Wind field prediction method and device

Abstract

The invention provides a wind field prediction method and device, and the method comprises the steps: determining a target region corresponding to a prediction instruction in response to the prediction instruction; acquiring initial simulation conditions, simulation boundary conditions and a three-dimensional fluid model of the target area, wherein the simulation boundary conditions comprise inlet and outlet boundary conditions and earth surface boundary conditions; performing simulation processing according to the simulation initial condition, the simulation boundary condition and the three-dimensional fluid model of the target area to obtain wind field forecast data of the target area; the wind field forecast data comprises wind direction distribution forecast information and wind speed forecast information; and optimizing the wind field forecast data based on the real-time meteorological data of the target area by using a preset fusion algorithm to obtain optimized wind field forecast data. And an accurate wind field prediction result can be obtained.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a wind field prediction method and device. Background Art

[0002] Wind field prediction technology is one of the core technologies in meteorological science, environmental engineering, aerospace and other fields. Its accuracy is directly related to the decision-making reliability of key scenarios such as meteorological disaster warning, wind energy resource development, and safe aircraft takeoff and landing.

[0003] At present, traditional wind forecasting methods mostly rely on numerical weather forecast models, which simulate the evolution of large-scale circulation by solving a set of atmospheric dynamics equations and combine parameterization schemes to approximate complex processes such as cloud microphysics and surface fluxes. However, these models often have certain limitations when dealing with complex terrain and local climate, resulting in insufficient forecast accuracy. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a wind field prediction method and device that can obtain accurate wind field prediction results. The specific solution is as follows:

[0005] A wind field prediction method, comprising:

[0006] In response to the prediction instruction, determining a target area corresponding to the prediction instruction;

[0007] Acquiring simulation initial conditions, simulation boundary conditions, and a three-dimensional fluid model of the target area, wherein the simulation boundary conditions include inlet and outlet boundary conditions and surface boundary conditions;

[0008] Performing simulation processing based on the simulation initial conditions, simulation boundary conditions, and three-dimensional fluid model of the target area to obtain wind field forecast data for the target area; the wind field forecast data includes wind direction distribution forecast information and wind speed forecast information;

[0009] The wind farm forecast data is optimized based on the real-time meteorological data of the target area using a preset fusion algorithm to obtain optimized wind farm forecast data.

[0010] In the above method, optionally, the process of obtaining the simulation initial conditions of the target area includes:

[0011] Acquiring meteorological data, terrain data, and grid configuration information of the target area; the grid configuration information includes a grid type and a grid refinement strategy;

[0012] The meteorological data, terrain data and grid configuration information of the target area are used as initial conditions for simulation of the target area.

[0013] Optionally, the above method includes performing simulation processing based on the simulation initial conditions, simulation boundary conditions, and three-dimensional fluid model of the target area to obtain wind field forecast data for the target area, including:

[0014] Determining a simulation time step corresponding to the target area;

[0015] The preset HPC computing resources are called to perform parallel calculations on the simulation initial conditions, simulation boundary conditions, and three-dimensional fluid model of the target area according to the simulation time step to generate wind field forecast data for the target area.

[0016] Optionally, the above method uses a preset fusion algorithm to optimize the wind farm forecast data based on the real-time meteorological data of the target area to obtain the optimized wind farm forecast data, including:

[0017] A fusion algorithm selected from a Kalman filter algorithm, a weighted average algorithm, and a machine learning algorithm is used to optimize the wind farm forecast data based on the real-time meteorological data of the target area to obtain optimized wind farm forecast data.

[0018] The above method may optionally include optimizing the wind farm forecast data based on the real-time meteorological data of the target area using a preset fusion algorithm, and obtaining the optimized wind farm forecast data, further comprising:

[0019] Obtain actual wind field observation results;

[0020] Optimize the parameters of the fusion algorithm according to the actual observation results.

[0021] The above method may optionally include optimizing the wind farm forecast data based on the real-time meteorological data of the target area using a preset fusion algorithm, and obtaining the optimized wind farm forecast data, further comprising:

[0022] Determine the forecast time scale selected by the user;

[0023] The optimized wind field forecast data is output on a preset display interface according to the forecast time scale.

