Downscaling bridge wind temperature coupling numerical simulation method considering real terrain

Abstract

The invention provides a downscaling bridge wind temperature coupling numerical simulation method considering a real terrain, and belongs to the technical field of bridge structure design. Comprising the steps of determining a calculation area of a mesoscale model according to a bridge geographic position, and adopting multi-layer grid nesting to improve calculation precision; obtaining terrain static data and initial meteorological field data, and configuring a physical scheme; establishing a geometric model of the small-scale model through three-dimensional modeling software; according to a small-scale model, calculating a domain inlet surface form, and performing fitting block division and extraction point planning; extracting inlet wind speed and temperature data by adopting a Cressman interpolation method through the mesoscale model, and fitting by adopting a least square method to obtain a small-scale model inlet surface wind speed and temperature function; according to the material characteristics of the bridge structure, loading a geometric model and an inlet surface wind speed and temperature function, establishing a small-scale model by using a CFD method, and solving environmental wind, a temperature field and structural response; the simulation authenticity is improved, and the redundancy of simulation work is reduced.

Description

Technical Field

[0001] The invention provides a downscaled bridge wind-temperature coupling numerical simulation method considering real terrain, belonging to the technical field of bridge structure design. Background Art

[0002] With the continuous development of bridge engineering, the design, construction and maintenance of bridge structures are facing more and more complex environmental factors. Among them, wind and temperature are the main environmental factors that have adverse effects on bridges, significantly affecting the safety and durability of bridge structures. In the actual environment, especially in complex terrains such as rugged mountainous areas, there is a complex process of atmospheric dynamics and thermal coupling. Air pressure will change due to temperature. Hot air rises and cold air falls. The pressure difference leads to the formation of wind. At the same time, wind affects the temperature and thus the air pressure, thus forming a wind-temperature coupling effect in the environment. In addition, most rugged mountainous areas have steep terrain and undulating terrain, which makes the adjacent valleys have significant height differences. At the same time, due to the diverse landforms, the wind field and temperature field distribution characteristics are extremely complex. For bridge structures, the temperature field in the natural environment mainly transfers heat to the structure through convection heat transfer and radiation heat transfer (the process of heat exchange between two objects with different temperatures and non-contacting each other through electromagnetic waves). The structure is constrained, resulting in the generation of temperature-induced effects (structural deformation caused by uneven temperature differences), which may cause the structure to deform too much or even crack. The wind field directly acts on the surface of the structure and is easily affected by the terrain environment, thus affecting the wind field distribution near the structure, and then affecting the convective heat transfer process on the surface of the structure. Therefore, ignoring the real terrain conditions and the wind-temperature coupling in the environment may lead to a large deviation between the simulation results and the actual situation, thus failing to reflect the real response of the structure.

[0003] In order to analyze the response of bridges under wind-temperature coupling in complex environments, it is particularly important to develop a downscaled bridge wind-temperature coupling numerical simulation method that considers the real terrain.

[0004] CN114139263A discloses a bridge wind-temperature coupling numerical simulation method that takes into account the local wind field of the bridge deck. The height of the thermal boundary layer on the surface of the bridge structure is calculated through the thermal boundary layer theory formula; the wind speed at the height of the thermal boundary layer is solved through the wind field model; the wind speed data at the bridge entrance and the thermal boundary layer height are fitted by the least squares method to obtain the reduction coefficient; the ambient temperature of the bridge site and the thermal radiation value of the bridge are obtained; a bridge wind-temperature coupling numerical model is established, and the parameters of the bridge wind-temperature coupling numerical model are fitted; and the temperature field of the bridge structure is solved using the finite element method. This method has the following technical disadvantages:

[0005] (1) The cost of temperature acquisition is high and not realistic enough: The acquisition of environmental temperature relies heavily on theoretical formulas and requires the actual measurement of the daily maximum and minimum temperatures. The temperature sample density is low and cannot reflect the actual temperature physical field.

