Regional wind field prediction method, device, equipment and storage medium

Abstract

The present application discloses a regional wind field prediction method, device, equipment and storage medium, which relates to the field of weather forecast technology, including: using a mesoscale weather model to simulate the terrain wind field of a target area based on a preset meteorological reanalysis data set and preset static high-resolution geographic data to obtain mesoscale wind field data and performing block interpolation processing on the mesoscale wind field data to obtain a target interpolation result, inputting the target interpolation result into a preset LBM microscale model to obtain a stable wind field prediction result and determine the corresponding current result acquisition time; using the mesoscale weather model to obtain the mesoscale simulation result corresponding to the target area at the current result acquisition time, and jumping to the step of obtaining mesoscale wind field data until the current result acquisition time is the target prediction stop time to obtain the target wind field prediction result of the target area. In this way, a high-precision wind field prediction result can be obtained to enhance the accuracy of weather forecasts under complex terrain.

Description

Technical Field

[0001] The present invention relates to the technical field of weather forecasting, and in particular to a regional wind field prediction method, device, equipment and storage medium. Background Art

[0002] Traditional weather forecast models, such as the Weather Research and Forecasting Model (WRF), can provide weather forecast information for large areas. However, in complex terrain conditions (such as urban and mountainous areas), forecast accuracy is often insufficient due to a lack of detailed simulation of local terrain features. In particular, in urban areas with densely populated buildings and in mountainous terrain, wind speed and direction vary greatly, making it difficult for traditional forecast models to capture these details. Furthermore, existing weather forecast models are typically mesoscale models, using large spatial grids (typically 3-10 km), which cannot provide high-resolution meteorological data. This is a significant shortcoming in areas requiring precise management, such as urban transportation planning, air quality monitoring, and disaster emergency response. To improve forecast accuracy, some studies have attempted to use high-resolution computational fluid dynamics models for global weather simulation, but this results in significant computational cost and resource waste, making it difficult to implement in practical applications. Therefore, the question of how to obtain high-precision wind field forecasts remains unanswered. Summary of the Invention

[0003] In view of this, the present invention aims to provide a regional wind field prediction method, apparatus, device, and storage medium that can obtain high-precision wind field prediction results to enhance the accuracy of weather forecasts in complex terrain. The specific scheme is as follows:

[0004] In a first aspect, the present application discloses a regional wind field prediction method, comprising:

[0005] Using a mesoscale weather model to simulate the terrain wind field of the target area based on a preset meteorological reanalysis dataset and preset static high-resolution geographic data to obtain a mesoscale simulation result corresponding to the initial time, and performing data processing on the mesoscale simulation result based on a preset data analysis and processing method to obtain mesoscale wind field data;

[0006] Performing block interpolation processing on the mesoscale wind field data to obtain a target interpolation result, and inputting the target interpolation result into a preset LBM microscale model to drive the preset LBM microscale model to perform wind field simulation on the target area until a stable wind field prediction result is obtained, and determining the corresponding current result acquisition time;

[0007] The mesoscale weather model is used to obtain the mesoscale simulation result corresponding to the target area at the current result acquisition moment, and the step of performing data processing on the mesoscale simulation result based on the preset data analysis and processing method to obtain mesoscale wind field data is jumped to, until the current result acquisition moment corresponding to the stable wind field prediction result is the target prediction stop moment to obtain the target wind field prediction result of the target area.

[0008] Optionally, the using of a mesoscale weather model to perform terrain wind field simulation on a target area based on a preset meteorological reanalysis dataset and preset static high-resolution geographic data to obtain a mesoscale simulation result corresponding to an initial moment, and performing data processing on the mesoscale simulation result based on a preset data analysis and processing method to obtain mesoscale wind field data, includes:

[0009] Obtain an ERA5 meteorological reanalysis dataset and preset static high-resolution geographic data for the target area, and use a mesoscale weather model to perform a three-layer nested mesoscale simulation on the ERA5 meteorological reanalysis dataset and the preset static high-resolution geographic data to obtain a mesoscale simulation result corresponding to an initial time point;

[0010] interpolating the target wind speed variable extracted from the mesoscale simulation result to a preset pressure altitude layer to obtain a wind field variable;

[0011] The wind field variables are converted into three-dimensional space coordinates to obtain target mesoscale wind speed variables, and the target mesoscale wind speed variables are stored in a CSV format to obtain mesoscale wind field data.

[0012] Optionally, performing block interpolation processing on the mesoscale wind field data to obtain a target interpolation result includes:

[0013] The mesoscale wind field data is divided into blocks based on a preset data block method to obtain a target number of first-level interpolation sub-regions, and the first-level interpolation sub-regions are interpolated using a preset multi-level parallel interpolation coupling algorithm to obtain interpolated sub-region results; the preset multi-level parallel interpolation coupling algorithm includes a Cressman interpolation method and a cubic spline interpolation method;

[0014] The sub-region results located in the LBM micro-scale calculation domain are screened out from the interpolated sub-region results to obtain a first-level interpolation result, and the first-level interpolation result is interpolated based on the preset data blocking method and the preset multi-level parallel interpolation coupling algorithm to obtain a target interpolation result.

[0015] Optionally, the block processing of the mesoscale wind field data based on a preset data block method to obtain a target number of first-level interpolation sub-regions includes:

[0016] A CSV data file corresponding to a target resolution is extracted from the mesoscale wind field data to obtain wind field data to be interpolated, and the wind field data to be interpolated is two-dimensionally divided based on the number of MPI processes corresponding to the wind field data to obtain the number of first-level interpolation sub-regions of the MPI processes.

