Optimization Study of Outdoor Activity Space Wind Environment in Residential Areas Based on Spatial Syntax and Computational Fluid Dynamics Simulation

Abstract

In the context of increasing global energy shortages and climate change, the human living environment, as a crucial component of residents’ daily lives, has garnered growing attention from the academic community. Research on residential environments is vital for promoting the sustainable development of urban construction and constitutes an important aspect of sustainable development studies. This study focuses on the optimization strategy for the outdoor activity space wind environment in the Xihuayuan residential area in Lanzhou city, utilizing spatial syntax analysis and Computational Fluid Dynamics (CFD) simulation technology. Firstly, the outdoor activity space is analyzed for visibility and spatial accessibility using DepthmapX0.6 software. Then, the outdoor wind environment in the residential area is simulated using PHOENICS 2018 software, and the analysis is conducted on outdoor spaces with a poor wind environment in terms of high accessibility. The results indicate that residents’ outdoor comfort in these spaces is poor, highlighting the urgent need for improvement in the wind environment. This research attempts to optimize the wind environment in high-accessibility spaces within the residential area by improving building layout, orientation, and height. The simulation results after optimization demonstrate an increase in the overall average wind speed to 1.44 m/s, with the proportion of spaces with a good wind environment in high-accessibility areas during summer rising from 33.4% to 59.2%. The optimization strategy effectively improves the wind environment in high-accessibility areas of the residential area.

1. Introduction

Accelerating urbanization has increased China’s urban population to 59.58% in 2019 [1]. Rapid urbanization has disrupted urban metabolism, exacerbating the imbalance between resource supply and demand. Consequently, issues such as air pollution and the urban heat island effect have also contributed to the deterioration of urban living environments [2]. As primary human habitats, cities are focal points for implementing sustainable development goals [3]. However, some scholars argue that the resources available to cities and their environmental impacts far exceed the cities’ own capacities for regeneration, suggesting that urban development is inherently unsustainable. This study aims to simulate and analyze the wind environment of residential outdoor activity spaces and provide reasonable recommendations for improving wind conditions. The goal is to foster positive impacts on the sustainability of certain urban areas. As reported by national statistics, public activity spaces are instrumental in shaping the urban living environment, thereby facilitating urbanization and enhancing the quality of life. Residential areas, serving as the primary spatial framework, are essential for resident interaction, social engagement, and community building. These areas, together with the built environment, are intricately linked to human health and well-being. Outdoor public activity spaces, pivotal for daily social interactions among residents, are equipped with essential facilities for recreation, exercise, and leisure. Such spaces address not only the material and cultural needs of the residents but also support their spiritual well-being.

Wind environment refers to the flow conditions of air within and around buildings and their impacts. It plays a crucial role in atmospheric dispersion, regulating urban temperatures, and improving the urban thermal environment. With the rapid development of computer and numerical simulation technologies, Computational Fluid Dynamics (CFD) has emerged as a novel method for simulating wind environments in residential areas, widely adopted by scholars. Research on residential wind environments has been conducted internationally for some time. Early studies have demonstrated that wind speed plays a dominant role in human thermal comfort [4]. Additionally, CFD simulation technology has been employed to evaluate residential wind environments with notable results. For instance, as early as 1986, Hanson and colleagues used three-dimensional simulations to study wind flow around buildings, proposing more accurate predictive models, particularly for complex building layouts [5]. Subsequently, Blocken and Carmeliet (2004) further investigated how high-rise buildings can introduce higher wind speeds at pedestrian levels, highlighting the importance of considering external wind environments in architectural design [6]. With increasing attention to sustainable building design, Asfour (2010) examined the impact of various building group configurations on outdoor wind environments. His CFD simulations revealed how building layouts and orientations significantly affect airflow and ventilation potential, especially in hot climates [7]. Following this, Kubota et al. (2008) used wind tunnel testing to explore the relationship between building density and pedestrian wind speeds, providing guidelines for achieving ideal wind environments in residential areas [8]. Entering the new millennium, Tsang et al. (2012) studied the effects of high-rise building size, spacing, and base structure on pedestrian wind environments through wind tunnel experiments, deepening the understanding of the relationship between building design and wind environments [9]. In recent years, Jones et al. (2004) combined wind tunnel and CFD simulations to investigate pedestrian wind environments in high-rise residential areas in Hong Kong. They found that while both methods provided similar predictions of wind flow patterns, there were still discrepancies in some details [10]. Additionally, Aeinehvand et al. (2021) focused on natural ventilation in hot and humid climates and proposed design solutions to enhance ventilation efficiency through simulation studies, providing a scientific basis for optimizing building ventilation design [11]. The urban heat island (UHI) effect and wind environment significantly impact residents’ comfort and health. Shahmohamadi et al. (2011) investigated the health implications of UHI and proposed strategies to mitigate its adverse effects by improving transportation infrastructure, increasing green spaces, and enhancing the reflectivity of building materials [12]. These recommendations are particularly relevant in arid climates, where urban environments present unique microclimatic conditions [13]. Oke (1988) examined the influence of urban canyon layer climates on wind environments, highlighting the critical role of building layout and geometry in optimizing urban wind conditions [14]. Berkovic et al. (2012) further explored the impact of various shading strategies on courtyard thermal comfort, noting the significant roles of wind speed and shading in enhancing thermal comfort [15]. Pearlmutter et al. (2005) introduced an integrated method combining open-air scale models with empirical observations to evaluate urban surface thermal exchanges [16]. This approach benefits from the flexibility of modeling and the reliability of field observations, effectively simulating and predicting thermal exchanges and energy flows in urban microclimates. Middel et al. (2014) investigated the effects of urban form and landscape design on mid-afternoon microclimates, demonstrating how different urban forms and landscaping can influence local temperature and comfort conditions [17].