[0006] (2) The wind field model is not realistic and accurate enough: The simulation of the structural wind field does not take into account the real environmental factors (such as the coupling effect between temperature and wind field) and the influence of terrain. The inlet wind profile of the wind field model is not realistic and cannot reflect the real wind field distribution.

[0007] (3) The analysis of temperature-induced effects of structures is not accurate enough: After using this technology to obtain the temperature field of the structure, it is necessary to extract the temperature load and then load the structure. In this process, only limited discrete temperature load boundaries can be extracted, and the load field extraction points are not comprehensive enough; the bidirectional coupling between the load field and the structure is not considered, and only the result of the unidirectional action of the temperature load field on the structure can be reflected. Summary of the invention

[0008] The present invention provides a downscaled bridge wind-temperature coupling numerical simulation method considering the real terrain. The technical problems to be solved by the present invention are:

[0009] (1) Obtaining real, accurate, and detailed ambient temperature and wind speed: First, the real mesoscale ambient temperature and wind speed of the required time and area are obtained through WRF model simulation; a small-scale model is established based on the bridge structure and terrain through the CFD method, where the inlet boundary conditions are obtained by extracting the temperature and wind speed results in the mesoscale and fitting them with the least squares method, and the fitting process is divided into blocks according to the terrain trend of the inlet surface to improve the simulation accuracy; the simulation area of ​​the mesoscale model is large enough to make the inlet conditions of the small-scale model more real and reasonable, thereby driving the calculation and solution of the small-scale model to obtain the real ambient wind field and temperature field at any position in the simulation domain. The above process takes into account the real terrain factors and the real ambient wind-temperature coupling process.

[0010] (2) Consideration of real-time bidirectional coupling between load field and structure: In the small-scale model, the bidirectional coupling between environment and structure is considered, so that the wind-temperature coupling effect of the environment and the response of the structure can be considered simultaneously during the calculation. There is no need to calculate the environmental field first and then load it on the structure, which improves the calculation accuracy and reduces the complexity of the simulation work.

[0011] The specific technical solution is: a downscaling bridge wind-temperature coupling numerical simulation method considering real terrain, including the following steps:

[0012] (1) The calculation area of ​​the mesoscale model is determined according to the geographical location of the bridge, and multi-layer grid nesting is used to improve the calculation accuracy.

[0013] A three-layer bidirectional nested grid (d01, d02, d03) scheme is used for simulation. The horizontal resolution of the grid from outside to inside is 9km, 3km, and 1km, respectively, that is, the outermost d01 grid spacing is 9km. The number of vertical grid layers in the computational domain is set to 50 layers, the top pressure in the model is set to 5000Pa, and the default bottom pressure in the mesoscale model is 1000Pa. The vertical grid is encrypted below 1km, 14 layers of encrypted grid are set, and the bottom grid height is 25m;

[0014] The time integration step was set to 1 s, and the Lambert projection scheme was used for map projection during simulation.

[0015] (2) Obtain static terrain data and initial meteorological field data and configure the physical solution.

[0016] In the WRF model calculation process, two types of initial data need to be input first, namely static terrain data and initial meteorological field data. The complete three-dimensional initial data in the calculation area are obtained through the built-in integration and interpolation programs of the model. The data of the initial meteorological field needs to select the file of the target simulation period and load it in the WRF model. The planetary boundary layer scheme adopts the YSU scheme, the RRTM longwave radiation scheme, the Dudhia shortwave radiation scheme, the Goddard microphysical process scheme, the Betts-Miller-Janjic cumulus convection parameterization scheme, the Monin-Obukhov (modified MM5) near-surface parameterization scheme, and the Noah land surface process scheme. Only the first layer of grids opens the cumulus convection scheme.

[0017] (3) Establish the geometric model of the small-scale model through 3D modeling software.

[0018] The geometric models in the small-scale model are divided into terrain models and structural models.

