A method for evaluating wind environment of track station area by integrating WRF / CALMET simulation and FAI calculation

Abstract

The application discloses a kind of track site area wind environment evaluation method of integration WRF / CALMET simulation and FAI calculation, using multiscale composite simulation method-WRF / CALMET mode to simulate wind field, obtain detailed wind field information in city and around subway station, simultaneously, introduce wind area index FAI this index, by constructing geographic information system GIS database, using digital elevation model DEM, building data and simulated meteorological data to calculate the FAI of station area;Finally, select average wind speed and wind area index FAI two wind environment evaluation indexes to evaluate the wind environment of station area.The application can realize the fast simulation of high-precision wind field information of station area and the comprehensive evaluation of wind environment of station area.

Description

Technical Field

[0001] This invention belongs to the field of urban wind environment simulation, specifically involving a method for evaluating the wind environment of rail station areas that integrates WRF / CALMET simulation and FAI calculation. Background Technology

[0002] In recent years, the public has become increasingly concerned about the negative impacts of climate change on the urban living environment, and urban development is facing increasingly severe climate challenges. Rapid urbanization and massive construction activities have led to changes in the microclimate and the emergence of problems such as the heat island effect, smog, and ventilation issues, all of which have greatly reduced the quality of urban space and the living standards of residents.

[0003] Rail transit stations, as the connecting points between the urban external space and the rail transit network, attract urban resources significantly due to their unique locations, thus generating a clustering effect. However, this clustering effect also leads to high building density, diversified population activities, and adverse impacts on the urban climate.

[0004] Urban wind environment is one of the dominant influencing factors of urban microclimate, and high-density building clusters often deteriorate the surrounding wind environment. A good wind environment can alleviate urban thermal environment problems, help disperse air pollutants, and create a healthy and comfortable living environment.

[0005] Existing research on wind environment assessment primarily covers key indicators such as wind speed, wind direction, and wind frequency. Assessment methods are mainly divided into two categories:

[0006] One approach involves physical measurement or simulation methods to directly acquire wind field data. Existing research focuses on meteorological numerical simulation, selecting different simulation tools depending on the research scale. For example, regional scales favor models like WRF and RAMS, while street-level scales commonly use CFD and ENVI-met. With technological advancements, multi-scale coupled simulations such as RAMS / UCM, WRF / CALMET, and WRF / CFD are gradually becoming more widely used. These simulations can take into account both the regional background wind and the impact of urban buildings at the macro-scale, improving the accuracy and comprehensiveness of the research. However, these studies are also limited by factors such as the difficulty in acquiring high-precision surface data and the computational power required, making it difficult to consider the impact of buildings on the wind environment.

[0007] The second method is the surface parameter analysis method, which mainly relies on theory and models to assess the impact of building layout on the wind environment. Among them, the windward area index (FAI) is a core indicator that effectively reflects the obstruction of wind flow by buildings.

[0008] However, most existing windward area indices are based on uniform wind direction information and lack more precise wind field information. Summary of the Invention

[0009] Objective: To address the shortcomings of existing technologies, this invention proposes a wind environment assessment method for urban rail transit station areas. It combines numerical simulation of the wind environment with surface parameter assessment methods. This solves the problems of existing standalone numerical wind environment measurements being limited by computing power and data accuracy, making it difficult to fully consider the impact of street layout and building design on the wind environment. It also addresses the issue of inaccurate initial wind field information in existing surface parameter methods. This method enables rapid simulation of high-precision wind field information for station areas and a comprehensive evaluation of the wind environment in these areas.

[0010] To achieve the above-mentioned technical objectives, the present invention employs the following technical means:

[0011] A wind environment assessment method for rail transit station areas integrating WRF / CALMET simulation and FAI calculation is proposed. The method employs a multi-scale composite simulation approach—the WRF / CALMET model—to simulate wind fields and obtain detailed wind field information for the entire city and the vicinity of subway stations. Simultaneously, the windward area index (FAI) is introduced. A GIS database is constructed, and the FAI for the study station area is calculated using a digital elevation model (DEM), building data, and simulated meteorological data. Finally, average wind speed and windward area index are selected as two indicators for descriptive analysis, analyzing their numerical and spatial distribution characteristics. k-means cluster analysis is then applied to both indicators to comprehensively evaluate the wind environment of the station area. The specific steps include:

[0012] S1. Urban wind environment simulation based on WRF / CALMET model:

[0013] S11, Mesoscale WRF Model:

[0014] The WRF model was used for simulation analysis. The initial field and lateral boundary conditions of the WRF model were derived from ERA5 reanalysis data.