In China, research in this field has relatively recent origins. In terms of theoretical exploration, researchers have begun applying CFD technology to simulate and analyze wind environments in residential areas. For instance, in 2010, Yang Li utilized CFD technology to simulate airflow in some buildings and the design center of Tongji University, demonstrating its practical value in residential area planning and design [18]. In 2019, Zeng Suiping and his colleagues conducted a comprehensive exploration of the ventilation efficiency of various residential area modules from multiple perspectives. They proposed a formula describing the relationship between building volume density and the urban ventilation environment. This contribution provides a theoretical foundation for urban planning efforts. Furthermore, researchers have proposed various wind environment evaluation indicators and explored methods to optimize residential area layouts [19]. In 2020, Yao Haoyi and his colleagues proposed using the calm wind rate as an evaluation indicator. They conducted CFD simulation analyses to assess the wind environment advantages and disadvantages of different layout patterns, providing recommendations for the planning and layout of medium-volume residential communities [20]. In the research on the relationship between architectural spatial configurations and the wind environment, in 2013, Song Dexuan utilized CFD technology to simulate the relationship between different architectural spatial combinations and the wind environment. They summarized general patterns and provided significant guiding insights for the future planning of residential environments. These studies provide theoretical support from the perspectives of architectural design and planning to enhance the quality of the wind environment in residential areas [21]. In terms of simulating and analyzing actual projects, in 2013, Qian Yi and his colleagues conducted a numerical simulation to study the wind environment in residential neighborhoods in Wuhan. They found that building layout and wind direction significantly influence the wind environment. They advocate for the use of numerical simulations combined with urban wind rose diagrams to inform rational planning layouts, thus optimizing the wind environment in residential neighborhoods [22]. In 2016, Ye Zongqiang and his team conducted a study on the wind environment evaluation and planning strategies for large-scale residential areas in Xi’an city. Through field measurements and orthogonal experimental design, they proposed planning strategies related to building layout and spatial height control. The aim was to enhance the environmental quality of large residential areas [23]. In 2017, Cao Xiangming and his team pointed out in their research that the layout issues of high-rise building areas in Xi’an severely affect the quality of the urban climate environment. They proposed strategies such as the rational planning of urban ventilation corridors and controlling building height and density to improve the quality of the wind environment [24]. Jiuhong Zhang and Xiaoqian Zhang (2021) investigated the pedestrian-level wind environment in Shenyang’s residential areas, where increasing urbanization and high-rise buildings have significantly impacted wind conditions. By simulating various architectural layouts, they proposed certain configurations that could enhance the wind environment [25]. Xiang Liu et al. (2023) employed entropy-based TOPSIS methods combined with CFD simulation techniques to study the wind environment of residential complexes in the hot-summer and cold-winter climate zones. By analyzing the wind environment of a residential area in Changsha under the most unfavorable conditions, they explored optimal design strategies. Their findings highlighted the greater significance of winter wind conditions over summer, particularly regarding building layout [26].

The fundamental principle of spatial syntax involves dividing space into scales and spatial hierarchies to describe the structural characteristics of urban space. The spatial referent it denotes is a concept that provides a more intuitive and rational description of abstract relationships such as topology and geometry [27]. In 2004, Zhang Yu and his colleagues introduced and analyzed the theory, methods, and practices of spatial syntax. They particularly focused on analyzing the morphological variables of spatial syntax and its practical spatial analysis techniques, emphasizing that spatial syntax is a method for quantitatively analyzing spatial configurations [28]. In 2014, Chiaradia and colleagues addressed the limitations of spatial syntax in urban planning and proposed improvement pathways. They noted that while spatial syntax enjoys high recognition in the field of architecture, its application in urban planning is subject to controversy. However, with the refinement of theory and the development of the discipline, the application of spatial syntax in urban planning has gradually strengthened [29]. In 2015, Zhang Ji explored the application of spatial syntax in spatial evaluation during the implementation of planning, using the core area of Hangzhou Qianjiang New Town as an example. He proposed a new method for spatial evaluation in planning, which involves interpreting spatial self-organization patterns to assess the alignment between spatial performance and planning expectations [30]. In 2019, Huang Kai and colleagues conducted a study on sustainable conservation of urban historical environments based on spatial syntax, using the Xiguan historical district in Guangzhou as a case study. They emphasized the significance of spatial syntax as a theoretical method linking urban spatial structure with socio-economic activities, providing a research framework applicable to the sustainable conservation of historical environments [31].

In summary, research on residential area wind environments encompasses various aspects, including the utilization of CFD technology, the proposition of wind environment evaluation indicators, and the optimization of building layouts. Different studies address distinct facets of residential area wind environment issues and suggest corresponding planning strategies. These investigations offer valuable insights for urban planners and designers, furnishing theoretical and methodological support for augmenting the quality of residential area wind environments and ameliorating urban livability. Nonetheless, current research still exhibits limitations, such as a restricted sample scope and the necessity for further refinement of research methods and evaluation indicators.

Spatial syntax, as a quantitative research method, can facilitate the comprehension of the relationship between urban spatial structure and functional layout, furnishing a scientific basis for planning decisions. In the examination of wind environments in residential outdoor activity spaces, spatial syntax can quantify the impact of spatial structure on wind environments, guiding design and planning endeavors to enhance residents’ quality of life.

In current research on urban residential wind environments, most studies focus primarily on either wind environment simulation or space syntax analysis, with few investigating the coupling of wind environment with spatial accessibility. Existing research often overlooks the importance of high-accessibility spaces in residents’ daily lives and the impact of these spaces’ wind environments on resident comfort. This study, using the Xihuayuan residential area in Lanzhou as a case, integrates space syntax with CFD simulation techniques. This approach not only enables a detailed analysis of how residential space structures affect wind environments but also effectively identifies high-accessibility and low-accessibility spaces within the residential area using space syntax, and precisely predicts and optimizes the wind environment of these spaces through CFD simulation. By adjusting building layouts, orientations, and heights, a more generalized and effective strategy for optimizing residential wind environments is proposed, aimed at enhancing outdoor activity comfort for residents and promoting the sustainable development of residential areas.