[0019] The process of establishing the terrain model is as follows: obtain the digital elevation model (DEM) of the bridge site area, import it into the Globalmapper software for slicing, and obtain the terrain data information of the small-scale simulation domain. The small-scale model uses the (x, y, z) three-dimensional coordinate form. Before loading it into the small-scale model, spatial coordinate conversion is required, as shown in the following formulas (1) to (2). Then, the three-dimensional modeling software Rhino is used to establish the terrain surface.

[0020]

[0021] In the formula are the longitude and latitude coordinates in the WGS84 coordinate system, (x, y) are the coordinates in the Mercator projection coordinate system, R is the mean radius of the earth, and e is the first eccentricity of the earth ellipsoid.

[0022] The structural model is a solid model created by the 3D modeling software Rhino based on the structural geometry information.

[0023] (4) According to the inlet surface morphology of the small-scale model calculation domain, fitting block division and extraction point planning are carried out.

[0024] When extracting the results of the mesoscale model, the fitting blocks are divided and the extraction points are planned according to the terrain trend and the size of the small-scale model inlet surface area.

[0025] (5) Through the mesoscale model, the Cressman interpolation method is used to extract the inlet wind speed and temperature data, and the least squares method is used to fit the inlet wind speed and temperature functions of the small-scale model.

[0026] The Cressman interpolation calculation formula is as follows:

[0027]

[0028] In the formula, w ij is the weight between the interpolation point and the observation point, d ij is the Euclidean distance between the interpolation point and the observation point, D is the step function threshold, v i represents the interpolation result of the interpolation point, v j represents the observed value of the sample point, and n represents the number of sample points.

[0029] The least squares method is used to approximate the function of the extracted data to obtain the fitting function of the inlet surface. The least squares method implementation process is as follows:

[0030] Given data α0(x), α1(x), …, α n (x) Find the function that minimizes the sum of squared errors as shown in formula (5) from the function class The function obtained is the least squares solution of this function class under this set of data.

[0031]

[0032] (6) According to the material properties of the bridge structure, the geometric model and the inlet surface wind speed and temperature functions are loaded, and the small-scale model is established using the CFD method to solve the ambient wind, temperature field and structural response.

[0033] The geometric model was meshed using ANSYS software. The geometric model was imported into ANSYS software and further processed in the SpaceClaim module. First, the surface quality of the model was checked and repaired, then data coupling interaction was established between the fluid and solid contact surfaces, the boundary surfaces were grouped, and then meshing was performed in the Fluentmeshing module. The entire model uses a tetrahedral unstructured grid. The terrain boundary, bridge tower structure, and nearby fluid domains were meshed. The first layer of mesh height on the terrain surface is 0.1m, and the boundary layer growth rate is 1.1.

[0034] The CFD software Fluent was used to establish a small-scale model. Velocity-inlet and temperature-inlet were selected for the inlet boundary conditions. The initial information was set by writing a UDF (User Define Function) program based on the fitted inlet surface wind speed and temperature functions. The fluid was considered to be fully developed at the outlet, and the north side of the calculation model was set as the pressure outlet boundary (Pressure-outlet). The ground and bridge tower structure surfaces of the calculation model were set as non-slip wall boundaries (Non-slip Wall); the top surface of the model and the east and west boundary surfaces were set as symmetry boundaries (Symmetry).

[0035] Perform FSI (Fluid Structure Interaction) definition of the structural wall boundary and the spatial domain.

[0036] The calculation selects the pressure-based steady-state solver suitable for low-speed incompressible flow, the turbulence model and turbulence solution select SST k-ω and SIMPLEC algorithm, the discrete format selects the second-order upwind format, and the standard wall function selects.

[0037] The technical effects of the present invention are as follows:

[0038] A mesoscale model is established through the WRF mode, taking into account the real environmental terrain and rich meteorological physics schemes, and the real mesoscale environmental temperature and wind speed of the required time and area are obtained. In addition, since the simulation area of ​​the mesoscale model is large enough, the inlet conditions of the small-scale model are more realistic and reasonable, which in turn drives the calculation and solution of the small-scale model.

