A lateral boundary turbulence synthesis method for numerical simulation of wind fields in complex terrain

Abstract

The present invention provides a lateral boundary turbulence synthesis method for numerical simulation of complex terrain wind field, including: determining the microscale calculation domain boundary of the meso-microscale coupling mode, obtaining the lateral boundary surface coordinates of the microscale calculation domain; establishing a mesoscale model simulation of the complex terrain wind field, determining the longitude and latitude and geographic coordinates corresponding to the reference point of the meso-microscale coupling mode; converting the lateral boundary surface coordinates of the microscale calculation domain into the longitude and latitude and altitude corresponding to the mesoscale calculation domain according to the terrain projection mode and the reference point position; extracting the wind speed, temperature and turbulent kinetic energy time series data corresponding to the lateral boundary surface coordinates of the microscale calculation domain from the mesoscale numerical simulation results; synthesizing the lateral boundary turbulence time series pulsating wind speed of the microscale calculation domain for microscale numerical simulation. The present invention improves the simulation accuracy of unsteady flow in complex terrain wind field based on the meso-microscale coupling mode, and is more suitable for the refined evaluation of wind resources in complex terrain.

Description

Technical Field

[0001] The invention relates to the technical field of wind power generation, and in particular to a lateral boundary turbulence synthesis method for numerical simulation of wind fields in complex terrains. Background Art

[0002] Wind energy is a clean energy that occupies an important position in the global energy structure. In order to improve the efficiency of wind energy utilization and reduce costs, it is necessary to deeply understand the characteristics of wind resources in wind farms. Due to the high construction cost of deep-sea wind farms and the fact that the rich wind resources in flat terrain have been gradually developed, the development of wind farms in complex mountains is a better choice for wind farm construction. Complex terrain wind farms are affected by terrain and atmospheric conditions, and there are phenomena such as high turbulence, separation and atmospheric circulation. It is a great challenge to accurately evaluate wind farm resources. The microscale numerical simulation method based on computational fluid dynamics (CFD) technology can accurately reflect the flow details of wind farms, including the wake field generated by wind turbine disturbances, the wind acceleration effect caused by terrain, and the local flow separation and high turbulence of the mountain, but it cannot describe the macroscopic state of wind farms driven by the atmosphere; the mesoscale simulation method based on the mesoscale meteorological research and forecasting (WRF) model can better reflect the impact of the local atmosphere on wind farms, but due to the large grid scale, the impact of terrain and wind turbines on wind farms cannot be accurately modeled. In order to take into account the simulation of both the macroscopic atmospheric state and the microscopic flow details, it is necessary to establish a multi-scale wind resource assessment method from the mesoscale (hundreds of meters to hundreds of kilometers) to the microscale (several meters to thousands of meters), so as to finely assess the wind resources in complex terrain and provide wind field information input for wind turbine layout and wind farm operation control.

[0003] When the unsteady simulation of complex terrain wind field is performed based on the mesoscale-microscale coupling mode, the wind speed fluctuation in the mesoscale simulation result is very small due to the insufficient turbulence generation of the mesoscale numerical simulation method, which cannot be directly used for microscale boundary driving. It is necessary to consider the turbulence generation method in the microscale numerical simulation to improve the accuracy of the mesoscale-microscale coupling method. At present, there are two main types of turbulence generation methods in microscale numerical simulation. The first type is to generate turbulence through the near-wall perturbation method. The advantage of this method is that the generated turbulent coherent structure is very close to the real atmospheric boundary layer flow, but this method is time-consuming and computationally expensive. In addition, this method needs to construct a periodic boundary during the turbulence generation process. When simulating complex terrain wind field, it is affected by the terrain undulation, and it is difficult to construct a periodic boundary. The second type is to generate turbulence through artificial synthesis method. The advantages of this method are fast turbulence generation rate and strong controllability of turbulence characteristics. However, this method is heavily dependent on the atmospheric boundary layer wind speed profile model, turbulence intensity profile model, turbulence integral scale model and pulsation power spectrum model when generating turbulent wind, and there is a lack of accurate analytical solutions of the above engineering models in complex terrain wind field. Therefore, when performing unsteady simulation of complex terrain wind farms based on the meso-microscale coupling model, the above two methods are limited and are not suitable for turbulence generation in the microscale computational domain. It is necessary to develop a turbulence synthesis method in the microscale computational domain that can truly reproduce the atmospheric boundary layer turbulence while taking into account the turbulence generation efficiency and numerical simulation accuracy. Summary of the invention

[0004] The purpose of the present invention is to provide a lateral boundary turbulence synthesis method for numerical simulation of complex terrain wind fields, which synthesizes the lateral boundary turbulent wind in the microscale computational domain based on the wind speed and turbulent kinetic energy time series data obtained by mesoscale numerical simulation, so that the inlet and outlet boundary turbulent wind is closer to the real physical state of the atmospheric boundary layer; overcomes the limitation of generating turbulence by inlet boundary conditions of simple wind speed profile or single wind direction; has high turbulence generation efficiency and is not affected by the terrain undulation at the boundary; is more suitable for the refined assessment of wind resources in complex terrain, and improves the simulation accuracy of unsteady flow in complex terrain wind fields based on meso-microscale coupling mode.