2. Materials and Methods

2.1. Study Case Analysis

The research focuses on the outdoor activity space wind environment in the Xihuayuan residential area in Lanzhou. Situated at the intersection of Dunhuang Road and Huoxing Street in the Qilihe District of Lanzhou city, the Xihuayuan residential area comprises a total of 24 high-rise residential buildings and a kindergarten. Some of the buildings feature ground-floor commercial podiums (Figure 1).

To enhance the clarity of the spatial characteristics of the buildings, a sectional view was utilized to illustrate the vertical heights of each structure, with appropriate markings for differentiation (Figure 2). Specifically, the sectional view indicates that there are 13 high-rise residential buildings with 11 floors each (1#, 2#, 3#, 4#, 5#, 6#, 7#, 8#, 10#, 12#, 14#, 15#, and 17#), each reaching a height of 33.7 m; 1 building with 18 floors (19#), reaching a height of 54.5 m; 4 buildings with 21 floors each (11#, 20#, 21#, and 24#), each standing at 63.2 m; 3 buildings with 24 floors each (9#, 13#, and 18#), each with a height of 72.5 m; and 3 buildings with 28 floors each (16#, 22#, and 23#), each soaring to a height of 84.1 m. The total area of the study site spans approximately 11.7 hectares and is characterized by a grid layout oriented towards the south.

2.2. Accessibility Study of Outdoor Public Spaces Based on Spatial Syntax

The basic principle of spatial syntax involves the division of space into scales and spatial hierarchies, with the aim of describing the structural characteristics of urban space. The spatial referent it denotes is a concept that provides a more intuitive and rational description of abstract relationships such as topology and geometry.

Visual integration refers to the total visual depth of an element Di, starting from a certain element Di in the system to the shortest meter distance path to other elements, and the sum of paths after exhausting all other elements. On this basis, Mean Depth (Di), RA (Di), and RRA (Di) are successively calculated, and then the reciprocal is taken to obtain visual integration [HH] (DI) [32].

Visual integration is defined as the degree of connectivity within a spatial system. A higher value indicates that the element requires fewer turns to see other elements in the system, and the paths from this element to reach other elements are more straightforward, indicating better accessibility. Conversely, a lower value in visual integration analysis indicates that the element requires more turns to see other elements in the system, and the paths from this element to reach other elements are more convoluted, indicating poorer accessibility.

2.2.1. Data and Information

This study utilized satellite imagery from LocaSpace Viewer 4.2.2 in 2023 as the data source. The study area underwent processing in CAD 2016 software to delineate the residential area boundaries and building outlines, which were then saved in DXF format to generate the spatial syntax analysis base map. The base map was subsequently imported into Depthmap-Beta 1.0 software for Visibility Graph Analysis (VGA) of outdoor public spaces within the residential area. The results of spatial visual integration, represented by the Integration Map, were depicted using a color spectrum, where warmer colors (red) indicate higher values and cooler colors (blue) indicate lower values.

To conduct a more in-depth study of the wind environment in outdoor activity spaces within residential communities, several representative outdoor areas were selected as the primary subjects of this research. These areas, designated as Zones A, B, C, D, and E, are key locations for residents’ daily activities, and their wind environment is crucial for both comfort and safety. The selected zones are distributed across various locations within the community, covering a range of microclimate conditions and representing different distances, orientations, and exposure levels relative to buildings. Given the impact of building layout on wind flow, these areas may experience ventilation issues or be affected by unsuitable wind speeds, making them highly sensitive to adverse wind conditions. By analyzing these representative zones, the study provides targeted recommendations for optimizing the wind environment of outdoor activity spaces in the community, aiming to enhance overall resident comfort and outdoor activity experiences.

2.2.2. Analysis of Simulation Results

Analysis of the Visibility Graph Analysis (VGA) of outdoor public spaces within the residential area yielded the results of spatial visual integration, as shown in Figure 2. In the plan, the viewpoints are represented by grid-divided small squares. To facilitate further analysis, the study area was roughly numbered, and

Table 1. Spatial topological data for each analysis point.

Point number	A1	A2	A3
Integration [HH]	11.9	12.6	10.1
Point number	B1	B2	B3
Integration [HH]	9.3	9.0	10.4
Point number	C1	C2	C3
Integration [HH]	13.6	12.8	14.1
Point number	D1	D2	D3
Integration [HH]	9.1	8.6	10.2

presents the spatial topological data for each analysis point.

From the analysis (Figure 3), it is observed that Area C corresponds to the central plaza of the residential area and is adjacent to the main entrance. The color representing its integration is the warmest, indicating that this space has the widest field of view. Based on the spatial topological data for each analysis point, it is found that Area C3 has the highest integration value (integration [HH]), suggesting that the central plaza space is open and has a broad field of vision, thus offering the highest accessibility. This area is more conducive for residents to gather and engage in outdoor activities, and it is more easily perceived. The integration value gradually decreases from the central plaza towards the outer areas, and the color transitions from warm to cool. This is because Areas B and D are both clustered activity areas between buildings, resulting in a strong sense of enclosure and a narrower field of view, leading to lower integration and poorer accessibility but better privacy. Area A2, which is the intersection space, has a slightly higher integration value than Areas A1 and A3, indicating a better visual range.

2.3. Study on Residential Area Wind Environment Based on CFD Simulation

2.3.1. Selection of CFD Numerical Simulation Software

The selection of appropriate simulation software for CFD analysis should be based on the research object. In this study, PHOENICS, with its FLAIR module, is chosen due to its compatibility with the spatial scale of the research object. Moreover, PHOENICS has been extensively validated, demonstrating its advantages in accuracy and prediction capability of results.

2.3.2. Data and Analysis

The meteorological data used in this study are sourced from the Chinese Standard Meteorological Database (CSWD). EPW format data files containing temperature, wind speed, wind direction, and solar trajectory data for all 8760 h of the year were obtained using the Ladybug plugin within Rhino 7 software’s Grasshopper platform. The CSWD data include measured meteorological data for all 8760 h of the year.