[0039] Through the CFD method, a small-scale model was established according to the actual bridge structure and terrain. The inlet boundary conditions were obtained by extracting the temperature and wind speed results in the mesoscale model and fitting them using the least squares method. In the fitting process, block fitting was performed according to the terrain trend of the inlet surface. This can to a certain extent solve the boundary wind speed distortion problems caused by excessively high surface wind speed and missing wind speed values ​​that generally exist when setting the model inlet, thereby improving the simulation accuracy.

[0040] In the small-scale model, the two-way coupling between the environment and the structure is taken into account, so that the wind-temperature coupling results of the environment and the response of the structure can be obtained simultaneously during the calculation, which improves the simulation authenticity and reduces the complexity of the simulation work. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 is a flow chart of the method of the present invention;

[0042] Figure 2 It is a schematic diagram of a multi-layer nested grid of a simulation area of ​​the present invention;

[0043] Figure 3 It is a schematic diagram of establishing the overall geometric model of the present invention;

[0044] Figure 4 It is a schematic diagram of the entry surface block and data extraction point distribution of the present invention;

[0045] Figure 5 It is a schematic diagram of the overall watershed grid of the present invention;

[0046] Figure 6 is a schematic diagram of a ground surface grid of the present invention;

[0047] Figure 7 is a schematic diagram of a boundary layer grid of the present invention;

[0048] Figure 8 is a schematic diagram of a bridge tower structure grid of the present invention;

[0049] Fig. 9 It is a schematic diagram of setting the solution parameters of the present invention. DETAILED DESCRIPTION

[0050] The specific technical solution of the present invention is described in conjunction with the accompanying drawings.

[0051] like Figure 1 As shown, a downscaling bridge wind-temperature coupling numerical simulation method considering real terrain includes the following steps:

[0052] (1) The calculation area of ​​the mesoscale model is determined according to the geographical location of the bridge, and multi-layer grid nesting is used to improve the calculation accuracy.

[0053] The mesoscale model is suitable for simulation studies with a resolution of 1 to 10 km. It uses a three-layer bidirectional nested grid (d01, d02, and d03) scheme for simulation. The horizontal resolutions of the grids from outside to inside are 9 km, 3 km, and 1 km, respectively, that is, the outermost d01 grid spacing is 9 km. The number of vertical grid layers in the computational domain is set to 50 layers, the top layer pressure in the model is set to 5000Pa, and the default bottom layer pressure in the mesoscale model is 1000Pa. Since the wind field within the atmospheric boundary layer height range is mainly affected by near-ground radiation, and the simulation height in the small-scale model is generally about 1km, the vertical grid is encrypted below 1km, and 14 layers of encrypted grids are set, with the lowest grid height of 25m. Improve the vertical resolution of the model in the near-ground layer to better transmit effective data for the small-scale wind-temperature coupling model.

[0054] The time integration step was set to 1 s, and the Lambert projection scheme was used for map projection during simulation.

[0055] The mesoscale model simulation domain has multiple layers of nested grids such as Figure 2 shown.

[0056] (2) Obtain static terrain data and initial meteorological field data and configure the physical solution.

[0057] In the calculation process of WRF model, two types of initial data need to be input first, namely static terrain data and initial meteorological field data. The complete three-dimensional initial data in the calculation area are obtained through the built-in integration and interpolation programs of the model. The data are obtained from the open source websites https: / / www2.mmm.ucar.edu / wrf / users / download / and https: / / rda.ucar.edu / datasets / . The data of the initial meteorological field need to select the file of the target simulation period and load it in the WRF model. The planetary boundary layer scheme adopts the YSU scheme, the RRTM longwave radiation scheme, the Dudhia shortwave radiation scheme, the Goddard microphysical process scheme, the Betts-Miller-Janjic cumulus convection parameterization scheme, the Monin-Obukhov (modified MM5) near-surface parameterization scheme, and the Noah land surface process scheme, in which the cumulus convection scheme is only turned on in the first layer of grids. The specific physical scheme parameter settings are shown in Table 1.

[0058] Table 1 Physical scheme parameter settings

[0059]

[0060]

[0061] (3) Establish the geometric model of the small-scale model through 3D modeling software.