[0015] S12, Downscaling CALMET mode:

[0016] Based on the large-scale meteorological data provided by the WRF model, the CALMET model is used for downscaling. The CALMET model includes two core components: a diagnostic wind field module and a micrometeorological module.

[0017] First, using the diagnostic wind field module, the CALMET model uses meteorological field data with a resolution of 1 km from the WRF model simulation results as the initial guessed wind field. Based on this, by considering topographic dynamics adjustment, slope flow adjustment, and topographic blocking effect, the CALMET model can generate a first-step wind field that is more adapted to the complexity of the terrain. Subsequently, the CALMET model further incorporates measured observation data and, through fine interpolation and smoothing, finally generates wind field information with higher accuracy.

[0018] In the micro-meteorology module, the CALMET model uses a parametric method to describe the structure of the atmospheric boundary layer. By performing targeted wind field diagnosis on the study area, the CALMET model uses a 100m grid spacing and divides the grid according to the study area. The lowest layer in the vertical direction is located 10m above the ground, thereby achieving a refined simulation of the wind field in the study area.

[0019] S13. Wind field information extraction and processing:

[0020] After completing the wind environment simulation based on the WRF / CALMET model, the simulation results data were extracted and processed. Wind speed, wind direction and wind frequency information were extracted from the model output and converted into ncl format files for Python to read.

[0021] First, by writing Python scripts, we read the wind speed, wind direction, and wind frequency data generated by the WRF and CALMET models, and then use the NumPy and SciPy libraries in the Python environment to calculate the average wind speed and wind direction.

[0022] After obtaining the preliminary processed wind speed, wind direction, and wind frequency data, ArcGIS software is used for geospatial processing and analysis to generate a wind speed distribution map within the study area. The average wind speed is then extracted and recalculated using a raster calculation tool.

[0023] S2. Calculation of the windward area index (FAI) for the area surrounding the site based on WRF / CALMET wind field information:

[0024] S21. Wind direction and frequency data are combined with wind field data simulated by the WRF / CALMET model to calculate the wind frequency for each wind direction. The calculation formula is as follows:

[0025] ;

[0026] In the formula: A m The frequency of wind occurrence at direction m;

[0027] K m This represents the number of times the wind was recorded at direction m.

[0028] C represents the number of times calm winds occurred;

[0029] S22. Selection of raster format and resolution for FAI calculation in the site area:

[0030] Using ArcGIS netting tool, select a 30m grid building to calculate the windward area index;

[0031] S23. Calculation of Windward Area Index (FAI): The formula for calculating the windward area index is as follows:

[0032] ;in,

[0033] ;

[0034] The windward area index of a building represents a single wind direction, which is the ratio of the area of ​​the building's windward surface perpendicular to a certain wind direction to the area of ​​the land on which the building is located. It is also known as the FAI for a specific wind direction.

[0035] The frequency represents a specific wind direction. The larger the windward area index, the stronger the wind resistance of the area, and vice versa.

[0036] It is the average area of ​​the research site;

[0037] Indicates wind direction;

[0038] It is the total area of ​​each building or hillside projected onto a plane perpendicular to the direction of the incoming wind;

[0039] Represents the width of the building's projection onto a plane perpendicular to the direction of the incoming wind;

[0040] This represents the height of a building's projection onto a plane perpendicular to the direction of the incoming wind.

[0041] Next, using a Geographic Information System (GIS), the windward area index (FAI) data was visualized and the average FAI value for each station was calculated.

[0042] In step S11, the parameterization schemes for the WRF model simulation include: microphysical process scheme, atmospheric radiation scheme, land surface process scheme, planetary boundary layer scheme, and cumulus convection scheme.

[0043] The Thompson microphysical process scheme was selected to simulate the formation process of clouds and precipitation.