[0024] A wind field prediction device, comprising:

[0025] a determining unit, configured to determine, in response to the prediction instruction, a target area corresponding to the prediction instruction;

[0026] an acquisition unit, configured to acquire simulation initial conditions, simulation boundary conditions, and a three-dimensional fluid model of the target area, wherein the simulation boundary conditions include inlet and outlet boundary conditions and surface boundary conditions;

[0027] a simulation unit, configured to perform simulation processing based on the simulation initial conditions, simulation boundary conditions, and three-dimensional fluid model of the target area to obtain wind field forecast data for the target area; the wind field forecast data includes wind direction distribution forecast information and wind speed forecast information;

[0028] An execution unit is used to optimize the wind farm forecast data based on the real-time meteorological data of the target area using a preset fusion algorithm to obtain optimized wind farm forecast data.

[0029] In the above device, optionally, the acquisition unit includes:

[0030] An acquisition subunit, configured to acquire meteorological data, terrain data, and grid configuration information of the target area; the grid configuration information includes a grid type and a grid refinement strategy;

[0031] The first execution subunit is configured to use the meteorological data and terrain data and grid configuration information of the target area as initial conditions for simulation of the target area.

[0032] In the above device, optionally, the simulation unit includes:

[0033] A first determining subunit is used to determine a simulation time step corresponding to the target area;

[0034] The generation subunit is used to call preset HPC computing resources, perform parallel calculations on the simulation initial conditions, simulation boundary conditions and three-dimensional fluid model of the target area according to the simulation time step, and generate wind field forecast data for the target area.

[0035] In the above device, optionally, the execution unit includes:

[0036] The second execution subunit is used to optimize the wind field forecast data based on the real-time meteorological data of the target area by using a fusion algorithm selected from the group consisting of a Kalman filter algorithm, a weighted average algorithm, and a machine learning algorithm to obtain optimized wind field forecast data.

[0037] Based on the wind field prediction method and device provided by the above-mentioned embodiment of the present invention, in response to a prediction instruction, the target area corresponding to the prediction instruction is determined; the simulation initial conditions, simulation boundary conditions and three-dimensional fluid model of the target area are obtained, and the simulation boundary conditions include inlet and outlet boundary conditions and surface boundary conditions; simulation processing is performed according to the simulation initial conditions, simulation boundary conditions and three-dimensional fluid model of the target area to obtain wind field forecast data of the target area; the wind field forecast data includes wind direction distribution prediction information and wind speed prediction information; the wind field forecast data is optimized based on the real-time meteorological data of the target area using a preset fusion algorithm to obtain optimized wind field forecast data. Applying the method provided by the embodiment of the present invention, by combining terrain conditions and real-time meteorological data to perform wind field prediction and optimization, the accuracy of wind field forecast data can be effectively improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] 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.

[0039] Figure 1 A flow chart of a wind field prediction method provided by the present invention;

[0040] Figure 2 A flowchart of a process for obtaining simulation initial conditions of a target area provided by the present invention;

[0041] Figure 3 A schematic diagram of a wind farm prediction control process provided by the present invention;

[0042] Figure 4 This is a structural schematic diagram of a wind field prediction device provided by the present invention. DETAILED DESCRIPTION

[0043] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0044] In this application, the terms "comprises," "comprising," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not preclude the presence of additional identical elements in the process, method, article, or apparatus that includes the element.

[0045] The present invention provides a wind field prediction method, which can be applied to electronic equipment, which can be a computer device, such as a server or a server cluster, etc. The flow chart of the method is as follows: Figure 1 As shown, the method includes:

[0046] S101: In response to a prediction instruction, determining a target area corresponding to the prediction instruction.

[0047] In this embodiment, the prediction instruction may be an instruction triggered by a prediction area selected by a user, or may be an instruction automatically triggered during the running of the application.

[0048] S102: Acquire simulation initial conditions, simulation boundary conditions, and a three-dimensional fluid model of the target area, wherein the simulation boundary conditions include inlet and outlet boundary conditions and surface boundary conditions.

[0049] In this embodiment, the simulation initial conditions may include temperature, humidity, air pressure, terrain data, grid configuration information, etc. of the target area.

[0050] Optionally, terrain data of the target area can be acquired using a high-resolution digital elevation model. The terrain data may include information such as altitude, slope, and land cover type. The land cover type may include forest, city, and farmland. In some embodiments, terrain data of the target area may also be acquired using lidar technology to further improve the accuracy of the terrain data.

[0051] In this embodiment, the inlet and outlet boundary conditions in the simulation boundary conditions may include information such as wind speed and wind direction.

[0052] In this embodiment, the surface boundary conditions in the simulation boundary conditions may include information such as surface roughness and heat flux.

[0053] In this embodiment, the three-dimensional fluid model can be selected based on the characteristics of the target area, and the characteristics of the target area may include flow type, turbulence intensity, geometric complexity, near-wall flow requirements, flow stability, etc.