The meteorological data for wind direction and wind speed in Lanzhou city throughout the year were processed and filtered using the Ladybug plugin. Additionally, a frequency analysis of wind direction was conducted, resulting in the creation of a wind rose diagram for the entire year (Figure 4). It illustrates the wind direction distribution across different wind speed levels. In the diagram, colors ranging from blue to red represent wind speed intervals from low to high. Multiple closed curves indicate the frequency of different wind directions, with each curve representing a frequency of 1.0% (approximately 50 h). As shown in Figure 4, the wind environment characteristics in Lanzhou are predominantly characterized by moderate-to-low wind speeds with no significant directional bias, indicating a need for further acquisition and analysis of wind speed and direction data for specific seasons.

According to national climate zoning, Lanzhou falls into a cold region. Considering the wind environment characteristics of Lanzhou, it is concluded that residential area design should account for summer ventilation and heat protection, and winter insulation and wind protection. Additionally, the cold climate conditions in Lanzhou during winter may lead to the accumulation of cold air within building clusters, creating cold air corridors or localized microclimates. Considering outdoor comfort in winter is beneficial for reducing heat loss from buildings and promoting the sustainable development of urban living environments.

Therefore, the predominant wind directions and average wind speeds for summer (June, July, and August) and winter (December, January, and February) are used as the simulation data for the wind environment analysis in this study.

From the summer wind rose diagram for Lanzhou (Figure 5), it is evident that the predominant wind direction during the summer is between north and northeast, with NNE being the dominant direction. From the winter wind rose diagram for Lanzhou (Figure 6), it can be observed that the predominant wind direction during the winter is also between north and northeast, with ENE being the dominant direction. Further precise calculations were conducted on the wind direction data using common methods. The formula used for this calculation is

$$D_{avg}={\sum }_{\mathcal{i}=1}^n{\displaystyle \frac{D_{\mathcal{i}}}{n}}$$

(1)

In the formula, $D_{avg}$ represents the mean wind direction, $D_{\mathcal{i}}$ represents the wind direction azimuth of each sample, and $n$ represents the number of samples.

The average wind speed under the predominant wind direction for summer is calculated to be 1.21 m/s; while, for winter, it is 0.7 m/s. These values are used as boundary reference data for the PHOENICS wind environment simulation.

2.3.3. Wind Environment Evaluation Criteria

Residential wind environment is a key indicator for evaluating the ecological environment of residential areas, primarily based on the subjective perception of individuals at a height of 1.5 m above ground level. The “Green Building Evaluation Standard” GB/T50378-2019 explicitly outlines the quantitative criteria for wind environment evaluation, which serves as the basis for this study’s evaluation [33]. Wind speed and wind pressure are adopted as the evaluation standards.

Under typical summer wind speed and wind direction conditions, the wind speed at a height of 1.5 m above ground level around buildings should be greater than 1 m/s to prevent the occurrence of local vortices or windless areas. Additionally, the pressure difference between the inner and outer surfaces of more than 50% of operable windows should exceed 0.5 Pa.

Under typical winter wind speed and wind direction conditions, the wind speed at a height of 1.5 m above ground level around buildings should be less than 5 m/s. In outdoor resting areas and children’s recreational areas, the wind speed should be less than 2 m/s. Except for the first row of windward buildings, the pressure difference between the windward and leeward surfaces of buildings should not exceed 5 Pa.

In conclusion, the established comfort evaluation criteria for the wind environment in this study aim to meet the wind protection needs of residential areas in both winter and summer seasons. The comfortable wind speed range for pedestrian areas is 1.0 to 5.0 m/s.

2.3.4. Model Establishment

Based on Google satellite maps and field surveys, the map was vectorized using CAD software. To study the impact of architectural spatial patterns on wind environments, factors such as vegetation and structures were ignored during the importation process into SketchUp for modeling. The architectural outlines were appropriately simplified to establish the wind field model (Figure 7). Wind fields in the study are typically influenced by neighboring buildings. To closely simulate real conditions, neighboring buildings were also suitably modeled.

Selection of Mathematical Model

The turbulence model illustrates the movement of fluids within the field. Following the recommended standard k-model outlined in the “Civil Building Green Performance Calculation Standard”, simulations are conducted with a convergence accuracy of 0.0001 and 10,000 iterations, until convergence is achieved. This turbulence model approximately simulates the construction of first-order closed Reynolds stress, and its governing equations are represented by the Reynolds-averaged Navier–Stokes equations [34].

$${\displaystyle \frac{\partial U_i}{\partial {\mathcal{X}}_i}}=0$$

(2)

$$U_j{\displaystyle \frac{{\partial }_{U_i}}{{\partial }_{{\mathcal{X}}_j}}}=-{\displaystyle \frac{1}{\rho }}{\displaystyle \frac{\partial P}{\partial x_i}}+{\displaystyle \frac{\partial }{\partial x_j}}\left(\mu {\displaystyle \frac{{\partial }_{U_i}}{{\partial }_{{\mathcal{X}}_j}}}\right)+{\displaystyle \frac{\partial }{\partial x_j}}\left(\overline{{-\mathcal{U}}_i}\overline{{\mathcal{U}}_j}\right)$$

(3)

$$\overline{{-\mathcal{U}}_i}\overline{{\mathcal{U}}_j}=2{\complement }_{\mu }k{s}_{ij}-{\displaystyle \frac{2}{3}}k{s\delta }_{ij}$$

(4)

In the equation, $U$ represents the mean velocity component, $x_i$ and $x_j$ represent the coordinate directions, $\overline{{-\mathcal{U}}_{\mathrm{i}}}\overline{{\mathcal{U}}_{\mathrm{j}}}$ denotes the Reynolds stress term, $P$ signifies the mean pressure, $\rho$ stands for air density, ${\delta }_{ij}$ represents the eddy viscosity, and the constant ${\complement }_{\mu }$ is typically set to 0.087.