[0062] The geometric model in the small-scale model is divided into a terrain model and a structural model. The process of establishing the terrain model is as follows: obtain the digital elevation model (DEM) of the bridge site area, import it into the Globalmapper software for slicing, and obtain the terrain data information of the small-scale simulation domain. The DEM represents the ground elevation in the form of a set of ordered numerical arrays, which includes factors such as elevation, slope, and slope aspect. It can be obtained through the Geospatial Data Cloud Platform of the Computer Network Information Center of the Chinese Academy of Sciences (http: / / www.gscloud.cn). Since the spatial position of the mesoscale model and DEM data are both in the WGS84 coordinate system, the small-scale model uses the (x, y, z) three-dimensional coordinate form. Therefore, spatial coordinate conversion is required before loading into the small-scale model, as shown in the formulas (1) to (2). Then, the three-dimensional modeling software Rhino is used to establish the terrain surface.

[0063]

[0064] In the formula are the longitude and latitude coordinates in the WGS84 coordinate system, (x, y) are the coordinates in the Mercator projection coordinate system, R is the mean radius of the earth, and e is the first eccentricity of the earth ellipsoid.

[0065] The structural model is a solid model created by the 3D modeling software Rhino based on the structural geometry information. The final geometric model is as follows Figure 3 shown.

[0066] (4) According to the inlet surface morphology of the small-scale model calculation domain, fitting block division and extraction point planning are carried out.

[0067] When extracting the results of the mesoscale model, the fitting blocks should be divided and the extraction points should be planned according to the terrain trend and the size of the small-scale model inlet surface area. The inlet surface fitting block division and data extraction point distribution are shown in the figure below. Figure 4 As shown in the figure, P1 to P16 only represent the schematic extraction point positions, that is, 16 mesoscale result wind profiles and temperature profiles, and the number of horizontal and vertical extraction points should be set according to the simulation requirements. Especially in complex canyon terrain with large terrain undulations, this number of wind profiles can reflect good accuracy in the small-scale model boundary assignment in the geometric height coordinate system, and to a certain extent solve the boundary wind speed distortion problem of excessive surface wind speed and missing wind speed values ​​that generally exists when setting the model entrance.

[0068] (5) Through the mesoscale model, the Cressman interpolation method is used to extract the inlet wind speed and temperature data, and the least squares method is used to fit the inlet wind speed and temperature functions of the small-scale model.

[0069] Since the D03 simulation area accuracy of the mesoscale wind-temperature coupling model is 1 km, when extracting the data of the required point in the result file, the point may not be exactly located at the center or grid point of the grid, but at a certain position within the grid. Therefore, it is necessary to use the spatial data interpolation method to obtain the wind speed and temperature data corresponding to the required data extraction point.

[0070] The Cressman interpolation method is a grid point value estimation method. Its basic idea is to multiply all observations around the interpolation point by a weight coefficient, and then use the weight value as a unit step function to segment the space. This method is easy to implement and calculate, can simultaneously consider the influence of direction and distance, can better handle small samples and irregularly distributed data, and is insensitive to isolated values ​​compared to some other interpolation algorithms. Overall, it has the significant advantages of simplicity and efficiency.

[0071] The calculation formula is as follows:

[0072]

[0073] In the formula, w ij is the weight between the interpolation point and the observation point, d ij is the Euclidean distance between the interpolation point and the observation point, D is the step function threshold, v i represents the interpolation result of the interpolation point, v j represents the observed value of the sample point, and n represents the number of sample points.

[0074] The least squares method is used to approximate the function of the extracted data to obtain the fitting function of the inlet surface. The least squares method implementation process is shown in the following example. Given the data α0(x), α1(x), …, α n (x), need to Find the function that minimizes the sum of squared errors as shown in formula (5) from the function class The function obtained is the least square solution of this function class under this set of data. This process can be realized by Matlab software.