[0054] Optionally, a RANS model can be combined with a k-ε or k-ω turbulence model to obtain a three-dimensional fluid model of the target area. The three-dimensional fluid model includes surface friction factors, heat exchange factors, and turbulent diffusion factors.

[0055] In an embodiment provided in the present application, based on the above solution, optionally, the process of obtaining the simulation initial conditions of the target area is as follows: Figure 2 As shown, including:

[0056] S201: Acquire meteorological data, terrain data, and grid configuration information of the target area; the grid configuration information includes a grid type and a grid refinement strategy.

[0057] In this embodiment, the grid type may include at least one of a structured grid, an unstructured grid, or a hybrid grid. The grid type may be selected based on the terrain complexity of the target area.

[0058] Optionally, the mesh refinement strategy includes mesh refinement in key areas of the target region (e.g., around buildings, on ridges, etc.) to improve computational accuracy, and the number of meshes does not exceed a set threshold to avoid wasting computational resources.

[0059] S202: Using the meteorological data, terrain data and grid configuration information of the target area as initial conditions for simulation of the target area.

[0060] S103: Perform simulation processing according to the simulation initial conditions, simulation boundary conditions and three-dimensional fluid model of the target area to obtain wind field forecast data of the target area; the wind field forecast data includes wind direction distribution forecast information and wind speed forecast information.

[0061] In one embodiment provided in the present application, based on the above solution, optionally, performing simulation processing according to the simulation initial conditions, simulation boundary conditions, and three-dimensional fluid model of the target area to obtain wind field forecast data of the target area includes:

[0062] Determining a simulation time step corresponding to the target area;

[0063] The preset HPC computing resources are called to perform parallel calculations on the simulation initial conditions, simulation boundary conditions, and three-dimensional fluid model of the target area according to the simulation time step to generate wind field forecast data for the target area.

[0064] In this embodiment, the simulation time step may be selected based on the changing characteristics of the flow field in the target area and the calculation stability.

[0065] Optionally, HPC computing resources can decompose the simulation initial conditions (such as initial wind speed and temperature field distribution) and simulation boundary conditions (such as terrain constraints and large-scale circulation forcing) of the target area into multiple sub-computational units according to the spatial domain. Each sub-unit independently performs local flow field evolution calculations within the simulation time step based on the three-dimensional fluid model to generate wind field forecast data for the target area.

[0066] S104: Optimizing the wind farm forecast data based on the real-time meteorological data of the target area using a preset fusion algorithm to obtain optimized wind farm forecast data.

[0067] In this embodiment, the fusion algorithm includes at least one of a Kalman filter algorithm, a weighted average algorithm, and a machine learning algorithm.

[0068] By applying the method provided in the embodiment of the present invention, the accuracy of wind farm forecast data can be effectively improved by combining terrain conditions with real-time meteorological data to perform wind farm prediction and optimization.

[0069] In one embodiment provided in the present application, based on the above solution, optionally, optimizing the wind farm forecast data based on the real-time meteorological data of the target area using a preset fusion algorithm to obtain the optimized wind farm forecast data includes:

[0070] A fusion algorithm selected from a Kalman filter algorithm, a weighted average algorithm, and a machine learning algorithm is used to optimize the wind farm forecast data based on the real-time meteorological data of the target area to obtain optimized wind farm forecast data.

[0071] In this embodiment, a dynamic prediction model is constructed based on the Kalman filter framework. Multidimensional state variables, including temperature, wind speed, and air pressure, and their corresponding observation vectors are defined. The error covariance parameters for each element are initialized through analysis of historical meteorological data. Next, a three-dimensional flow model is used to infer future wind field conditions. The prediction error range is corrected based on the terrain and structures (including rocket models) in the three-dimensional flow model. For example, boundary condition corrections are used to reflect the dynamic impact of structures (including rocket launchers) on airflow. The predicted values are then compared with the observed data in real time. Dynamic weighting is applied when the residual exceeds a preset threshold.

[0072] In some embodiments, a Kalman filter algorithm can be used to construct a state-space model, defining meteorological elements such as temperature, wind speed, and air pressure as state variables. Historical data is then combined to establish a covariance matrix characterizing the correlation between these elements, providing statistical constraints for the initial prediction. When using a computational fluid dynamics (CFD) model to deduce the future spatiotemporal evolution of the wind field, a three-dimensional high-precision terrain model and the geometric parameters of buildings (including special structures such as rocket launchers) are introduced to correct the range of prediction deviations. During the data fusion phase, a weighted average algorithm is used to dynamically assign trust weights between theoretically estimated results and measured data through a multi-dimensional comparison of real-time meteorological observation data and predicted values. This process continuously optimizes the weight allocation strategy through a machine learning algorithm.