[0005] In order to achieve the above technical objectives, the technical solution adopted by the present invention is:

[0006] A lateral boundary turbulence synthesis method for numerical simulation of complex terrain wind field simulation, the lateral boundary turbulence synthesis method comprising the following steps:

[0007] Step 1: According to the actual terrain altitude difference and atmospheric boundary layer thickness of the wind farm area under study, the microscale calculation domain boundary of the meso-microscale coupling model is determined, and the lateral boundary surface coordinates of the microscale calculation domain are obtained;

[0008] Step 2: Obtain mesoscale static terrain data and meteorological data, determine the longitude and latitude and geographic coordinates corresponding to the reference point of the meso-microscale coupling model, simulate the complex terrain wind field based on the mesoscale WRF model, and obtain the mesoscale predicted wind speed field;

[0009] Step 3: According to the terrain projection method and the reference point position, the lateral boundary surface coordinates of the microscale calculation domain are converted into the longitude and latitude and altitude corresponding to the mesoscale predicted wind speed field, and the nested domain data of the mesoscale calculation domain including the wind farm is regularly saved to obtain the mesoscale numerical simulation results;

[0010] Step 4, extracting the wind speed, temperature and turbulent kinetic energy time series data corresponding to the lateral boundary surface coordinates of the microscale calculation domain from the mesoscale numerical simulation results;

[0011] Step 5, using the wind speed, temperature and turbulent kinetic energy time series data obtained in step 4, synthesize the turbulent time series pulsating wind speed at the lateral boundary of the microscale calculation domain for microscale numerical simulation.

[0012] Step 1 further comprises:

[0013] According to the actual terrain altitude difference and atmospheric boundary layer thickness in the wind farm area under study, the scope of the microscale calculation domain of the meso-microscale coupling model is determined;

[0014] Obtain high-precision terrain corresponding to the micro-scale computational domain, generate a background grid, and then encrypt the local grid and near-ground grid in the micro-scale computational domain to generate a micro-scale computational domain grid;

[0015] The grid node coordinates at the lateral boundaries of the microscale computational domain are extracted and converted into grid surface center coordinates.

[0016] Step 2 further includes:

[0017] Obtain mesoscale static terrain data and meteorological data, and determine the mesoscale calculation domain range and parameter configuration;

[0018] The center points of the mesoscale computational domain and the microscale computational domain are determined as the meso-microscale coupling reference points, and the longitude, latitude and terrain coordinates of the meso-microscale coupling reference points are obtained.

[0019] Step 3 further includes:

[0020] The lateral boundary coordinates of the microscale computational domain are moved so that the reference point of the microscale computational domain is located at (0,0). The microscale terrain is assumed to be in the UTM projection coordinate system. The geographic coordinates (x, y, z) of the lateral boundary surface of the microscale computational domain are converted to the longitude, latitude and altitude corresponding to the mesoscale computational domain according to the reference point position and terrain projection method.

[0021] Furthermore, in step 4, the spatial inverse distance interpolation method is used to extract the time series data of wind speed, temperature and turbulent kinetic energy on the lateral boundary surface of the microscale calculation domain from the mesoscale numerical results; the horizontal flow direction (from west to east), horizontal span direction (from south to north) and vertical direction (from bottom to top) of the wind speed can be respectively expressed as u , v , w to characterize; temperature and turbulent kinetic energy are respectively T and TKE Indicates; where, for the horizontal flow direction wind speed component u , construct the following interpolation function:

[0022] ;

[0023] ;

[0024] ;

[0025] ;

[0026] ;

[0027] ;

[0028] In the formula, f u ( i , j , k ), f u ( i , j , k+1 ), f u ( i+1 , j , k ), f u ( i+1 , j , k+1 ), f u ( i , j+1 , k ), f u ( i , j +1 , k+1 ), f u ( i+1 , j+1 , k ),f u ( i+1 , j+1 , k+1 ) is the wind speed component in the horizontal direction at each grid point u data; i , j and k are the horizontal, vertical and vertical grid numbers of the mesoscale computational domain; lon, lat and zcol are the longitude, latitude and altitude of the mesoscale computational domain grid; of_lat, of_lon and of_lz are the longitude, latitude and altitude of the microscale computational domain boundary surface coordinates; f u,interp is the wind speed component in the horizontal direction u The interpolation function of f u,00 , f u,10 , f u,01 , f u,11 and r u,m is an intermediate variable;

[0029] Similarly, for the horizontal wind speed component v and the vertical wind speed component w ,temperature T and turbulent kinetic energy TKE Construct their interpolation functions f v,interp , f w,interp , f T,interp and f TKE,interp , the interpolation function expression is the same as the wind speed variable u .

