threshold, effectively capturing and channeling winds better than baseline and square designs. This validates prior research highlighting the advantages of these aspect ratios.
• Inclined geometries, particularly the inclined rectangle, demonstrated the best overall performance, with average speeds of 1.54 m/s and a 23.07% ventilation improvement. This aligns with studies attributing enhanced wind capture to inclined openings.
• While some designs achieved ventilation gains, all revealed minor-to-moderate flow non-uniformities, as indicated by standard deviation and CV values up to 31.81%. Previous investigations have linked such variations to potential indoor air quality and comfort issues. Optimization of dimensional parameters and proportions using CFD could help address this unevenness through techniques like promoting cross-breeze induction, supporting modeling-guided design refinement.
• Quantitative evaluations using numerical modeling provide objective insights into configurations that best support thermal comfort through natural ventilation optimization. These results validate the pursuit of customized geometric solutions tailored to local climate conditions as an effective way to passively improve indoor environmental quality in subway stations. This passive approach contributes to energy sustainability by reducing reliance on mechanical ventilation systems. Better entrance designs can improve air quality and comfort in a more sustainable and energy-efficient way, reducing the need for mechanical systems that consume a lot of power.

While the CFD simulations provided valuable insights, their inherent limitations due to simplified models necessitate real-world validation through physical wind tunnel testing. Additionally, the implementation of wider entrances or multiple entrance options in existing stations may be constrained by cost considerations. However, incorporating these features in new constructions would be highly beneficial.

To address these limitations and further optimize ventilation, future research will focus on the orientation of entrances, as it directly influences the effectiveness of wind capture and the subsequent natural ventilation process. Studies will investigate how different entrance orientations can optimize the use of prevailing winds, thereby enhancing the station’s ventilation efficiency and reducing energy consumption. Additionally, optimizing the promising inclined rectangle and cylinder rectangle designs using CFD, aiming to minimize local wind speed variations and ensure even airflow distribution.

Exploration of alternative entrance geometries and the impact of cross-breeze circulation on ventilation performance will also be undertaken. Furthermore, validating CFD results through physical wind tunnel testing and conducting follow-up user surveys to evaluate implemented improvements and gather feedback are crucial next steps, as well as further improvements to the accuracy of the CFD-based model. Incorporating energy consumption analysis into these studies will provide a comprehensive understanding of the potential savings and efficiency improvements.

By addressing these limitations and pursuing the proposed future work, we can make significant contributions to improving natural ventilation, reducing energy consumption, and enhancing thermal comfort for passengers within the Gamal subway station and similar environments.