[0044] The atmospheric radiation scheme selected is the RRTMG long-wave and short-wave radiation scheme, in which the long-wave radiation scheme uses the fast radiative transfer model; the short-wave radiation scheme is selected to simulate the radiative transfer process in the atmosphere, including the absorption, scattering and emission processes of solar radiation and Earth's long-wave radiation.

[0045] The Noah-NP land surface process scheme was selected to simulate the energy and moisture exchange processes between the Earth's surface and the atmosphere, thus improving the accuracy of land surface process simulation.

[0046] The planetary boundary layer scheme selected is the Yonsei University planetary boundary layer scheme, which is used to describe the turbulent exchange process within the atmospheric boundary layer.

[0047] The Kain-Fritsch cumulus convection scheme was selected to simulate the development of convective clouds, especially in areas with frequent convective activity.

[0048] In the micro-meteorological module of step S12, the simulation data is based on digital elevation model data with a resolution of 30m and land use data with a resolution of 30m, and finally wind field information with a resolution of 100m is obtained.

[0049] In step S23, the windward area index (FAI) is calculated for each of the 146 station areas.

[0050] Beneficial effects:

[0051] First, the method for evaluating the wind environment of rail transit station areas based on the combination of WRF / CALMRT model simulation and GIS windward area index (FAI) calculation provided in this embodiment of the invention achieves the acquisition of urban wind field information with an accuracy of 100m within a city range of tens of kilometers through WRF / CALIMET simulation. At the same time, the simulated wind field information is used as boundary conditions to calculate the windward area index (FAI) of the station area. The two are combined to comprehensively evaluate the wind environment of the station.

[0052] Second, this method combines the two, which solves the problems of WRF / CALMET model surface data being difficult to consider urban buildings and computer computing power limitations, and also solves the problem of inaccurate wind field data in FAI calculation, thus realizing rapid simulation of high-precision wind field information in the site area and comprehensive evaluation of the wind environment in the site area. Attached Figure Description

[0053] Figure 1 The flowchart for constructing the wind environment assessment model of this invention;

[0054] Figure 2Wind speed distribution map of Nanjing City simulated by WRF / CALMET — spatial distribution of wind speed calculated by GIS using Kriging interpolation;

[0055] Figure 3 This is a schematic diagram of the wind environment assessment model for the site area. Detailed Implementation

[0056] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0057] like Figure 1 As shown, the wind environment assessment method for railway station areas integrating WRF / CALMET simulation and FAI calculation provided by this invention mainly includes two steps: first, urban wind environment simulation is performed based on the WRF / CALMET model; second, the windward area index (FAI) of the area surrounding the station is calculated by combining the simulated wind field information; and finally, the wind environment is assessed by combining both. The following detailed description of each step is provided with reference to specific embodiments:

[0058] Step 1: Urban wind environment simulation based on WRF / CALMET model

[0059] ① Mesoscale WRF model

[0060] This paper uses WRF version 4.2, with initial field and lateral boundary conditions derived from ERA5 reanalysis data. The simulation period spans three months, from August to October 2022. The simulation was conducted after the summer rainy season, when wind speeds are relatively stable and more valuable for reference. The simulation region is centered at (31.95°N, 118.77°E), and a three-layer nested grid of 27 km × 9 km × 3 km was used. Figure 2 The resolutions were 9 km, 3 km, and 1 km, with grid numbers of 120×120, 166×166, and 154×323, respectively. The parameterization schemes for the WRF simulations were Thompson microphysics process scheme, RRTMG long-wave and short-wave radiation scheme, Noah-MP land surface process scheme, YSU planetary boundary layer scheme, and Kain-Fritsch cumulus convection scheme, respectively, with specific settings as shown in the table below.