[0030] Step 5 further includes:

[0031] Generate a series of random numbers from -1 to 1;

[0032] Generate turbulent wind time series data based on wind speed and turbulent kinetic energy time series data:

[0033] ;

[0034] ;

[0035] ;

[0036] Where: k rA random number between -1 and 1; k 0 is the extracted turbulent kinetic energy time series data; U x,0 , U y,0 and U z,0 is the extracted wind speed time series data; U x , U y and U z is the synthetic microscale lateral boundary surface turbulent wind speed;

[0037] Import the synthesized turbulent wind and temperature data into the microscale computational domain boundary; specifically:

[0038] Using the time interpolation method, the time step between any two moments is assigned using the linear interpolation method:

[0039] ;

[0040] ;

[0041] Where: s is the time interpolation factor; t c Calculate the time for the current step; t s is the sampling start time; t e Sampling end time; Φ ref is the current time step variable; Φ s The variable value corresponding to the sampling start time; Φ e The variable value corresponding to the sampling end time;

[0042] The triangulated network / linear interpolation method is used to project the mesoscale space points onto the boundary plane. By constructing a two-dimensional triangular profile, the center of the discrete grid within the corresponding triangle on the boundary surface is found, and the variable distribution on the grid is calculated based on the weight relationship between the center of the surface and the three vertices:

[0043] ;

[0044] Where: ω A ,ω B and ω C are the weight values ​​of the three vertices respectively; Value A 、Value B and ValueC They are the variable values ​​corresponding to the three vertices; Value P is the variable value corresponding to the center of the face.

[0045] Furthermore, in step 5, the microscale lateral boundary conditions adopt the inlet and outlet boundary conditions that vary with time: when the flow is determined to be an outflow, the boundary is set to zero normal gradient; when the flow is determined to be an inflow, the boundary is set to the mesoscale data that varies with time.

[0046] Compared with the prior art, the present invention has the following beneficial effects:

[0047] First, the lateral boundary turbulence synthesis method for numerical simulation of complex terrain wind fields of the present invention synthesizes turbulence based on the results of mesoscale numerical simulation, and takes into account the atmospheric boundary layer turbulence in the microscale lateral boundary conditions, so that the inlet boundary turbulent wind is closer to the real atmospheric boundary layer physical state; it better overcomes the limitations of generating turbulence by inlet boundary conditions of simple wind speed profiles or single wind direction, is more suitable for the refined assessment of wind resources in complex terrain, and improves the simulation accuracy of unsteady flow in complex terrain wind fields based on the meso-microscale coupling model.

[0048] Second, the lateral boundary turbulence synthesis method for numerical simulation of complex terrain wind farms of the present invention can realize the refined evaluation of wind resources in complex terrain wind farms, increase the power generation of wind turbines and reduce economic losses.

[0049] Third, the lateral boundary turbulence synthesis method for numerical simulation of complex terrain wind field simulation of the present invention is simple to operate, easy to encode, simple to generate, and saves a lot of time cost. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 It is the topography of a certain wind farm;

[0051] Figure 2 It is a high-precision terrain of a wind farm in the UTM projection coordinate system (the terrain data comes from SRTM 30);

[0052] Figure 3 It is the micro-scale computational domain grid of a certain wind farm;

[0053] Figure 4 It is a schematic diagram of the spatial interpolation method of the lateral boundary of the microscale computational domain;

[0054] Figure 5a This is the instantaneous wind speed cloud map of the microscale computational domain during the mesoscale-microscale coupling simulation of a certain wind farm (based on the lateral boundary turbulence synthesis method)

[0055] Figure 5bThis is the instantaneous wind speed cloud map of the microscale computational domain during the microscale coupled simulation of a certain wind farm (no synthetic turbulence on the lateral boundary)

[0056] Figure 6a It is a comparison of radar observation data of a wind farm and the results of meso-microscale coupled simulation (average wind speed profile);

[0057] Figure 6b It is a comparison of radar observation data of a certain wind farm and the results of meso-microscale coupled simulation (turbulence intensity profile);

[0058] Figure 7 It is a flow chart of the lateral boundary turbulence synthesis method for numerical simulation of wind fields in complex terrains of the present invention. DETAILED DESCRIPTION

[0059] The embodiments of the present invention are further described in detail below with reference to the accompanying drawings.

[0060] See also Figure 7 The embodiment of the present invention discloses a lateral boundary turbulence synthesis method for numerical simulation of a complex terrain wind field, comprising:

[0061] 1. Obtain the lateral boundary surface coordinates of the microscale computational domain

[0062] According to the actual terrain altitude difference and atmospheric boundary layer thickness in the wind farm area under study, the microscale calculation domain range of the meso-microscale coupling model is determined. The microscale calculation domain range can be 6H×6H×H, where H is the atmospheric boundary layer thickness, which is determined by the atmospheric stability, 6H corresponds to the horizontal direction, and H corresponds to the vertical direction.

[0063] Obtain high-precision terrain corresponding to the micro-scale computational domain from SRTM 30 data or other high-precision elevation data (the terrain range is slightly larger than the wind farm area under study), and then save it in stl format. High-precision terrain can also come from field survey data.

[0064] After obtaining the high-precision terrain, the blockMesh tool of OpenFOAM is first used to generate the background mesh, and then the snappyHexMesh tool is used to complete the local mesh and near-ground mesh encryption process in the micro-scale calculation domain;

[0065] The coordinates of the microscale computational domain boundary are extracted using the OpenFOAM data monitoring function. The microscale numerical simulation parameters are simply configured and initialized, and the example is run in advance. After obtaining the coordinates of the lateral boundary grid nodes of the microscale computational domain, they are converted into the coordinates of the center of the grid surface.