[0017] Optionally, the interpolation processing of the first-level interpolation sub-region using a preset multi-level parallel interpolation coupling algorithm to obtain an interpolated sub-region result includes:

[0018] The first-level interpolation sub-region is horizontally interpolated using the Cressman interpolation method to obtain the sub-region result after the first-level interpolation processing, and the first-level interpolation sub-region result is vertically interpolated using the cubic spline interpolation method to obtain the sub-region result after interpolation.

[0019] Optionally, performing horizontal interpolation processing on the first-level interpolation sub-region by using the Cressman interpolation method to obtain a sub-region result after the first-level interpolation processing includes:

[0020] An interpolation influence radius and a smoothing factor in the Cressman interpolation method are determined, and interpolation calculations are performed on each preset pressure altitude layer of the first-level interpolation sub-region based on the interpolation influence radius and the smoothing factor to obtain a sub-region result after the first-level interpolation processing.

[0021] Optionally, the step of selecting a sub-region result located in the LBM micro-scale computational domain from the interpolated sub-region results to obtain a first-level interpolation result, and interpolating the first-level interpolation result based on the preset data blocking method and the preset multi-level parallel interpolation coupling algorithm to obtain a target interpolation result includes:

[0022] Screening each of the interpolated sub-regions to determine a sub-region within the LBM micro-scale calculation region, then merging the results of the sub-regions within the LBM micro-scale calculation region and storing them as a CSV data file to obtain a first-level interpolation result;

[0023] Performing two-dimensional division processing on the wind field data to be interpolated based on the number of MPI processes corresponding to the first-level interpolation result to obtain corresponding second-level interpolation sub-areas;

[0024] The Cressman interpolation method and the cubic spline interpolation method are used to perform horizontal interpolation processing and vertical interpolation processing on the second-level interpolation sub-region respectively to obtain the target interpolation result.

[0025] In a second aspect, the present application discloses a regional wind field prediction device, comprising:

[0026] A wind field data processing module is used to perform terrain wind field simulation on the target area using a mesoscale weather model based on a preset meteorological reanalysis dataset and preset static high-resolution geographic data to obtain a mesoscale simulation result corresponding to an initial time, and to perform data processing on the mesoscale simulation result based on a preset data analysis and processing method to obtain mesoscale wind field data;

[0027] A wind field prediction module is used to perform block interpolation processing on the mesoscale wind field data to obtain a target interpolation result, and input the target interpolation result into a preset LBM microscale model to drive the preset LBM microscale model to perform wind field simulation on the target area until a stable wind field prediction result is obtained, and determine the corresponding current result acquisition time;

[0028] A step jump module is used to use the mesoscale weather model to obtain the mesoscale simulation result corresponding to the target area at the current result acquisition moment, and jump to the step of performing data processing on the mesoscale simulation result based on the preset data analysis and processing method to obtain mesoscale wind field data, until the current result acquisition moment corresponding to the stable wind field prediction result is the target prediction stop moment to obtain the target wind field prediction result of the target area.

[0029] In a third aspect, the present application discloses an electronic device, comprising:

[0030] Memory, used to store computer programs;

[0031] A processor is used to execute the computer program to implement the aforementioned regional wind field prediction method.

[0032] In a fourth aspect, the present application discloses a computer-readable storage medium for storing a computer program, which implements the aforementioned regional wind field prediction method when executed by a processor.

[0033] It can be seen that in the present application, a mesoscale weather model is used to perform terrain wind field simulation on the target area based on a preset meteorological reanalysis data set and preset static high-resolution geographic data to obtain a mesoscale simulation result corresponding to the initial moment, and the mesoscale simulation result is processed based on a preset data analysis and processing method to obtain mesoscale wind field data; the mesoscale wind field data is block interpolated to obtain a target interpolation result, and the target interpolation result is input into a preset LBM microscale model to drive the preset LBM microscale model to perform wind field simulation on the target area until a stable wind field prediction result is obtained, and the corresponding current result acquisition time is determined; the mesoscale weather model is used to obtain the mesoscale simulation result corresponding to the target area at the current result acquisition time, and jump to the step of performing data processing on the mesoscale simulation result based on the preset data analysis and processing method to obtain mesoscale wind field data, until the current result acquisition time corresponding to the stable wind field prediction result is the target prediction stop time to obtain the target wind field prediction result of the target area. Specifically, the wind data output by the mesoscale weather model is extracted. Using a multi-level parallel interpolation coupling algorithm, the coarse-grid data extracted from the mesoscale weather model simulation results is converted into the fine-grid data required by the LBM model. This enables microscale, high-resolution simulation calculations, resulting in highly accurate wind field forecasts. This significantly reduces computational costs while maintaining high computational efficiency and high-precision meteorological simulations, thereby improving the accuracy of weather forecasts in complex terrain conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are merely embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0035] Figure 1 This is a flow chart of a regional wind field prediction method disclosed in this application;

[0036] Figure 2 This is a schematic diagram of a specific terrain wind field simulation method disclosed in this application;

[0037] Figure 3 This is a schematic diagram of a specific mesoscale wind field data disclosed in this application;

[0038] Figure 4 This is a specific wind field data block interpolation processing flow chart disclosed in this application;

[0039] Figure 5 This is a schematic diagram of a specific wind field data segmentation result disclosed in this application;

[0040] Figure 6 A schematic diagram of a specific Cressman interpolation influence domain disclosed in this application;