According to classical turbulence theory, the expressions for turbulent kinetic energy and dissipation rate are as follows:

$$Q_{sgs}={\int }_{k_c}^{\infty }E\left(k\right)dk$$

(5)

$${\rm E}=2\mathsf{\nu }{\int }_0^{\infty }{k}^{2}E\left(k\right)dk$$

(6)

Define Calculation Domain

The computational domain’s setting must be described in maximum detail to accurately capture the flow field within the building complex. The boundaries of the computational domain significantly influence the realism of the flow simulation. According to the “Green Building Performance Calculation Standard for Civil Buildings” in China, the requirements for selecting the computational domain for CFD simulation of the outdoor wind environment are as follows: the vertical height from the top of the object building to the upper boundary of the computational domain should be greater than 5H; the distance from the outer edge of the object building to the horizontal boundary of the computational domain should be greater than 5H; and where H is the height value of the building within the computational domain. By calculating the range of the residential area and surrounding buildings, the model is positioned at the center of the computational domain with dimensions of 2890 × 2540 m × 291 m for simulation [35].

Mesh Partitioning for Calculation

The speed and accuracy of simulation calculations are influenced by the number and type of computational grids. Grids that are too sparse or too dense can lead to inaccurate simulation results. In this simulation, the PARSOL method developed by PHOENICS was employed to partition the entire computational domain. After multiple adjustments, the grids were ultimately set at three different levels: the minimum grid size was 5 m × 5 m within the residential area; the intermediate grid size was 20 m × 20 m within the surrounding buildings; and the background grid size was 70 m × 70 m covering the entire simulated area, with a uniform distribution from the center to the edges.

Setting Boundary Conditions

Since the wind properties at pedestrian height in residential areas are characterized by incompressible flow, this study utilizes the velocity inlet boundary provided by the PHOENICS software. In the atmospheric boundary layer, gradient winds are influenced by surface roughness and height, and the calculation formula is as follows:.

$$V=V_0({{\displaystyle \frac{h}{h_0}})}^n$$

(7)

The equation is as follows: V represents the wind speed at height h, V₀ represents the wind speed at height h₀, h represents the height above the ground, and n represents the ground roughness.

According to the “Code for Design of Building Structures” GB 50009-2012, ground roughness can be classified into four categories: A, B, C, and D, with corresponding values as shown in

Table 2. Values of roughness coefficient α.

Category	Regulation	α
A	Near-sea surfaces and islands, coasts, lake shores, and desert areas	0.12
B	Fields, countryside, forests, hills, and sparsely populated towns with scattered houses	0.16
C	Urban areas with dense clusters of buildings	0.22
D	Urban areas with dense clusters of tall buildings	0.30

. In this simulation process, the roughness value is taken as 0.22 [36]. Additionally, since the study focuses solely on wind environment simulation, other parameters such as temperature are set to default values.

2.4. Feasibility Verification of Simulation Software

To validate the rationality and feasibility of the selected wind environment simulation software and solution process, a measurement period from 1 December 2023, from 10:00 to 17:00 has been established for this study. During this time frame, the wind speed was observed to be at level one from the northeast.

To comprehensively evaluate the wind environment of outdoor activity spaces, this study has set up multiple monitoring points in different types of spaces. The selection of these points is based on the following principles: First, monitoring points were placed in ventilation corridors between buildings (Zone A) to capture wind speed variations within these wind flow passages. Second, monitoring points were set up in main outdoor activity areas (Zone C) to assess natural ventilation conditions. Third, monitoring points were established in areas with specific wind speed requirements, such as children’s play zones (Zone E). Finally, monitoring points were placed in areas with potential significant wind speed fluctuations, such as building shadow zones (Zone D) and community entrances (Zone B), to record sudden changes in wind speed. This ensures that the monitoring points represent the overall wind environment characteristics of the study area, with all points set at a pedestrian height of 1.5 m.

This study used the GM8902 digital anemometer from Baozhi to monitor wind speed. The device has a measurement range of 0.3 m/s to 30 m/s and an accuracy of ±(2% + 0.1 m/s). To ensure the accuracy and consistency of the data, the equipment was calibrated according to the manufacturer’s recommendations before data collection, confirming that the device provides reliable data for the measurements.

Data collection was conducted over multiple time periods to capture the dynamic changes in the daytime wind environment. The specific time frame was from 10:00 to 17:00. During this period, data were recorded at a frequency of every 60 min. This arrangement allows for a comprehensive reflection of wind environment characteristics at different times throughout the day.

Turbulence intensity is a key indicator for measuring the stability of airflow. To assess the stability of airflow in outdoor spaces, this study calculated the turbulence intensity at each monitoring point. Turbulence intensity is determined by the variability of wind speed, and the formula used is

$$I={\displaystyle \frac{u^{\prime }}{\overline{U}}}$$

(8)

In the formula, u′ represents the standard deviation of wind speed, and $\overline{U}$ denotes the average wind speed. Variations in turbulence intensity can reveal the impact of building layout on the wind environment, particularly with respect to pedestrian comfort and safety.

The calculated turbulence intensities are as follows: 10.30% at Point A, 3.38% at Point B, 7.70% at Point C, 2.82% at Point D, and 3.25% at Point E. In areas with higher turbulence intensity, if the simulation results accurately reflect the variability and average wind speed of the measured data, it can be considered that the model has high accuracy and validity in complex wind environments.

Table 3. Actual wind speed measurements at monitoring points (unit: m/s).