[0061]

[0062] ② Downscaling CALMET model:

[0063] CALMET comprises a diagnostic wind field module and a micrometeorology module. The diagnostic wind field module uses a 1km WRF meteorological field as the initial inferred wind field, adjusting for topographic dynamics, slope flow, and topographic blocking effects to generate the first-stage wind field. After importing observational data, interpolation and smoothing are performed to generate the final wind field. The micrometeorology module then describes the boundary layer structure using parametric methods. A refined diagnosis of the wind field in Nanjing city is performed, with a horizontal grid of 600×500 pixels and a grid spacing of 100m. Vertically, it is divided into 13 layers, with the lowest layer at a 10m elevation. The simulation data uses 30m DEN data and 30m land use data, with a resolution of 100m. Downscaling calculations are then performed to obtain wind field information at a resolution of 100m.

[0064] ③ Data extraction and processing of wind environment simulation results:

[0065] The data was processed using Python tools to read, interpolate, and post-process the WRF and CALMET patterns, and the monthly average and three-month average wind speed were calculated. Hourly wind direction data from August to October were also extracted.

[0066] Based on this, the wind field information data was processed using ArcGIS 10.4 to obtain average wind speed point feature data for the Nanjing city area, totaling 612,400 data points. These data points containing average wind speed information were used as the raw data for spatial interpolation. The Kriging interpolation tool, included in the Geostatistical Analyst toolbox of ArcGIS 10.4, was used to perform spatial interpolation of the wind speed information, obtaining wind speed information for the Nanjing city area and the station locations, such as... Figure 2 As shown.

[0067] Step 2: Calculate FAI (Factor Area Intelligence) for the area surrounding the Geographic Information System (GIS) site:

[0068] Based on hourly wind direction data from WRF-CALMET simulations, this study used Python tools to extract hourly wind direction data from August to October, and then filtered the wind direction data for each station area, calculating wind frequency data for 16 wind directions for each station. This wind frequency data was used as the wind field boundary condition for the windward area index. Using spatial analysis and raster calculation tools from a Geographic Information System (GIS), natural terrain and urban buildings were taken into account to calculate the windward area index for the areas surrounding the rail transit stations.

[0069] Using the above method, the windward area index of 146 subway stations was calculated, and the results were visualized in GIS. Figure 3 .

[0070] The formula for calculating the windward area index is:

[0071] ,in,

[0072] ,

[0073] The windward area index of a building represents a single wind direction, which is the ratio of the area of ​​the building's windward surface perpendicular to a certain wind direction to the area of ​​the land on which the building is located. It is also known as the FAI for a specific wind direction.

[0074] The frequency represents a specific wind direction. The larger the windward area index, the stronger the wind resistance of the area, and vice versa.

[0075] It is the average area of ​​the research site;

[0076] Indicates wind direction;

[0077] It is the total area of ​​each building or hillside projected onto a plane perpendicular to the direction of the incoming wind;

[0078] Represents the width of the building's projection onto a plane perpendicular to the direction of the incoming wind;

[0079] This represents the height of a building's projection onto a plane perpendicular to the direction of the incoming wind.

[0080] .