[0041] Figure 7 A schematic diagram of a vertical profile of wind speed disclosed in this application;

[0042] Figure 8 A specific first-level interpolation result data screening flow chart disclosed in this application;

[0043] Figure 9 This is a schematic diagram of a specific regional wind field prediction result disclosed in this application;

[0044] Figure 10 This is a schematic structural diagram of a regional wind field prediction device disclosed in this application;

[0045] Figure 11 This is a structural diagram of an electronic device disclosed in this application. DETAILED DESCRIPTION

[0046] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0047] In order to improve forecast accuracy, existing technologies attempt to use high-resolution computational fluid dynamics models for global weather simulations. However, this results in a huge waste of computational resources and is difficult to implement in practical applications. This application specifically introduces a regional wind field prediction method that can achieve high-precision meteorological simulations while saving computing resources.

[0048] See also Figure 1 As shown, the embodiment of the present application discloses a regional wind field prediction method, including:

[0049] Step S11: Using a mesoscale weather model, a terrain wind field simulation is performed on the target area based on a preset meteorological reanalysis data set and preset static high-resolution geographic data to obtain a mesoscale simulation result corresponding to the initial moment, and the mesoscale simulation result is processed based on a preset data analysis and processing method to obtain mesoscale wind field data.

[0050] In this embodiment, the mesoscale weather model is used to simulate the terrain wind field of the target area based on the preset meteorological reanalysis data set and the preset static high-resolution geographic data to obtain the mesoscale simulation result corresponding to the initial moment, and the mesoscale simulation result is processed based on the preset data analysis and processing method to obtain the mesoscale wind field data, including: obtaining the ERA5 meteorological reanalysis data set and the preset static high-resolution geographic data of the target area, and using the mesoscale weather model to perform a three-layer nested mesoscale simulation on the ERA5 meteorological reanalysis data set and the preset static high-resolution geographic data to obtain the mesoscale simulation result corresponding to the initial moment; interpolating the target wind speed variable extracted from the mesoscale simulation result to the preset pressure height layer to obtain the wind field variable; converting the wind field variable to three-dimensional space coordinates to obtain the target mesoscale wind speed variable, and storing the target mesoscale wind speed variable in CSV (Comma-Separated Values, a file format for storing tabular data in plain text) format to obtain the mesoscale wind field data. In general, a mesoscale weather model WRF is established to simulate complex terrain wind fields. For example Figure 2 As shown in the figure, taking the A County area as an example, the ERA5 (fifth generation ECMWF atmospheric reanalysis of the globalclimate, that is, the fifth generation medium-term weather forecast center climate reanalysis data) reanalysis data set is used as the initial field data, and the static high-resolution geographic data is used as the terrain and land use data. Figure 2 A three-layer nested high-resolution simulation is performed on the simulation area in the simulation area. The horizontal resolution of the outermost grid is 9km, and the grid size is 40×40; the horizontal resolution of the second sub-grid is 3km, and the grid size is also 40×40; the horizontal resolution of the innermost grid is 1km, and the grid size is 28×28. The vertical top of the simulated layer is 5hPa, and the number of vertical layers is 38. The simulation time is: 2024-10-01_00:00:00~2024-10-01_12:00:00, a total of 12 hours, and the time step of the output result is 60 minutes. Then, as shown in the following example, Figure 3As shown in the figure, the wind field data calculated by the mesoscale WRF model is processed, that is, the mesoscale wrfout result file in NC (Network Common Data Format, i.e. NetCDF file) format is analyzed and processed. Specifically, first, the required wind speed variables U, V, W, U10, V10, etc. are extracted from the wrfout file and interpolated to 100hPa, 125hPa, 150hPa, 175hPa, 200hPa, 225hPa, 250hPa, 300hPa, 350hPa, 400hPa, 450hPa, 500hPa, 550hPa, 600hPa, 650hPa, 700hPa, 750hPa, 775hPa, 800hPa, 825hPa, 850hPa, The 27 pressure layers (875hPa, 900hPa, 925hPa, 950hPa, 975hPa, and 1000hPa) were analyzed. The extracted wind field variables (U, V, W, U10, and V10), represented by spatial indices (longitude, latitude, and pressure layer), were then converted into wind speed components (Ux, Uy, and Uz) represented by three-dimensional spatial coordinates (x, y, and z). x and y are still longitude and latitude, and z is the altitude after the pressure layer conversion. The 10-meter wind speed variables for U10 and V10 were integrated into the Ux, Uy, and Uz wind speed components. Finally, the extracted mesoscale wind field data was stored in CSV format for subsequent interpolation.

[0051] Step S12: Perform block interpolation processing on the mesoscale wind field data to obtain a target interpolation result, and input the target interpolation result into a preset LBM microscale model to drive the preset LBM microscale model to perform wind field simulation on the target area until a stable wind field prediction result is obtained, and determine the corresponding current result acquisition time.

[0052] In this embodiment, Figure 4 As shown, the mesoscale wind field data is block-processed based on a preset data blocking method to obtain a target number of first-level interpolation sub-regions, and the first-level interpolation sub-regions are interpolated using a preset multi-level parallel interpolation coupling algorithm to obtain interpolated sub-region results; the preset multi-level parallel interpolation coupling algorithm includes a Cressman interpolation method and a cubic spline interpolation method; the sub-region results located in the LBM (LatticeBoltzmannMethod) microscale calculation domain are screened out from each of the interpolated sub-region results to obtain a first-level interpolation result, and the first-level interpolation results are interpolated based on the preset data blocking method and the preset multi-level parallel interpolation coupling algorithm to obtain a target interpolation result.