[0066] 2. Determine the longitude and latitude and coordinates of the meso-microscale coupling reference point

[0067] Obtain mesoscale static terrain data and meteorological data, determine the mesoscale calculation domain range and parameter configuration scheme, and establish a complex terrain wind field simulation method based on the mesoscale WRF model. The climate input data for mesoscale numerical simulation uses the fifth generation data (ERA5) of the European Weather Forecast Center (ECMWF), and the static terrain data uses the data of the United States Geological Survey (USGS).

[0068] Determine the mesoscale-microscale coupling reference point and obtain the latitude, longitude and topographic coordinates of the reference point. The reference point is located at the center of the mesoscale and microscale computational domains.

[0069] 3. Convert the lateral boundary coordinates of the microscale computational domain to the latitude, longitude and altitude corresponding to the mesoscale computational domain

[0070] According to the reference point position and terrain projection method, the geographic coordinates (x, y, z) of the lateral boundary surface of the microscale computational domain are converted to the longitude, latitude and altitude corresponding to the mesoscale computational domain. The lateral boundary coordinates of the microscale computational domain are moved so that the horizontal and vertical coordinates of the microscale computational domain reference point are located at (0, 0). The microscale terrain is agreed to be in the UTM projection coordinate system.

[0071] 4. Extract the time series data of wind speed, temperature and turbulent kinetic energy on the lateral boundary surface of the microscale calculation domain from the mesoscale calculation results

[0072] During the mesoscale numerical simulation, the nested domain data of the mesoscale computational domain including the wind farm is saved every 10 minutes. A spatial inverse distance interpolation method is established to extract the time series data of wind speed, temperature and turbulent kinetic energy on the lateral boundary surface of the microscale computational domain from the mesoscale numerical results.

[0073] For the horizontal wind speed component u , construct the following interpolation function

[0074] ;

[0075] ;

[0076] ;

[0077] ;

[0078] ;

[0079] ;

[0080] In the formula, f u ( i , j , k ), fu ( i , j , k+1 ), f u ( i+1 , j , k ), f u ( i+1 , j , k+1 ), f u ( i , j+1 , k ), f u ( i , j +1 , k+1 ), f u ( i+1 , j+1 , k ), f u ( i+1 , j+1 , k+1 ) is the wind speed component in the horizontal direction at each grid point u data; i , j and k are the horizontal, vertical and vertical grid numbers of the mesoscale computational domain; lon, lat and zcol are the longitude, latitude and altitude of the mesoscale computational domain grid; of_lat, of_lon and of_lz are the longitude, latitude and altitude of the microscale computational domain boundary surface coordinates; f u,interp is the wind speed component in the horizontal direction u The interpolation function of f u,00 , f u,10 , f u,01 , f u,11 and r u,m is an intermediate variable.

[0081] For the horizontal wind speed component v and the vertical wind speed component w ,temperature T and turbulent kinetic energy TKE Their interpolation functions can also be constructed similarly fv,interp , f w,interp , f T,interp and f TKE,interp , the interpolation function expression is the same as the wind speed variable u .

[0082] 5. Synthesize turbulent wind at the lateral boundary of the microscale computational domain

[0083] Generate a series of random numbers from -1 to 1, and generate turbulent wind at the boundary of the microscale computational domain based on the wind speed and turbulent kinetic energy time series data:

[0084] ;

[0085] ;

[0086] ;

[0087] Where: k r A random number between -1 and 1; k 0 is the turbulent kinetic energy time series data extracted in step 4; U x,0 , U y,0 and U z,0 is the wind speed time series data (10-minute data) extracted in step 4; U x , U y and U z is the synthetic microscale lateral boundary surface turbulent wind speed (second-level data);

[0088] Since the wind speed and turbulent kinetic energy data extracted in step 4 have a large time resolution (minute level), it is necessary to first generate short-term turbulent wind (second-level data), and then establish a suitable time and space interpolation method to import the synthesized turbulent wind and temperature data into the microscale calculation domain boundary. The lateral boundary conditions of the microscale calculation domain adopt the inlet and outlet boundary conditions that vary with time. When the flow is judged as outflow (i.e., outlet boundary), the boundary is set to zero normal gradient; when the flow is judged as inflow (i.e., inlet boundary), the boundary is set to mesoscale data that varies with time.

[0089] Time interpolation method:

[0090] For the time step between any two moments, linear interpolation is used to assign values:

[0091] ;

[0092] ;

[0093] Where: s is the time interpolation factor; t c Calculate the time for the current step; t s is the sampling start time; t e Sampling end time. Φ ref is the current time step variable (which can be wind speed, temperature, turbulent kinetic energy, etc.); Φ s The variable value corresponding to the sampling start time; Φ e The variable value corresponding to the sampling end time;

[0094] Spatial interpolation method:

[0095] The triangulated network / linear interpolation method is used to project the mesoscale space points onto the boundary plane. By constructing a two-dimensional triangular profile, the center P of the discrete grid within the corresponding triangle on the boundary surface is found, and the variable distribution on the grid is calculated based on the weight relationship between the center of the surface and the three vertices:

[0096] ;

[0097] Where: ω A ,ω B and ω C are the weight values ​​of the three vertices respectively; Value A 、Value B and Value C They are the variable values ​​corresponding to the three vertices (which can be wind speed, temperature, turbulent kinetic energy, etc.) ;Value P is the variable value corresponding to the center of the face.