Claims

1. A method for evaluating the wind environment in railway station areas that integrates WRF / CALMET simulation and FAI calculation, characterized in that, A multi-scale composite simulation method—the WRF / CALMET model—was used to simulate the wind field, obtaining detailed wind field information for the entire city and the areas surrounding subway stations. Simultaneously, the windward area index (FAI) was introduced. A GIS database was constructed, and the FAI for the study area was calculated using a digital elevation model (DEM), building data, and simulated meteorological data. Finally, average wind speed and windward area index were selected as two indicators for descriptive analysis, analyzing their respective numerical and spatial distribution characteristics. k-means cluster analysis was then applied to both indicators to comprehensively evaluate the wind environment of the station area. The specific steps included are as follows: S1. Urban wind environment simulation based on WRF / CALMET model: S11, mesoscale WRF model; The WRF model was used for simulation analysis. The initial field and lateral boundary conditions of the WRF model were derived from ERA5 reanalysis data. S12, downscaled CALMET model; Based on the large-scale meteorological data provided by the WRF model, the CALMET model is used for downscaling. The CALMET model includes two core components: a diagnostic wind field module and a micrometeorological module. First, using the diagnostic wind field module, the CALMET model uses meteorological field data with a resolution of 1 km from the WRF model simulation results as the initial guessed wind field. Based on this, by considering topographic dynamics adjustment, slope flow adjustment, and topographic blocking effect, the CALMET model can generate a first-step wind field that is more adapted to the complexity of the terrain. Subsequently, the CALMET model further incorporates measured observation data and, through fine interpolation and smoothing, finally generates wind field information with higher accuracy. In the micro-meteorology module, the CALMET model uses a parametric method to describe the structure of the atmospheric boundary layer. By performing targeted wind field diagnosis on the study area, the CALMET model uses a 100m grid spacing and divides the grid according to the study area. The lowest layer in the vertical direction is located 10m above the ground, thereby achieving a refined simulation of the wind field in the study area. S13, Wind field information extraction and processing; After completing the wind environment simulation based on the WRF / CALMET model, the simulation results data were extracted and processed. Wind speed, wind direction and wind frequency information were extracted from the model output and converted into ncl format files for Python to read. First, by writing Python scripts, we read the wind speed, wind direction, and wind frequency data generated by the WRF and CALMET models, and then use the NumPy and SciPy libraries in the Python environment to calculate the average wind speed and wind direction. After obtaining the preliminary processed wind speed, wind direction, and wind frequency data, ArcGIS software is used for geospatial processing and analysis to generate a wind speed distribution map within the study area. The average wind speed is then extracted and recalculated using a raster calculation tool. S2. Calculation of the windward area index (FAI) for the area surrounding the site based on WRF / CALMET wind field information: S21. Wind direction and frequency data are combined with wind field data simulated by the WRF / CALMET model to calculate the wind frequency for each wind direction. The calculation formula is as follows: ； In the formula: A m The frequency of wind occurrence at direction m; K m This represents the number of times the wind was recorded at direction m. C represents the number of times calm winds occurred; S22. Selection of raster format and resolution for FAI calculation in the site area: Using ArcGIS netting tool, select a 30m grid building to calculate the windward area index; S23. Calculation of Windward Area Index (FAI): The formula for calculating the windward area index is as follows: ,in, ； The windward area index of a building represents a single wind direction, which is the ratio of the area of ​​the building's windward surface perpendicular to a certain wind direction to the area of ​​the land on which the building is located. It is also known as the FAI for a specific wind direction. The frequency represents a specific wind direction. The larger the windward area index, the stronger the wind resistance of the area, and vice versa. It is the average area of ​​the research site; Indicates wind direction; It is the total area of ​​each building or hillside projected onto a plane perpendicular to the direction of the incoming wind; Represents the width of the building's projection onto a plane perpendicular to the direction of the incoming wind; This represents the height of a building's projection onto a plane perpendicular to the direction of the incoming wind. Next, using a Geographic Information System (GIS), the windward area index (FAI) data was visualized and the average FAI value for each station was calculated. By using WRF / CALMET with 100m resolution wind field data as FAI input and combining it with a 30m grid to calculate FAI, high-precision wind field information of the station area was rapidly simulated and a comprehensive evaluation of the wind environment of the station area was achieved. In step S11, the parameterization schemes for the WRF model simulation include: microphysical process scheme, atmospheric radiation scheme, land surface process scheme, planetary boundary layer scheme, and cumulus convection scheme. The Thompson microphysical process scheme was selected to simulate the formation process of clouds and precipitation. The atmospheric radiation scheme selected is the RRTMG long-wave and short-wave radiation scheme, in which the long-wave radiation scheme uses the fast radiative transfer model; the short-wave radiation scheme is selected to simulate the radiative transfer process in the atmosphere, including the absorption, scattering and emission processes of solar radiation and Earth's long-wave radiation. The Noah-NP land surface process scheme was selected to simulate the energy and moisture exchange processes between the Earth's surface and the atmosphere, thus improving the accuracy of land surface process simulation. The planetary boundary layer scheme selected is the Yonsei University planetary boundary layer scheme, which is used to describe the turbulent exchange process within the atmospheric boundary layer. The Kain-Fritsch cumulus convection scheme was selected to simulate the development process of convective clouds. In the micro-meteorological module of step S12, the simulation data is based on digital elevation model data with a resolution of 30m and land use data with a resolution of 30m, and finally wind field information with a resolution of 100m is obtained. In step S23, the windward area index (FAI) is calculated for each of the 146 station areas.

Images