[0053] Specifically, the mesoscale wind field data is partitioned based on a preset data partitioning method to obtain a target number of first-level interpolation sub-regions, including: extracting a CSV data file corresponding to a target resolution from the mesoscale wind field data to obtain wind field data to be interpolated, and performing two-dimensional partitioning on the wind field data to be interpolated based on the number of MPI processes corresponding to the wind field data to be interpolated to obtain the number of MPI (Message Passing Interface, a message passing programming model for writing parallel computers) processes, the first-level interpolation sub-regions. That is, the input original data is a 1km resolution wind field CSV data file extracted from the WRF result file, and the entire interpolation area is partitioned according to the number of MPI processes for first-level interpolation calculation. Assuming that there are 64 MPI processes (i.e., 64 ranks) in total, the interpolation area is evenly divided into 64 sub-regions through two-dimensional partitioning. The partitioning is performed row by row, starting from the low-scale value area, such as Figure 5 As shown in the figure, the ranks assigned in the first row are rank 0 to rank 7, and the ranks assigned in the eighth row are rank 56 to rank 63. Each rank is responsible for two-step interpolation calculations (horizontal interpolation and vertical interpolation) for that region. If the edge area is not large enough to be assigned to a rank, that rank only processes the remaining edge area.

[0054] Specifically, the method uses a preset multi-level parallel interpolation coupling algorithm to perform interpolation processing on the first-level interpolation sub-region to obtain the interpolated sub-region result, including: using the Cressman interpolation method to perform horizontal interpolation processing on the first-level interpolation sub-region to obtain the first-level interpolation sub-region result, and using the cubic spline interpolation method to perform vertical interpolation processing on the first-level interpolation sub-region result to obtain the interpolated sub-region result. First of all, it should be noted that in the Cressman interpolation method, the contribution of each observation point (WRF grid point) to the target grid point is inversely proportional to its distance to the target point; Cressman interpolation introduces a fixed search radius (bounded influence domain), and only WRF grid points within this range will participate in the interpolation calculation; in order to control the influence of distant WRF grid points, a smoothing factor ( , usually 1 or a slightly smaller value), so that the weights of points farther away decay faster. The Cressman interpolation method is used to perform horizontal interpolation processing on the first-level interpolation sub-area to obtain the sub-area result after the first-level interpolation processing, including: determining the interpolation influence radius and smoothing factor in the Cressman interpolation method, and performing interpolation calculations on each preset pressure altitude layer of the first-level interpolation sub-area based on the interpolation influence radius and the smoothing factor to obtain the sub-area result after the first-level interpolation processing. Specifically, the distance matrix is ​​first calculated: for two sets, one is the observation point set P, and the other is the grid point set G, each point has its longitude and latitude coordinates, for example, the observation point p i With coordinates , grid point g j With coordinates For each grid point g j , calculate it and all observation points p i Great circle distance between . Here we use the Haversine formula:

[0055] ;

[0056] in, , ; R is the average radius of the earth, usually 6371km. Then, filter out valid points: filter out a set S of known points whose distance is less than or equal to the given influence radius R:

[0057] ;

[0058] Then, calculate the weights: For each valid known point (x i ,y i )∈S, calculate its weight w i :

[0059] ;

[0060] Among them, w i This is the weight of the i-th element, β is the smoothing factor, ε is a non-zero small positive number, d i is the distance. The smoothing factor is introduced to further control the influence of distant observation stations, so that the weights of points farther away decay faster. Furthermore, all weights are normalized so that their sum is 1:

[0061] ;

[0062] .

[0063] Where W is the sum of all element weights, is the normalized weight. Finally, use the normalized weight to calculate the effective known point z i The interpolation result z0 of the grid point (x0, y0) is obtained by weighted averaging the values ​​of

[0064] .

[0065] Using the above-mentioned Cressman interpolation method, each rank processes the horizontal interpolation calculations of all altitude layers within its assigned area in parallel. Figure 6 As shown in the figure, in the calculation example of County A, an interpolation influence radius of 0.05° and a smoothing factor of β=1 were used, and the first-level interpolation result with a resolution of 100m was obtained by interpolating the 27 altitude layers of the original data.

[0066] In this embodiment, after completing the horizontal interpolation, it is necessary to perform the vertical cubic spline interpolation calculation. That is, at the determined grid point position, for the given data points of multiple height layers (x i ,y i ) Construct a series of cubic polynomials S i (x), so that these polynomial segments are continuous and smoothly connected over the entire height interval. i (x) is of the form:

[0067] ;

[0068] Among them, a i , b i , c i , d i is the position coefficient of the interpolation function, which is determined by solving a system of linear equations, x is the position of the interpolation point, x i is the independent variable value of the i-th node.

[0069] Cubic spline interpolation requires conditions, zero-order continuity:

[0070] ;

[0071] .

[0072] in, For node x i The value of the cubic spline function at y i For a given data point; For nodes The value of the cubic spline function, is the next data point.

[0073] First-order continuity:

[0074] ;

[0075] .

[0076] in, For node x i Cubic spline function The first-order derivative value of is the node x i Cubic spline function The first derivative value of ; For nodes Cubic spline function The first-order derivative value of For nodes Cubic spline function The first derivative value of .

[0077] Second-order continuity:

[0078] ;

[0079] .