Time	Monitoring Point A	Monitoring Point B	Monitoring Point C	Monitoring Point D	Monitoring Point E
10:00	0.59	0.91	0.47	0.62	0.91
11:00	0.65	0.93	0.52	0.64	0.96
12:00	0.68	0.97	0.54	0.66	0.99
13:00	0.61	0.89	0.49	0.62	0.92
14:00	0.59	0.87	0.45	0.62	0.89
15:00	0.63	0.92	0.50	0.65	0.94
16:00	0.74	0.91	0.53	0.64	0.93
17:00	0.79	0.95	0.58	0.67	0.98

shows the actual wind speed measurements (in m/s) at the five monitoring points (A, B, C, D, and E) during different times of the day (from 10:00 to 17:00). The data indicate that wind speeds at different monitoring points vary over time, with Point B generally recording the highest wind speeds and Point C recording the lowest. This variation suggests that local environmental factors influence the distribution of wind speeds.

Table 4. Comparison between actual and simulated values of outdoor wind environment.

Monitoring Point Locations	Average Measured Wind Speed (m/s)	Simulated Wind Speed Value (m/s)	Error Rate
Monitoring Point A	0.66	0.62	6.1%
Monitoring Point B	0.91	0.87	4.4%
Monitoring Point C	0.51	0.48	5.9%
Monitoring Point D	0.64	0.60	6.3%
Monitoring Point E	0.94	0.91	3.2%

compares the average measured wind speeds with the simulated wind speeds at each monitoring point. The results show a high correlation between the measured and simulated values, with an error rate ranging from 3.2% to 6.3%. This indicates that the simulation model used in this study has a certain level of accuracy in predicting wind speeds, although minor discrepancies remain due to the model’s inability to fully capture the environmental complexities.

Figure 8 compares the wind speed measurements at the five monitoring points (A, B, C, D, and E) across different time periods. The line graph contrasts the average measured wind speeds with the simulated wind speeds. Light-colored bars represent the measured wind speeds at each monitoring point, while dark-colored bars represent the simulated wind speeds. It can be observed that the trends in measured and simulated wind speeds are generally consistent across the time periods, indicating that the simulation model demonstrates strong adaptability in predicting actual wind speeds and effectively reflects wind speed variations across different locations and times.

In the CFD simulation, although we ensured that the boundary conditions were as consistent as possible with the actual measurement conditions, some simplifications were made during the simulation process. These simplifications included idealizing building and terrain features to reduce computational complexity and improve simulation efficiency. It should be noted that during actual measurements, wind speed data collection was influenced by surrounding environmental factors such as outdoor vegetation and structures, and wind direction did not fully adhere to the climatic conditions preset in the simulation software. This has led to some discrepancies between the measured and simulated results. However, overall, the characteristics and trends of wind speed variations between the average measured wind speeds and simulated wind speeds at different monitoring points during winter are generally consistent. Additionally, the relative error between the simulation and measurement results is maintained below 10%. Therefore, the use of PHOENICS software for simulating the wind environment in residential outdoor public spaces effectively reflects the characteristics and trends of airflow within the residential area.

2.5. Simulation Results Analysis

2.5.1. Summer Conditions

Summer Wind Speed Analysis

The summer wind speed analysis reveals that during summer, under the influence of the northeast to north wind (as depicted in Figure 9), the architectural layout of the high-rise residential complex, with buildings higher in the north and lower in the south, obstructs the introduction of prevailing winds. The high-rise buildings on the east and north sides are the first to interact with the dominant winds. Subsequently, the buildings in the rear are situated within the wind shadow of the front buildings, resulting in obstruction to the wind flow within the complex. Consequently, there are no significant differences in wind speed within the area, and the overall wind speed is relatively low, averaging around 1.1 m/s outdoors.

Although no distinct areas of calm or turbulent airflow are observed, the high-rise buildings facing south are obstructed, creating weak wind zones where wind speeds are below 1 m/s. This obstruction hampers natural ventilation, leading to some pedestrian areas not meeting the required comfort standards for wind environments.

In the diagram, Areas A, B, C, and D all represent outdoor activity spaces within the residential complex. Under the influence of the northeast wind, Area A within the complex serves as a through-flow zone, with a wind speed of 1.6 m/s, significantly higher than other areas. Due to the higher initial wind speed between the tall buildings and their proximity to the ventilation openings of the complex, outdoor wind environments in areas A and B benefit from the wider space between the buildings, resulting in better ventilation.

Area C is a recreational square within the complex, surrounded by neatly arranged residential buildings, forming a somewhat enclosed space. Here, multi-directional airflows collide and weaken, leading to reduced airflow and relatively gentle wind speed variations, resulting in poorer wind environments.

As the airflow is obstructed by windward-facing buildings, Area D is located on the leeward side of the buildings, experiencing wind speeds below 1 m/s. This creates a wind shadow zone, hampering smooth ventilation and affecting the comfort of outdoor activities for residents.

Summer Wind Pressure Analysis

The simulation results of summer building wind pressure, as depicted in Figure 10, indicate that under the dominant wind direction in summer, the buildings on the northeast windward side obstruct the airflow, resulting in increased pressure and positive pressure on the windward side of the buildings. Meanwhile, airflow circulation leads to negative pressure on both sides and the leeward side of the buildings. With the exception of slight wind pressure variations observed in the windward and leeward sides of the front-row windward buildings and buildings 3, 4, and 11, the wind pressure difference between the windward and leeward sides of the remaining buildings exceeds 0.5 Pa, meeting the evaluation criteria set forth in this study.

2.5.2. Winter Conditions

Winter Wind Speed Analysis

According to the winter wind speed contour map (Figure 11), the average outdoor wind speed in winter is approximately 0.7 m/s. Under the influence of the northeast–north winter prevailing wind, turbulent winds are formed due to the superposition of incoming winds, resulting in areas with relatively high wind speeds between windward buildings. The high-rise buildings on the windward side obstruct the incoming winds to some extent, causing the wind speeds in the western part of the community to be generally lower than those in the eastern part. This leads to the formation of localized low-speed wind shadow areas.

Areas A, B, C, and D at pedestrian height within the community meet the basic requirements for residents to engage in outdoor activities. The wind speed in Area E is less than 2 m/s, meeting the wind environment standard for the children’s recreation area in this study. Large areas of stagnant wind are observed within the community, hindering the proper ventilation and dispersal of odors and pollutants such as garbage.