[0075]

[0076] (6) According to the material properties of the bridge structure, the geometric model and the inlet surface wind speed and temperature functions are loaded, and the small-scale model is established using the CFD method to solve the ambient wind, temperature field and structural response.

[0077] The geometric model was meshed using ANSYS software. The geometric model was imported into ANSYS software and further processed in the SpaceClaim module. First, the surface quality of the model was checked and repaired, and then data coupling interaction was established between the fluid and solid contact surfaces. The boundary surfaces were grouped and then meshed in the Fluentmeshing module. The entire model uses a tetrahedral unstructured grid. Due to the large fluctuations in the terrain in the computational domain, the wind field in the near-ground boundary layer is highly sensitive to the terrain, and the bridge tower scale is relatively small relative to the size of the overall model, so the terrain boundary, bridge tower structure, and nearby fluid domain are meshed. The first layer of grid height on the terrain surface is 0.1m, and the boundary layer growth rate is 1.1.

[0078] The grid division is as follows Figures 5 to 8 shown.

[0079] The CFD software Fluent was used to establish a small-scale model. Velocity-inlet and temperature-inlet were selected for the inlet boundary conditions. The initial information was set by writing a UDF (User Define Function) program based on the fitted inlet surface wind speed and temperature functions. The fluid was considered to be fully developed at the outlet, and the north side of the calculation model was set as the pressure outlet boundary (Pressure-outlet). The ground and bridge tower structure surfaces of the calculation model were set as non-slip wall boundaries (Non-slip Wall); the top surface of the model and the east and west boundary surfaces were set as symmetry boundaries (Symmetry).

[0080] Perform FSI (Fluid Structure Interaction) definition of the structural wall boundary and the spatial domain.

[0081] The calculation selects the pressure-based steady-state solver suitable for low-speed incompressible flow, the turbulence model and turbulence solution select SST k-ω and SIMPLEC algorithm, the discrete format selects the second-order upwind format, and the standard wall function selects.

[0082] The schematic diagram of solution parameters is as follows Fig. 9 shown.

Claims

1. A downscaling bridge wind-temperature coupling numerical simulation method considering real terrain, characterized in that: The following steps are involved: (1) Determine the calculation area of ​​the mesoscale model according to the geographical location of the bridge, and use multi-layer grid nesting to improve the calculation accuracy; (2) Obtain static terrain data and initial meteorological field data and configure the physical scheme; (3) Establish the geometric model of the small-scale model through 3D modeling software; (4) According to the inlet surface morphology of the small-scale model calculation domain, the fitting block is divided and the extraction point is planned; (5) Through the mesoscale model, the Cressman interpolation method is used to extract the inlet wind speed and temperature data, and the least squares method is used to fit the inlet wind speed and temperature functions of the small-scale model; (6) According to the material properties of the bridge structure, the geometric model and the inlet surface wind speed and temperature functions are loaded, and the small-scale model is established using the CFD method to solve the ambient wind, temperature field and structural response.

2. According to the downscaling bridge wind-temperature coupling numerical simulation method considering real terrain according to claim 1, it is characterized in that: The specific method of step (1) is: A three-layer bidirectional nested grid scheme is used for simulation. The three layers of bidirectional nested grids are d01, d02, and d03. The horizontal resolutions of the grids from outside to inside are 9km, 3km, and 1km, respectively, that is, the grid spacing of the outermost d01 is 9km; the number of vertical grid layers in the calculation domain is set to 50 layers, the top layer air pressure in the model is set to 5000Pa, and the default bottom layer air pressure in the mesoscale model is 1000Pa; the vertical grid is encrypted below 1km, 14 layers of encrypted grids are set, and the bottom layer grid height is 25m; The time integration step was set to 1 s, and the Lambert projection scheme was used for map projection during simulation.