[0080] in, For node x i Cubic spline function The second-order derivative value of is the node x i Cubic spline function The second-order derivative value of For nodes Cubic spline function The second-order derivative value of For nodes Cubic spline function The second-order derivative value of .

[0081] Natural boundary conditions:

[0082] ;

[0083] .

[0084] in, is the cubic spline function at the first node x0 The second-order derivative value of is the last node x n Cubic spline function The second-order derivative value of .

[0085] After all ranks have completed the horizontal interpolation calculations for all altitude layers in their areas, that is, after all the wind speed data for all 27 altitude layers of all interpolation grid points in the entire area have been calculated, the second step of the two-step interpolation, the vertical interpolation calculation, begins. Each rank independently processes the vertical interpolation of each interpolation point in its assigned area, that is, for each interpolation grid point, a cubic spline interpolation is performed along the height direction. The vertical profile of the wind speed of a single interpolation point is shown in the figure below. Figure 7 As shown. After all ranks complete the cubic spline interpolation calculation in their respective areas, each rank writes its local results into a CSV file, and the main process rank0 selects the rank containing the LBM calculation area, and merges and extracts the local results containing the LBM simulation area and stores them into a CSV file to prepare for the subsequent second-level high-resolution interpolation. For example, in the A County case, a total of 64 ranks performed block interpolation calculations on the entire WRF simulation area, which generated 64 sub-region result CSV files. After regional screening of all ranks, only the sub-region results of ranks 27, 28, 35, and 36 contain the LBM micro-scale calculation area, so only these sub-region results are merged and stored to obtain the first-level coarse-resolution interpolation results.

[0086] In this embodiment, the sub-region results located in the LBM micro-scale calculation domain are screened out from the interpolated sub-region results to obtain the first-level interpolation results, and the first-level interpolation results are interpolated based on the preset data blocking method and the preset multi-level parallel interpolation coupling algorithm to obtain the target interpolation results, including: screening each of the interpolated sub-regions to determine the sub-regions located in the LBM micro-scale calculation region, and then merging the sub-region results located in the LBM micro-scale calculation region, and storing them as a CSV data file to obtain the first-level interpolation results; performing two-dimensional division processing on the wind field data to be interpolated based on the number of MPI processes corresponding to the first-level interpolation results to obtain the corresponding second-level interpolation sub-regions; using the Cressman interpolation method and the cubic spline interpolation method to perform horizontal interpolation processing and vertical interpolation processing on the second-level interpolation sub-region respectively to obtain the target interpolation results. Specifically, the first-level interpolation result file is subjected to regional data screening, and only the LBM calculation area data is extracted. As Figure 8As shown, because the LBM microscale wind field calculation area is much smaller than the WRF simulation area, the data area of ​​the first-level coarse-resolution interpolation results is still larger than the LBM microscale calculation area. Therefore, a second result screening is required. This involves directly filtering the data grid points. Grid data that exists only in the LBM microscale calculation area is selected from the first-level coarse-resolution results to serve as the initial data for the second-level high-resolution interpolation. Then, using the same blocking strategy as the first-level interpolation, the interpolation area is evenly divided into two dimensions based on the number of MPI processes. Each rank is responsible for the two-step interpolation calculation in the horizontal and vertical directions of its partitioned area. The computational process is the same as for the first-level coarse-resolution interpolation. However, because the second-level interpolation resolution is very high, for example, reaching 2.5m in the County A example, the second-level high-resolution interpolation requires a large number of cores. In the Changsha County example, the grid size for the second-level high-resolution interpolation is 1701×1201×255, with a total grid number of 520,939,755. 1024 cores were used in this calculation. Furthermore, the same calculation process as the first-level coarse-resolution interpolation calculation is adopted. Each rank independently processes the cubic spline interpolation along the height direction for all interpolation grid points within its assigned area. That is, the cubic spline interpolation calculation is performed using the data of all 27 height layers at this interpolation grid position, resulting in a continuous cubic inequality S(z) along the z (height direction) at this grid position. After all ranks complete the cubic spline interpolation calculation for their assigned areas, each rank stores the local results of its area in a distributed manner. The main process rank0 summarizes all local results and stores them as a CSV file. The grid wind field data in this CSV file is the refined grid wind field data at the initial time.

[0087] Step S13: Use the mesoscale weather model to obtain the mesoscale simulation result corresponding to the target area at the current result acquisition moment, and jump to the step of performing data processing on the mesoscale simulation result based on the preset data analysis and processing method to obtain mesoscale wind field data, until the current result acquisition moment corresponding to the stable wind field prediction result is the target prediction stop moment to obtain the target wind field prediction result of the target area.

[0088] In this embodiment, the CSV file data obtained by the second step of multi-level parallel interpolation is written into the boundary initial condition file of the LBM micro-scale model, completing the data transfer and driving the LBM micro-scale model calculation until the wind field stabilizes. Then, the WRF simulation results at the same time are used as the original data for multi-level parallel interpolation calculation to obtain the refined grid wind field data at this time. This data is then used to update the boundary condition file of the LBM micro-scale model and continue to drive the LBM micro-scale model to perform calculations. For example, in the A County example, the interpolation results of the WRF mesoscale results are used as the initial boundary conditions of the LBM micro-scale model for calculation. When the calculation reaches 1100 seconds, a fully developed wind field is obtained. Since the calculation time is less than one hour, the result file of the next hour in the WRF simulation time series (relative to the result file at the time of the initial boundary conditions) is multi-level parallel interpolated to obtain the refined grid wind field data of the CSV file at this time. The data on the five surfaces of this file are read to update the boundary conditions of the LBM micro-scale model and continue to drive the model to perform calculations. After multiple updates of boundary conditions, the simulation results of the LBM microscale wind field model are finally obtained after calculation, which is also the wind field prediction result of the WRF-LBM multiscale coupling model. Among them, the wind field prediction results of the A County example are as follows: Figure 9 shown.