Winter Wind Pressure Analysis

Based on the winter wind pressure contour map (Figure 12), the buildings on the east side of the community are affected by differences in height and changes in wind direction due to building obstruction, resulting in higher wind pressure values on the east side compared to the west side. The buildings on the west side have almost no height difference, so the wind pressure distribution pattern is similar.

Compared to summer, the winter wind speed is lower, resulting in reduced wind pressure differences between buildings within the community, all of which are less than 5 Pa. The overall slight negative pressure state allows for the introduction of fresh air from outside, which is beneficial for ventilation within the community.

3. Results and Discussion

3.1. Comprehensive Analysis of Outdoor Space Wind Environment Based on Spatial Syntax

By overlaying the analysis maps of spatial visibility integration and wind environment, a comprehensive analysis of outdoor space utilization and wind environment in the residential area is obtained. The analysis maps for summer and winter wind environments (Figure 13 and Figure 14) are derived, and factors such as summer wind speed, winter wind speed, and spatial visibility integration are normalized.

$${\mathrm{x}}_{\mathrm{i}}={\displaystyle \frac{{\mathrm{a}}_{\mathrm{i}}-{\mathrm{a}}_{\mathrm{m}\mathrm{i}\mathrm{n}}}{{\mathrm{a}}_{\mathrm{M}\mathrm{A}\mathrm{X}}-{\mathrm{a}}_{\mathrm{M}\mathrm{I}\mathrm{N}}}}$$

(9)

In the formula, i represents the normalized values of summer wind speed, winter wind speed, and spatial visibility integration.

Analysis reveals

(1) Area A exhibits high spatial integration and a favorable summer wind environment, leading to high space utilization. Despite having a slightly lower spatial integration value (average 11.5) compared to Area C (average 13.5), Area A benefits from the central activity plaza’s penetration and its proximity to the secondary entrance, resulting in a spacious and highly integrated space. Serving as the main intersection space within the residential area, it facilitates residents’ daily activities, making it highly accessible and influential. As a semi-open space near the ventilation entrance of the residential area, it is influenced by surrounding high-rise buildings. The average summer wind speed is 1.6 m/s, providing a relatively comfortable outdoor wind environment. It promotes air circulation and helps to reduce the heat island effect; while, the average winter wind speed is 0.73 m/s, with a certain area of calm wind zone. This approach mitigates the impact of cold winds during winter, thereby offering residents a relatively comfortable outdoor environment. However, the presence of stagnant wind zones may potentially result in inadequate air circulation.

(2) Area B has a lower spatial integration, but the favorable summer winds attract a certain gathering of residents in this area. It constitutes a cluster of activities between buildings, with a spatial integration score (average 9.6) lower than the overall spatial integration score of the residential area (average 10.9). The strong sense of enclosure in the space results in narrow visibility and lower integration. However, being situated in the predominant wind direction with fewer obstructions from buildings, Area B benefits from a relatively wide space between buildings, resulting in a summer average wind speed of 1.44 m/s, providing a higher level of outdoor comfort for residents. In winter, the average wind speed is 1.04 m/s, indicating overall good ventilation.

(3) Area C features high spatial integration, providing a high level of outdoor comfort and serving as a frequent gathering spot among residents. Positioned as the central activity square and adjacent to the main entrance, Area C boasts an average spatial integration value of 13.5, the highest within the residential area. This high integration enhances accessibility and establishes it as the primary social hub for residents. Despite the orderly arrangement of residential buildings in the vicinity, the space maintains a certain level of enclosure. Various directional airflows interact and weaken each other within this area.

During the summer, the average wind speed in Area C is 1.08 m/s, with a smooth variation in wind speed, albeit with some areas of stagnant air. However, in winter, the average wind speed drops to 0.63 m/s, which is too low, resulting in poor wind conditions.

(4) Area D, designated as a clustered activity space between buildings, exhibits a lower spatial integration score (average value of 9.3) compared to the overall spatial integration score of the residential area (average value of 10.9). Consequently, it offers relatively lower accessibility but provides better privacy. Due to the blocking effect of tall buildings on incoming wind, both summer and winter wind speeds in this area are below 1 m/s, resulting in poor ventilation and the formation of a large stagnant air zone. This condition negatively impacts the outdoor comfort for residents.

According to the wind environment evaluation criteria in this study, pedestrian areas with wind speeds ranging from 1.0 to 5.0 m/s are considered spaces with good wind environments, while those outside this range are considered spaces with poor wind environments. Based on the analysis results, the outdoor public spaces in this study are categorized into four types: high accessibility with good wind environment, high accessibility with poor wind environment, low accessibility with good wind environment, and low accessibility with poor wind environment (as shown in

Table 5. Spatial classification based on accessibility and wind environment.

Category	Summer	Winter
High Accessibility with Good Wind Environment	A, C1, C2	A3
High Accessibility with Poor Wind Environment	C3	A1, A2, C
Low Accessibility with Good Wind Environment	B	B
Low Accessibility with Poor Wind Environment	D	D

).

Analysis of the four types of spaces reveals the following:

• High Accessibility with Good Wind Environment: In summer, this category accounts for 33.4% of the total area, with average wind speeds ranging from 1.07 to 1.65 m/s and a calm wind rate of 17.3%. In winter, it occupies 14.3% of the area, with an average wind speed of 1.02 m/s and a calm wind rate of 21%.
• High Accessibility with Poor Wind Environment: In summer, this category represents 25.8% of the total area, with an average wind speed of 0.96 m/s and a calm wind rate of 86.1%. In winter, it covers 44.9% of the area, with average wind speeds ranging from 0.53 to 0.78 m/s and a calm wind rate of 91.6%.
• Low Accessibility with Good Wind Environment: In summer, this category accounts for 21.7% of the total area, with average wind speeds ranging from 1.22 to 1.43 m/s and no calm wind zones. In winter, it also occupies 21.7% of the area, with average wind speeds ranging from 1.01 to 1.04 m/s and a calm wind rate of 29.5%.
• Low Accessibility with Poor Wind Environment: In summer, this category represents 19.1% of the total area, with average wind speeds ranging from 0.57 to 0.85 m/s, and all areas experiencing calm winds. In winter, it covers 19.1% of the area, with average wind speeds ranging from 0.48 to 0.57 m/s, and all areas experiencing calm winds.