3. The downscaling bridge wind-temperature coupling numerical simulation method considering real terrain according to claim 1 is characterized in that: The specific method of step (2) is: During the WRF model calculation process, two types of initial data need to be input first, namely static terrain data and initial meteorological field data. The complete three-dimensional initial data in the calculation area are obtained through the built-in integration and interpolation programs in the model; the initial meteorological field data needs to select the file of the target simulation period and load it in the WRF model; the planetary boundary layer scheme adopts the YSU scheme, the RRTM long-wave radiation scheme, the Dudhia short-wave radiation scheme, the Goddard microphysical process scheme, the Betts-Miller-Janjic cumulus convection parameterization scheme, the Monin-Obukhov near-surface parameterization scheme, and the Noah land surface process scheme, among which the cumulus convection scheme is only turned on for the first layer of grids.

4. The downscaling bridge wind-temperature coupling numerical simulation method considering real terrain according to claim 1 is characterized in that: The specific method of step (3) is: The geometric model in the small-scale model is divided into terrain model and structural model. The process of establishing the terrain model is as follows: obtain the digital elevation model DEM of the bridge site area, import it into the Globalmapper software for slicing, and obtain the terrain data information of the small-scale simulation domain; the small-scale model adopts the (x, y, z) three-dimensional coordinate form; before loading it into the small-scale model, spatial coordinate conversion is required, as shown in the following formulas (1) to (2); and then use the three-dimensional modeling software Rhino to establish the terrain surface; In the formula is the longitude and latitude coordinates in the WGS84 coordinate system, (x, y) is the coordinates in the Mercator projection coordinate system, R is the mean radius of the earth, and e is the first eccentricity of the earth ellipsoid; The structural model is a solid model created by the 3D modeling software Rhino based on the structural geometry information.

5. The downscaling bridge wind-temperature coupling numerical simulation method considering real terrain according to claim 1 is characterized in that: The specific method of step (4) is as follows: when extracting the results of the mesoscale model, first divide the fitting blocks and plan the extraction points according to the terrain trend and the size of the small-scale model inlet surface area.

6. The downscaling bridge wind-temperature coupling numerical simulation method considering real terrain according to claim 1 is characterized in that: The Cressman interpolation calculation formula in step (5) is as follows: In the formula, w ij is the weight between the interpolation point and the observation point, d ij is the Euclidean distance between the interpolation point and the observation point, D is the step function threshold, v i represents the interpolation result of the interpolation point, v j represents the observed value of the sample point, and n represents the number of sample points; The least squares method is used to approximate the function of the extracted data to obtain the fitting function of the inlet surface; the least squares method implementation process is as follows: Given data α0(x), α1(x), …, α n (x) Find the function that minimizes the sum of squared errors as shown in formula (5) from the function class n≤m, the function obtained is the least squares solution of this function class under this set of data:

7. The downscaling bridge wind-temperature coupling numerical simulation method considering real terrain according to claim 1 is characterized in that: The specific method of step (6) is: Use ANSYS software to mesh the geometric model; import the geometric model into ANSYS software, and further process the geometric model in the SpaceClaim module. First, check and repair the surface quality of the model, then establish data coupling interaction between the fluid and solid contact surfaces, group the boundary surfaces, and then enter the Fluentmeshing module for meshing; the entire model uses tetrahedral unstructured grids; mesh the terrain boundary, bridge tower structure, and nearby fluid domains; the first layer of grid height on the terrain surface is 0.1m, and the boundary layer growth rate is 1.1; The CFD software Fluent was used to establish a small-scale model. The velocity inlet and temperature inlet were selected when setting the inlet boundary conditions. The initial information was set by writing a UDF program based on the fitted inlet surface wind speed and temperature functions. The fluid was considered to be fully developed at the outlet, and the north side of the calculation model was set as the pressure outlet boundary. The ground and bridge tower structure surfaces of the calculation model were set as no-slip wall boundaries. The top surface of the model and the boundary surfaces on the east and west sides were set as symmetric boundaries. Perform FSI definition of structural wall boundaries and space domains; The calculation selects the pressure-based steady-state solver suitable for low-speed incompressible flow, the turbulence model and turbulence solution select SST k-ω and SIMPLEC algorithm, the discrete format selects the second-order upwind format, and the standard wall function selects.

Images