[0089] It can be seen that in this embodiment, a mesoscale weather model is used to perform terrain wind field simulation on the target area based on a preset meteorological reanalysis data set and preset static high-resolution geographic data to obtain a mesoscale simulation result corresponding to the initial moment, and the mesoscale simulation result is data-processed based on a preset data analysis and processing method to obtain mesoscale wind field data; the mesoscale wind field data is block-interpolated to obtain a target interpolation result, and the target interpolation result is input into a preset LBM microscale model to drive the preset LBM microscale model to perform wind field simulation on the target area until a stable wind field prediction result is obtained, and the corresponding current result acquisition time is determined; the mesoscale weather model is used to obtain the mesoscale simulation result corresponding to the target area at the current result acquisition time, and jump to the step of performing data processing on the mesoscale simulation result based on the preset data analysis and processing method to obtain mesoscale wind field data, until the current result acquisition time corresponding to the stable wind field prediction result is the target prediction stop time to obtain the target wind field prediction result of the target area. Specifically, the wind data output by the mesoscale weather model is extracted. Using a multi-level parallel interpolation coupling algorithm, the coarse-grid data extracted from the mesoscale weather model simulation results is converted into the fine-grid data required by the LBM model. This enables microscale, high-resolution simulation calculations, resulting in highly accurate wind field forecasts. This significantly reduces computational costs while maintaining high computational efficiency and high-precision meteorological simulations, thereby improving the accuracy of weather forecasts in complex terrain conditions.

[0090] refer to Figure 10 The present application also discloses a regional wind field prediction device, including:

[0091] The wind field data processing module 11 is used to use a mesoscale weather model to perform terrain wind field simulation on the target area based on a preset meteorological reanalysis data set and preset static high-resolution geographic data to obtain a mesoscale simulation result corresponding to the initial time, and to perform data processing on the mesoscale simulation result based on a preset data analysis and processing method to obtain mesoscale wind field data;

[0092] The wind field prediction module 12 is used to perform block interpolation processing on the mesoscale wind field data to obtain a target interpolation result, and input the target interpolation result into a preset LBM microscale model to drive the preset LBM microscale model to perform wind field simulation on the target area until a stable wind field prediction result is obtained, and determine the corresponding current result acquisition time;

[0093] The step jump module 13 is used to use the mesoscale weather model to obtain the mesoscale simulation result corresponding to the target area at the current result acquisition moment, and jump to the step of performing data processing on the mesoscale simulation result based on the preset data analysis and processing method to obtain mesoscale wind field data, until the current result acquisition moment corresponding to the stable wind field prediction result is the target prediction stop moment to obtain the target wind field prediction result of the target area.

[0094] In this example, wind data output by a mesoscale weather model is extracted. Using a multi-level parallel interpolation coupling algorithm, the coarse-grid data extracted from the mesoscale weather model simulation results is converted into the fine-grid data required by the LBM model. This enables microscale, high-resolution simulation calculations, resulting in highly accurate wind forecasts. This significantly reduces computational costs while maintaining high computational efficiency and high-precision meteorological simulations, thereby improving the accuracy of weather forecasts in complex terrain conditions.

[0095] In some specific embodiments, the wind farm data processing module 11 may specifically include:

[0096] a wind field simulation unit, configured to obtain an ERA5 meteorological reanalysis dataset and preset static high-resolution geographic data of a target area, and perform a three-layer nested mesoscale simulation on the ERA5 meteorological reanalysis dataset and the preset static high-resolution geographic data using a mesoscale weather model to obtain a mesoscale simulation result corresponding to an initial time point;

[0097] A wind field variable acquisition unit, configured to interpolate the target wind speed variable extracted from the mesoscale simulation result to a preset pressure altitude layer to obtain a wind field variable;

[0098] The wind field data storage unit is used to convert the wind field variables into three-dimensional space coordinates to obtain target mesoscale wind speed variables, and store the target mesoscale wind speed variables in CSV format to obtain mesoscale wind field data.

[0099] In some specific embodiments, the wind field prediction module 12 may specifically include:

[0100] a first data interpolation processing submodule, configured to perform block processing on the mesoscale wind field data based on a preset data block method to obtain a target number of first-level interpolation subregions, and to perform interpolation processing on the first-level interpolation subregions using a preset multi-level parallel interpolation coupling algorithm to obtain interpolated subregion results; the preset multi-level parallel interpolation coupling algorithm includes a Cressman interpolation method and a cubic spline interpolation method;

[0101] The second data interpolation processing submodule is used to filter out the sub-region results located in the LBM micro-scale calculation domain from each of the interpolated sub-region results to obtain a first-level interpolation result, and interpolate the first-level interpolation result based on the preset data blocking method and the preset multi-level parallel interpolation coupling algorithm to obtain a target interpolation result.

[0102] In some specific embodiments, the first data interpolation processing submodule may specifically include:

[0103] The wind field data blocking unit is used to extract a CSV data file corresponding to a target resolution from the mesoscale wind field data to obtain wind field data to be interpolated, and perform two-dimensional division processing on the wind field data to be interpolated based on the number of MPI processes corresponding to the wind field data to obtain the first-level interpolation sub-areas of the number of MPI processes.