Based on the above analysis, comparing the summer and winter wind speeds within the high-accessibility with poor wind environment areas reveals that both the winter wind speed and its variability are lower than those in summer. Therefore, to enhance outdoor comfort for residents, this study primarily focuses on optimizing the summer wind environment in areas characterized by high accessibility but poor wind conditions.

3.2. Optimization Strategies for High-Accessibility Space Wind Environment

Numerous scholars in the field have extensively documented the significant impacts of building layout, height, and orientation on the wind environment of residential areas. Despite the selection of an existing residential area as the simulation object for this study, the pursuit of enhancing the comfort of the wind environment in residential areas in Lanzhou city and providing valuable insights for future residential developments necessitates the simulation and optimization of the aforementioned influencing factors for the study object, followed by rigorous verification.

Through the wind environment simulation analysis of Xihuayuan residential area, it became evident that high-accessibility spaces with poor wind environments encounter specific challenges:

• Zone C, serving as the main central square within the residential area, boasts good accessibility but suffers from a poor wind environment due to obstruction by north-facing buildings and the architectural enclosure surrounding it.
• The windward-facing buildings, all taller than those positioned behind them, hinder efficient natural ventilation, consequently leading to the formation of stagnant wind zones.
• Buildings oriented perpendicular to the prevailing wind direction disrupt the wind field, resulting in areas characterized by reduced wind speeds.

Based on the identified issues and following multiple simulation verifications, the study proposes the following wind environment optimization strategies:

• Building Layout: Adjust the western buildings within the residential area (Buildings 7 and 11) southward by 7 m, thus establishing a staggered layout in conjunction with other residential buildings.
• Building Height: Reduce the height of north-facing tall buildings (Buildings 16, 22, and 23) to 72 m to facilitate the ingress of prevailing winds.
• Building Orientation: Enhance ventilation efficiency within the residential area by adjusting the orientation of certain northeast-facing windward buildings (Buildings 16 and 17) to a southwesterly direction by 15 degrees (SW15).

The optimized residential area layout is depicted in Figure 15.

The analysis of the simulated wind environment at a height of 1.5 m (Figure 16) indicates that, in response to the predominant summer winds in Lanzhou, the optimization adjustments implemented concerning building heights, orientations, and overall layout within the Xihuayuan residential area effectively mitigate issues related to low wind speeds and extensive stagnant zones. The average wind speed has notably increased from the original 1.16 m/s to 1.44 m/s, with stagnant zones now encompassing approximately 7.3% of the total outdoor activity area. Notably, the proportion of high-accessibility spaces exhibiting favorable wind environments has risen to 59.2%, representing a significant enhancement in wind environment quality.

Within the optimized spaces previously classified as featuring poor wind environments, wind speeds have improved and stabilized within the range of 1.41 m/s to 1.50 m/s, with the absence of any stagnant zones. Consequently, these improvements contribute substantially to enhancing outdoor comfort for residents.

4. Conclusions

This study employs space syntax analysis to examine the accessibility of public spaces within the study area and integrates CFD numerical simulation techniques to conduct a detailed analysis of the wind environment under the current building layout. Preliminary results reveal that high-accessibility areas experience excessively high calm wind rates and poorer outdoor comfort, indicating a pressing need for optimization of the regional wind environment. Based on the preliminary analysis results, key factors influencing the wind environment in the study area have been identified, including building layout, building height, and orientation. These factors have led to lower wind speeds in some high-accessibility areas, particularly with severe calm wind phenomena during both summer and winter. This provides a clear direction for subsequent optimization efforts. To address the identified issues, the primary optimization goals are to enhance the overall average wind speed in the residential area by adjusting building layouts, optimizing building heights, and modifying orientations. This approach aims to reduce the area of stagnant wind zones and improve the wind environment in high-accessibility spaces, thereby increasing outdoor comfort. Based on the preliminary analysis and simulation results, the following conclusions have been drawn:

• High-accessibility spaces within the study area exhibit a stillness rate of 57.3% in summer and 89.6% in winter, indicating poor outdoor comfort for residents in such areas. This underscores the pressing need for enhancing the wind environment.
• Through the optimization of factors such as building layout, height, and orientation, which influence the wind environment, the overall average wind speed in the residential area has risen to 1.44 m/s. Notably, the proportion of high-accessibility spaces with favorable wind environments in summer has increased by 25.8%. Additionally, the wind speed in high-accessibility spaces previously categorized as having poor wind environments has augmented and stabilized within the range of 1.41 m/s to 1.50 m/s. This optimization has effectively transformed them into high-accessibility spaces with favorable wind environments, consequently significantly enhancing outdoor comfort for residents.
• In both winter and summer, the prevailing wind direction in Lanzhou city is northeast to north–northeast. To ensure a favorable wind environment within the residential area, the overall layout should adhere to the principle that buildings facing north or east should be lower in height compared to those in other directions. Furthermore, whenever feasible, these buildings should be arranged in a stepped manner.

The primary contribution of this study lies in integrating space syntax with CFD numerical simulation techniques to propose an optimization framework that couples wind environment analysis with spatial accessibility. This approach provides significant insights for enhancing the comfort of urban residential environments and contributes to the sustainable development of urban living spaces.

However, this study has certain limitations, primarily due to its focus solely on wind environment optimization without incorporating factors such as temperature and humidity. This may impact the comprehensiveness of the findings. Future research could extend this work by including temperature and humidity factors and validating the applicability of the proposed method under various climatic conditions, aiming to provide more comprehensive outdoor comfort optimization strategies.