[0098] 6. Method Validation

[0099] Taking an actual wind farm in Yunnan, China as an example, the turbulence generation technology of the microscale computational domain lateral boundary in the meso-micro coupled numerical simulation is verified. Figure 1) is located in Yunnan Province, belonging to the hilly area. The wind farm terrain consists of multiple isolated peaks, with the highest elevation of 1794m and the lowest elevation of 1215m. The surface vegetation is mainly farmland, grassland and shrubs, and the average surface roughness length is 0.034m. The wind radar in the wind farm is located near the local mountain top (103.23080601°E, 23.86983333°N), and the radar location altitude is 1732m.

[0100] Step 1: Obtain the lateral boundary surface coordinates of the microscale computational domain

[0101] According to the actual terrain altitude difference and atmospheric boundary layer thickness in the wind farm area, the microscale calculation domain of the meso-microscale coupling model is determined. The microscale calculation domain height is 2 km, and the horizontal size is set to 5 km×5 km.

[0102] The high-precision terrain corresponding to the microscale computational domain is obtained from the SRTM 30 data (such as Figure 2 , the terrain range is slightly larger than the wind farm area under study), and then save it in stl format. To ensure the consistency of the mesoscale simulation output data and the microscale simulation boundary input data, Figure 2 The terrain shown is the terrain after the fusion of mesoscale and microscale terrain. The blue box in the figure indicates the boundary position of the microscale calculation domain.

[0103] After obtaining high-precision terrain, generate a microscale computational domain grid (such as Figure 3 ). First, the blockMesh tool of OpenFOAM is used to generate the background mesh, and then the snappyHexMesh tool is used to complete the local mesh and near-ground mesh encryption process in the microscale calculation domain.

[0104] Configure the microscale numerical simulation parameters, initialize, and run the example in advance. Use the OpenFOAM data monitoring function to monitor the side boundary parameters of the computational domain to obtain the grid node coordinates and convert them into the center coordinates of the grid surface. In this step, since the mesoscale parameters are unknown, the example assumes that the wind speed and temperature at the inlet boundary are fixed when running.

[0105] Step 2: Determine the longitude and latitude and coordinates of the meso-microscale coupling reference point

[0106] Obtain mesoscale static terrain data and meteorological data, determine the mesoscale calculation domain range and parameter configuration scheme, and establish a complex terrain wind field simulation method based on the mesoscale WRF model.

[0107] The climate input data for mesoscale numerical simulation is the fifth generation data (ERA5) of the European Centre for Medium-Range Weather Forecasts (ECMWF), with a spatial resolution of 0.25° and a temporal resolution of 1 hour. The static data of terrain type and terrain height are taken from the data of the United States Geological Survey (USGS), and its spatial resolution is 30". Four layers of nesting are adopted, and the projection mode is set to Lambert. The spatial horizontal resolutions are 27 km (d01), 9 km (d02), 3 km (d03) and 1 km (d04), respectively. The vertical height is 56 layers, which are encrypted near the ground, including 10 layers below 100m, 10 layers in the range of 100-500m, 5 layers in the range of 500-1000m, and 31 layers above 1000m. The simulation time period is 2023.09.16 0:00-2023.09.19 0:00 (UTC time), the center point of the simulation area is (23.8697°N, 103.2308°W), the WRF simulation time resolution is 30 s, 10 s, 3.33 s and 1.11 s, respectively, and the simulation result output interval is 10min.

[0108] The physical parameterization schemes used in the simulation process include: microphysical processes (WSM 6), longwave and shortwave radiation (RRTMG), near-surface layer (Monin-Obukhov), land surface processes (Unified Noah), planetary boundary layer (MYJ) and cumulus convection parameters (Kain-Fritsch).

[0109] Determine the mesoscale-microscale coupling reference point and obtain the latitude, longitude and terrain coordinates of the reference point. Select the center point of the mesoscale computational domain as the reference point, and the geographic coordinates of the point are (319853.46, 2640944.079) (in the UTM projection coordinate system).

[0110] Step 3: Convert the lateral boundary surface coordinates of the microscale computational domain into the latitude, longitude and altitude corresponding to the mesoscale computational domain

[0111] According to the reference point position and terrain projection method, the geographic coordinates (x, y, z) of the lateral boundary surface of the microscale computational domain are converted to the longitude, latitude and altitude corresponding to the mesoscale computational domain. The lateral boundary coordinates are moved so that the horizontal and vertical coordinates of the microscale computational domain reference point are located at (0, 0). The Python script is used for coordinate conversion. The reference point position of the wind farm is (23.8697°N, 103.2308°W) / (319853.46, 2640944.079), and the terrain projection method of the microscale computational domain is UTM.