[0104] In some specific embodiments, the first data interpolation processing submodule may specifically include:

[0105] A data interpolation unit is used to perform horizontal interpolation processing on the first-level interpolation sub-region using the Cressman interpolation method to obtain a sub-region result after the first-level interpolation processing, and to perform vertical interpolation processing on the sub-region result after the first-level interpolation processing using the cubic spline interpolation method to obtain a sub-region result after interpolation.

[0106] In some specific embodiments, the data interpolation unit can be specifically used to determine the interpolation influence radius and smoothing factor in the Cressman interpolation method, and based on the interpolation influence radius and the smoothing factor, perform interpolation calculations on each preset pressure altitude layer of the first-level interpolation sub-area to obtain the sub-area result after the first-level interpolation processing.

[0107] In some specific embodiments, the second data interpolation processing submodule may specifically include:

[0108] a first-level interpolation result determination unit, configured to screen the interpolated sub-regions to determine a sub-region within the LBM micro-scale calculation region, and then merge the results of the sub-regions within the LBM micro-scale calculation region and store them as a CSV data file to obtain a first-level interpolation result;

[0109] A region division unit, configured to perform two-dimensional division processing on the wind field data to be interpolated based on the number of MPI processes corresponding to the first-level interpolation result to obtain corresponding second-level interpolation sub-regions;

[0110] The target interpolation result obtaining unit is used to perform horizontal interpolation processing and vertical interpolation processing on the second-level interpolation sub-region using the Cressman interpolation method and the cubic spline interpolation method respectively to obtain a target interpolation result.

[0111] Furthermore, the embodiment of the present application also discloses an electronic device, Figure 11 This is a structural diagram of an electronic device 20 according to an exemplary embodiment. The content in the diagram should not be considered as any limitation to the scope of application of the present application.

[0112] Figure 11 This is a schematic diagram of the structure of an electronic device 20 provided in an embodiment of the present application. The electronic device 20 may specifically include: at least one processor 21, at least one memory 22, a power supply 23, a communication interface 24, an input / output interface 25, and a communication bus 26. The memory 22 is used to store a computer program, which is loaded and executed by the processor 21 to implement the relevant steps of the regional wind field prediction method disclosed in any of the aforementioned embodiments. Furthermore, the electronic device 20 in this embodiment may specifically be an electronic computer.

[0113] In this embodiment, the power supply 23 is used to provide operating voltage for each hardware device on the electronic device 20; the communication interface 24 can create a data transmission channel between the electronic device 20 and the external device. The communication protocol it follows is any communication protocol that can be applied to the technical solution of this application and is not specifically limited here; the input and output interface 25 is used to obtain external input data or output data to the outside world. Its specific interface type can be selected according to specific application needs and is not specifically limited here.

[0114] In addition, the memory 22, as a carrier for resource storage, can be a read-only memory, random access memory, disk or CD, etc. The resources stored thereon can include an operating system 221, a computer program 222, etc., and the storage method can be temporary storage or permanent storage.

[0115] The operating system 221 is used to manage and control the hardware devices and computer program 222 on the electronic device 20, and can be Windows Server, Netware, Unix, Linux, etc. In addition to including a computer program capable of implementing the regional wind field prediction method performed by the electronic device 20 disclosed in any of the aforementioned embodiments, the computer program 222 can further include computer programs capable of implementing other specific tasks.

[0116] Furthermore, this application also discloses a computer-readable storage medium for storing a computer program; wherein, when executed by a processor, the computer program implements the aforementioned regional wind field prediction method. The specific steps of this method can be referred to the corresponding contents disclosed in the aforementioned embodiments and will not be repeated here.

[0117] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from the other embodiments. Reference can be made to the descriptions of the identical or similar parts between the various embodiments. For the devices disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the descriptions are relatively simple, and the relevant parts can be referred to the descriptions of the methods.

[0118] Professionals may further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the above description has generally described the components and steps of each example according to their functions. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians may use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0119] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein may be implemented directly using hardware, a software module executed by a processor, or a combination of the two. The software module may be placed in random access memory (RAM), internal memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, a hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.

[0120] Finally, it should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or device. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of additional identical elements in the process, method, article, or device comprising the element.

[0121] The above is a detailed introduction to the technical solution provided by the present application. Specific examples are used herein to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method of the present application and its core idea. At the same time, for those skilled in the art, according to the ideas of the present application, there may be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. A regional wind field prediction method, characterized in that: include: Using a mesoscale weather model to simulate the terrain wind field of the target area based on a preset meteorological reanalysis dataset and preset static high-resolution geographic data to obtain a mesoscale simulation result corresponding to the initial time, and performing data processing on the mesoscale simulation result based on a preset data analysis and processing method to obtain mesoscale wind field data; Performing block interpolation processing on the mesoscale wind field data to obtain a target interpolation result, and inputting the target interpolation result into a preset LBM microscale model to drive the preset LBM microscale model to perform wind field simulation on the target area until a stable wind field prediction result is obtained, and determining the corresponding current result acquisition time; The mesoscale weather model is used to obtain the mesoscale simulation result corresponding to the target area at the current result acquisition moment, and the step of performing data processing on the mesoscale simulation result based on the preset data analysis and processing method to obtain mesoscale wind field data is jumped to, until the current result acquisition moment corresponding to the stable wind field prediction result is the target prediction stop moment to obtain the target wind field prediction result of the target area.

2. The regional wind field prediction method according to claim 1, characterized in that: The method comprises: performing a terrain wind field simulation on a target area using a mesoscale weather model based on a preset meteorological reanalysis data set and preset static high-resolution geographic data to obtain a mesoscale simulation result corresponding to an initial moment; and performing data processing on the mesoscale simulation result based on a preset data analysis and processing method to obtain mesoscale wind field data, including: Obtain an ERA5 meteorological reanalysis dataset and preset static high-resolution geographic data for the target area, and use a mesoscale weather model to perform a three-layer nested mesoscale simulation on the ERA5 meteorological reanalysis dataset and the preset static high-resolution geographic data to obtain a mesoscale simulation result corresponding to an initial time point; interpolating the target wind speed variable extracted from the mesoscale simulation result to a preset pressure altitude layer to obtain a wind field variable; The wind field variables are converted into three-dimensional space coordinates to obtain target mesoscale wind speed variables, and the target mesoscale wind speed variables are stored in a CSV format to obtain mesoscale wind field data.

3. The regional wind field prediction method according to claim 2, characterized in that: The performing block interpolation processing on the mesoscale wind field data to obtain a target interpolation result includes: The mesoscale wind field data is divided into blocks based on a preset data block method to obtain a target number of first-level interpolation sub-regions, and the first-level interpolation sub-regions are interpolated using a preset multi-level parallel interpolation coupling algorithm to obtain interpolated sub-region results; the preset multi-level parallel interpolation coupling algorithm includes a Cressman interpolation method and a cubic spline interpolation method; The sub-region results located in the LBM micro-scale calculation domain are screened out from the interpolated sub-region results to obtain a first-level interpolation result, and the first-level interpolation result is interpolated based on the preset data blocking method and the preset multi-level parallel interpolation coupling algorithm to obtain a target interpolation result.

4. The regional wind field prediction method according to claim 3, characterized in that: The block processing of the mesoscale wind field data based on a preset data block method to obtain a target number of first-level interpolation sub-regions includes: A CSV data file corresponding to a target resolution is extracted from the mesoscale wind field data to obtain wind field data to be interpolated, and the wind field data to be interpolated is two-dimensionally divided based on the number of MPI processes corresponding to the wind field data to obtain the number of first-level interpolation sub-regions of the MPI processes.

5. The regional wind field prediction method according to claim 4, characterized in that: The interpolation processing is performed on the first-level interpolation sub-region using a preset multi-level parallel interpolation coupling algorithm to obtain an interpolated sub-region result, including: The first-level interpolation sub-region is horizontally interpolated using the Cressman interpolation method to obtain the sub-region result after the first-level interpolation processing, and the first-level interpolation sub-region result is vertically interpolated using the cubic spline interpolation method to obtain the sub-region result after interpolation.

6. The regional wind field prediction method according to claim 5, characterized in that: The method of performing horizontal interpolation processing on the first-level interpolation sub-region by using the Cressman interpolation method to obtain a sub-region result after the first-level interpolation processing includes: An interpolation influence radius and a smoothing factor in the Cressman interpolation method are determined, and interpolation calculations are performed on each preset pressure altitude layer of the first-level interpolation sub-region based on the interpolation influence radius and the smoothing factor to obtain a sub-region result after the first-level interpolation processing.

7. The regional wind field prediction method according to any one of claims 4 to 6, characterized in that: The method further comprises: screening out sub-region results located in the LBM micro-scale computational domain from the interpolated sub-region results to obtain a first-level interpolation result, and interpolating the first-level interpolation result based on the preset data blocking method and the preset multi-level parallel interpolation coupling algorithm to obtain a target interpolation result, including: Screening each of the interpolated sub-regions to determine a sub-region within the LBM micro-scale calculation region, then merging the results of the sub-regions within the LBM micro-scale calculation region and storing them as a CSV data file to obtain a first-level interpolation result; Performing two-dimensional division processing on the wind field data to be interpolated based on the number of MPI processes corresponding to the first-level interpolation result to obtain corresponding second-level interpolation sub-areas; The Cressman interpolation method and the cubic spline interpolation method are used to perform horizontal interpolation processing and vertical interpolation processing on the second-level interpolation sub-region respectively to obtain the target interpolation result.

8. A regional wind field prediction device, characterized in that: include: A wind field data processing module is used to perform terrain wind field simulation on the target area using a mesoscale weather model based on a preset meteorological reanalysis dataset and preset static high-resolution geographic data to obtain a mesoscale simulation result corresponding to an initial time, and to perform data processing on the mesoscale simulation result based on a preset data analysis and processing method to obtain mesoscale wind field data; A wind field prediction module is used to perform block interpolation processing on the mesoscale wind field data to obtain a target interpolation result, and input the target interpolation result into a preset LBM microscale model to drive the preset LBM microscale model to perform wind field simulation on the target area until a stable wind field prediction result is obtained, and determine the corresponding current result acquisition time; A step jump module is used to use the mesoscale weather model to obtain the mesoscale simulation result corresponding to the target area at the current result acquisition moment, and jump to the step of performing data processing on the mesoscale simulation result based on the preset data analysis and processing method to obtain mesoscale wind field data, until the current result acquisition moment corresponding to the stable wind field prediction result is the target prediction stop moment to obtain the target wind field prediction result of the target area.

9. An electronic device, characterized in that: include: Memory, used to store computer programs; A processor, configured to execute the computer program to implement the regional wind field prediction method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that Used to store a computer program, which, when executed by a processor, implements the regional wind field prediction method according to any one of claims 1 to 7.

Images