[0112] Step 4: Extract the time series data of wind speed, temperature and turbulent kinetic energy on the lateral boundary surface of the microscale calculation domain from the mesoscale calculation results

[0113] During the mesoscale numerical simulation, the data of the fourth nested domain of the mesoscale computational domain are saved every 10 minutes. The time series data of wind speed, temperature and turbulent kinetic energy on the lateral boundary surface of the microscale computational domain are extracted from the mesoscale numerical results based on the spatial inverse distance interpolation method. The horizontal flow direction (from west to east), horizontal span direction (from south to north) and vertical direction (from bottom to top) of the wind speed can be respectively expressed as u , v , w to characterize; temperature and turbulent kinetic energy are respectively T and TKE express.

[0114] For the horizontal wind speed component u , construct the following interpolation function:

[0115] ;

[0116] ;

[0117] ;

[0118] ;

[0119] ;

[0120] ;

[0121] In the formula, f u ( i , j , k ), f u ( i , j , k+1 ), f u ( i+1 , j , k ), f u ( i+1 , j , k+1 ), f u ( i , j+1 , k ), f u ( i ,j +1 , k+1 ), f u ( i+1 , j+1 , k ), f u ( i+1 , j+1 , k+1 ) is the wind speed component in the horizontal direction at each grid point u data; i , j and k are the horizontal, vertical and vertical grid numbers of the mesoscale computational domain; lon, lat and zcol are the longitude, latitude and altitude of the mesoscale computational domain grid; of_lat, of_lon and of_lz are the longitude, latitude and altitude of the microscale computational domain boundary surface coordinates; f u,interp is the wind speed component in the horizontal direction u The interpolation function of f u,00 , f u,10 , f u,01 , f u,11 and r u,m is an intermediate variable.

[0122] For the horizontal wind speed component v and the vertical wind speed component w ,temperature T and turbulent kinetic energy TKE Their interpolation functions can also be constructed similarly f v,interp , f w,interp , f T,interp and f TKE,interp , the interpolation function expression is the same as the wind speed variable u .

[0123] Step 5: Synthesize turbulent wind at the lateral boundary of the microscale computational domain

[0124] First, a series of random number seeds from -1 to 1 are generated, and then the turbulent wind at the boundary of the microscale computational domain is generated according to the wind speed and turbulent kinetic energy time series data:

[0125] ;

[0126] ;

[0127] ;

[0128] Where: k r A random number between -1 and 1; k 0 is the turbulent kinetic energy time series data extracted in step 4; U x,0 , U y,0 and U z,0 is the wind speed time series data (10-minute data) extracted in step 4; U x , U y and U z is the synthetic microscale lateral boundary surface turbulent wind speed (second-level data, 5s per data, within a 10min time range, the wind speed mean remains unchanged, and the wind speed fluctuation amplitude is determined by the turbulent kinetic energy and random numbers obtained from the WRF results);

[0129] Since the wind speed and turbulent kinetic energy data extracted in step 4 have a large time resolution (10min time resolution), it is necessary to generate a shorter time turbulent wind (5s time resolution) first, and then establish a suitable time (for microscale numerical simulation, the time resolution is 0.01-1s) and spatial interpolation method to import the synthesized turbulent wind and temperature data into the microscale calculation domain boundary. The lateral boundary conditions of the microscale calculation domain adopt the inlet and outlet boundary conditions that vary with time. When the flow is judged as outflow (i.e., the outlet boundary), the boundary is set to zero normal gradient; when the flow is judged as inflow (i.e., the inlet boundary), the boundary is set to the time-varying mesoscale data.

[0130] Time interpolation method:

[0131] For the time step between any two moments (5s in this case), linear interpolation is used to assign values:

[0132] ;

[0133] ;

[0134] Where: s is the time interpolation factor; t c Calculate the time for the current step; t s is the sampling start time; t e Sampling end time. Φref is the current time step variable (which can be wind speed, temperature, turbulent kinetic energy, etc.); Φ s is the variable value corresponding to the sampling start time; Φ e is the variable value corresponding to the sampling end time;

[0135] Spatial interpolation method:

[0136] Using triangulation / linear interpolation method (such as Figure 4 As shown in the figure, the mesoscale space point is projected onto the boundary plane, and the face center of the discrete grid within the corresponding triangle on the boundary surface is found by constructing a two-dimensional triangular profile. The variable distribution on the grid is calculated based on the weight relationship between the face center and the three vertices:

[0137] ;

[0138] Where: ω A ,ω B and ω C They are the weight values ​​of the three vertices of the triangle; Value A 、Value B and Value C They are the variable values ​​corresponding to the three vertices of the triangle (which can be wind speed, temperature, turbulent kinetic energy, etc.) ;Value P is the variable value corresponding to the face center P.

[0139] In microscale numerical simulation, atmospheric thermal stability and the Coriolis force are considered in the momentum equation; the large eddy simulation (LES) is used as the turbulence model, and the sub-grid model is oneEqEddyABL, which is based on the oneEqEddy model and takes into account some corrections for the atmospheric boundary layer solution; the wall stress model is used for the wall.

[0140] PISO-SIMPLE is used to "implicitly" solve the momentum and pressure equations. The prediction-correction method uses the Rhie-Chow interpolation method, and the temperature equation is added to the correction loop to solve the temperature coupling problem. The use of a flux interpolation method similar to Rhie-Chow can avoid pressure-velocity decoupling. Time discretization uses the time domain backward difference method, and the gradient term, convection term, and divergence term are discretized using the Gaussian linear interpolation format. The Laplace term is discretized using the Gaussian linear interpolation correction format.

[0141] The method of the present invention and the method of lateral boundary non-turbulence synthesis were used to perform numerical simulation analysis on this case. After using the method of the present invention, the turbulent structure in the microscale calculation domain is obvious (such as Figure 5a ); The traditional method, that is, the lateral boundary turbulence synthesis method is not considered in the meso-microscale coupled simulation, and the mesoscale wind speed is directly used for driving. At this time, due to the insufficient turbulence generation of the mesoscale numerical simulation, and the inability of the microscale numerical simulation method to develop turbulence in a short time, the turbulence structure in the microscale numerical simulation results is not obvious (such as Figure 5b ), which is contrary to the characteristics of the real atmospheric boundary layer.

[0142] In order to quantitatively analyze the accuracy of the method of the present invention, the average wind speed profile and turbulence intensity profile at the radar position are analyzed. Figure 6a and Figure 6b When the lateral boundary turbulence synthesis is not considered, the average absolute error of wind speed is 1.38m / s, and the average absolute error of turbulence intensity is 0.0822; after the lateral boundary turbulence synthesis method is adopted, the average absolute error of wind speed is 0.45m / s, and the average absolute error of turbulence intensity is 0.0157. The accuracy of the numerical simulation results of the mesoscale and microscale coupling model has been improved.

[0143] comprehensive Figure 5a , Figure 5b , Figure 6a and Figure 6b It can be seen that the wind speed and turbulence intensity profiles obtained by the meso-microscale coupled numerical simulation of the lateral boundary turbulence synthesis method based on the numerical simulation of complex terrain wind fields are very consistent with the radar observation data, and can better simulate the special wind profiles and turbulence profiles of complex terrain; compared with the traditional method that does not consider the boundary turbulence synthesis, the atmospheric boundary layer turbulence obtained by the numerical simulation results of this method is closer to the real turbulent wind.

[0144] In summary, based on the need for lateral boundary turbulent wind synthesis during numerical simulation of complex terrain, the present invention proposes a lateral boundary turbulence synthesis method for numerical simulation of wind fields in complex terrain. Turbulence is synthesized based on the results of mesoscale numerical simulation, and the atmospheric boundary layer turbulence is considered in the microscale lateral boundary conditions, so that the inlet boundary turbulent wind is closer to the real physical state of the atmospheric boundary layer; it better overcomes the limitations of generating turbulence by inlet boundary conditions of simple wind speed profiles or single wind direction, and is more suitable for the refined assessment of wind resources in complex terrain, and improves the simulation accuracy of unsteady flow in wind fields in complex terrain based on the meso-microscale coupling model, thereby realizing the refined assessment of wind resources in wind farms in complex terrain, improving the power generation of wind turbines and reducing economic losses. The present invention is also simple to operate, easy to encode, simple to generate, and saves a lot of time and cost.

[0145] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes. The schemes in the embodiments of the present application may be implemented in various computer languages, for example, object-oriented programming language Java and literal scripting language JavaScript, etc.

[0146] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0147] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0148] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0149] Although the preferred embodiments of the present application have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present application.

[0150] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.

Claims

1. A lateral boundary turbulence synthesis method for numerical simulation of wind field in complex terrain, characterized in that: The lateral boundary turbulence synthesis method comprises the following steps: Step 1: According to the actual terrain altitude difference and atmospheric boundary layer thickness of the wind farm area under study, the microscale calculation domain boundary of the meso-microscale coupling model is determined, and the lateral boundary surface coordinates of the microscale calculation domain are obtained; Step 2: Obtain mesoscale static terrain data and meteorological data, determine the longitude and latitude and geographic coordinates corresponding to the reference point of the meso-microscale coupling model, simulate the complex terrain wind field based on the mesoscale WRF model, and obtain the mesoscale predicted wind speed field; Step 3: According to the terrain projection method and the reference point position, the lateral boundary surface coordinates of the microscale calculation domain are converted into the longitude and latitude and altitude corresponding to the mesoscale predicted wind speed field, and the nested domain data of the mesoscale calculation domain including the wind farm is regularly saved to obtain the mesoscale numerical simulation results; Step 4, extracting the wind speed, temperature and turbulent kinetic energy time series data corresponding to the lateral boundary surface coordinates of the microscale calculation domain from the mesoscale numerical simulation results; Step 5, using the wind speed, temperature and turbulent kinetic energy time series data obtained in step 4, synthesize the turbulent time series pulsating wind speed at the lateral boundary of the microscale calculation domain for microscale numerical simulation; In step 4, the spatial inverse distance interpolation method is used to extract the time series data of wind speed, temperature and turbulent kinetic energy on the lateral boundary surface of the microscale calculation domain from the mesoscale numerical results; among them, for the wind speed component u in the horizontal flow direction, the following interpolation function is constructed: r u,m =1 / ((lon(i+1,j)-lon(i,j))*(lat(i,j+1)-lat(i,j))); f u,interp =f u，00 *r u，m *((lon(i+1,j)-of_lon)*(lat(i,j+1)-of_lat))+f u,10 *r u,m *(of_lon-lon(i,j))*(lat(i,j+1)-of_lat)+f u，01 *r u,m *(lon(i+1,j)-of_lon)*(of_lat-lat(i,j))+f u,11 *r u，m *(of_lon-lon(i,j))*(of_lat-lat(i,j)); In the formula, f u (i,j,k),f u (i,j,k+1),f u (i+1,j,k),f u (i+1,j,k+1),f u (i,j+1,k),f u (i,j+1,k+1),f u (i+1,j+1,k),f u (i+1, j+1, k+1) is the data of the wind speed component u in the horizontal direction of each grid point; i, j and k are the horizontal, longitudinal and vertical grid numbers of the mesoscale calculation domain respectively; lon, lat and zcol are the longitude, latitude and altitude corresponding to the grid of the mesoscale calculation domain respectively; of_lat, of_lon, of_lz are the longitude, latitude and altitude corresponding to the coordinates of the boundary surface of the microscale calculation domain respectively; f u,interp is the interpolation function of the wind speed component u in the horizontal flow direction; f u,00 、f u,10 、f u,01 、f u,11 and r u,m is an intermediate variable; For the horizontal wind speed component v and the vertical wind speed component w, temperature T and turbulent kinetic energy TKE, their interpolation functions f are constructed. v,interp 、f w,interp 、f T,interp and f TKE,interp ; Step 5 further includes: Generate a series of random numbers from -1 to 1; Generate turbulent wind time series data based on wind speed and turbulent kinetic energy time series data: Where: k r is a random number between -1 and 1; k0 is the extracted turbulent kinetic energy time series data; U x,0 , U y,0 and U z,0 is the extracted wind speed time series data; U x , U y and U z is the synthetic microscale lateral boundary surface turbulent wind speed; Import the synthesized turbulent wind and temperature data into the microscale computational domain boundary; specifically: Using the time interpolation method, the time step between any two moments is assigned using the linear interpolation method: Φ ref =(1-s)Φ s +sΦ e ; Where: s is the time interpolation factor; t c Calculate the time for the current step; t s is the sampling start time; t e Sampling end time; Φ ref is the current time step variable; Φ s is the variable value corresponding to the sampling start time; Φ e is the variable value corresponding to the sampling end time; The triangulated network / linear interpolation method is used to project the mesoscale space points onto the boundary plane. By constructing a two-dimensional triangular profile, the center of the discrete grid on the boundary surface within the corresponding triangle range is found, and the variable distribution on the grid is calculated based on the weight relationship between the center of the surface and the three vertices A, B, and C of the triangle: Where: A ,ω B and ω C are the weight values ​​of the three vertices of the triangle; Value A 、Value B and Value C They are the variable values ​​corresponding to the three vertices of the triangle; Value P is the variable value corresponding to the center of the face.

2. The lateral boundary turbulence synthesis method for numerical simulation of complex terrain wind field simulation according to claim 1 is characterized in that: Step 1 further comprises: According to the actual terrain altitude difference and atmospheric boundary layer thickness in the wind farm area under study, the scope of the microscale calculation domain of the meso-microscale coupling model is determined; Obtain high-precision terrain corresponding to the micro-scale computational domain, generate a background grid, and then encrypt the local grid and near-ground grid in the micro-scale computational domain to generate a micro-scale computational domain grid; The grid node coordinates at the lateral boundaries of the microscale computational domain are extracted and converted into grid surface center coordinates.

3. The lateral boundary turbulence synthesis method for numerical simulation of complex terrain wind field simulation according to claim 1 is characterized in that: Step 2 further includes: Obtain mesoscale static terrain data and meteorological data, and determine the mesoscale calculation domain range and parameter configuration; The center points of the mesoscale computational domain and the microscale computational domain are determined as the meso-microscale coupling reference points, and the longitude, latitude and terrain coordinates of the meso-microscale coupling reference points are obtained.

4. The lateral boundary turbulence synthesis method for numerical simulation of complex terrain wind field simulation according to claim 1 is characterized in that: Step 3 further includes: The lateral boundary coordinates of the microscale computational domain are moved so that the reference point of the microscale computational domain is located at (0,0). The microscale terrain is assumed to be in the UTM projection coordinate system. The geographic coordinates (x, y, z) of the lateral boundary surface of the microscale computational domain are converted to the longitude, latitude and altitude corresponding to the mesoscale computational domain according to the reference point position and terrain projection method.

5. The lateral boundary turbulence synthesis method for numerical simulation of complex terrain wind field simulation according to claim 1 is characterized in that: In step 5, the microscale lateral boundary conditions adopt the inlet and outlet boundary conditions that vary with time: when the flow is judged to be outflow, the boundary is set to zero normal gradient; when the flow is judged to be inflow, the boundary is set to the mesoscale data that varies with time.

Images