[0073] In one embodiment provided in the present application, based on the above solution, optionally, after optimizing the wind farm forecast data based on the real-time meteorological data of the target area using a preset fusion algorithm and obtaining the optimized wind farm forecast data, the method further includes:

[0074] Obtain actual wind field observation results;

[0075] Optimize the parameters of the fusion algorithm according to the actual observation results.

[0076] In this embodiment, after outputting the optimized wind field forecast data, the actual wind field observation results of the target area are synchronously collected, and a parameter correction data set is constructed based on the spatiotemporal matching relationship between the observation results and the forecast data; through the back propagation mechanism in the machine learning model, the spatial distribution characteristics and temporal evolution laws of the prediction error are analyzed, and the covariance matrix parameters of the Kalman filter, the machine learning weight distribution coefficient and the turbulence model correction factor in the fusion algorithm are dynamically adjusted.

[0077] In some embodiments, an abnormal event trigger mechanism is set up to target the instantaneous strong disturbance characteristics generated during the rocket launch phase. When the observed wind speed mutation value exceeds a preset safety threshold, the boundary condition parameters of the local grid are optimized first to form a closed-loop optimization link of prediction-observation-correction, ensuring that the model parameters are adaptively updated as the actual wind field evolves dynamically.

[0078] In one embodiment provided in the present application, based on the above solution, optionally, after optimizing the wind farm forecast data based on the real-time meteorological data of the target area using a preset fusion algorithm and obtaining the optimized wind farm forecast data, the method further includes:

[0079] Determine the forecast time scale selected by the user;

[0080] The optimized wind field forecast data is output on a preset display interface according to the forecast time scale.

[0081] In this embodiment, the optimized wind field data may be displayed in a graphical manner, for example, wind speed, wind direction distribution, wind field animation, etc. may be displayed to facilitate user understanding and application.

[0082] Optionally, the user chooses to select wind forecast data at different time scales.

[0083] The wind field prediction method provided in the embodiments of this application can be applied to shallow wind forecasting. Specifically, through the integration of multiple technologies and refined modeling, dynamic simulation and real-time forecasting of shallow wind fields under complex terrain can be achieved. The specific implementation steps are as follows:

[0084] 1. Data Collection and Preprocessing

[0085] First, comprehensive meteorological data (including temperature, humidity, air pressure, wind speed, wind direction, and precipitation distribution) and terrain data (such as digital elevation models (DEMs), surface roughness, slope, and land cover types, such as forests, urban buildings, or farmland) are collected from meteorological stations, satellite remote sensing (such as weather radar and LiDAR), and a network of ground sensors over the target area. Historical meteorological data is also collected for model calibration and validation. Terrain data is acquired using high-resolution LiDAR technology to accurately characterize the three-dimensional features of complex terrain.

[0086] 2. 3D CFD modeling and meshing

[0087] Based on the flow characteristics of the target area (such as turbulence intensity and terrain complexity), an appropriate CFD model is selected. For shallow wind simulations, the RANS (Reynolds-averaged Navier-Stokes) model combined with the k-ε or k-ω turbulence model is preferred to balance computational efficiency and near-wall flow accuracy. In highly turbulent or transient flow scenarios, a hybrid LES (Large Eddy Simulation) or DES (Detached Eddy Simulation) model can be used. When constructing a three-dimensional flow model, physical processes such as surface friction, heat exchange, and turbulent diffusion must be considered, and boundary conditions (such as inlet and outlet wind speed / direction, surface roughness, and heat flux) must be set. Unstructured meshing technology is used for meshing, with localized mesh refinement performed in complex terrain areas to capture detailed variations in the wind field. A hierarchical meshing strategy is also used to control the overall mesh size to ensure a balance between computational efficiency and accuracy.

[0088] 3. Numerical simulation and parallel computing

[0089] Time-domain simulations are performed using CFD software (such as OpenFOAM and ANSYS Fluent) to calculate the wind speed and direction distribution and the spatiotemporal evolution of shallow winds in the target area. Parallel computing technologies (such as high-performance computing clusters (HPC)) are employed during the simulations to accelerate computational efficiency across large areas. Convergence checks are also performed to ensure the reliability of the results and avoid numerical fluctuations or instability.

[0090] 4. Data fusion and error correction

[0091] CFD simulation results are integrated with real-time meteorological observation data (such as wind speed and direction measured by weather stations and radar). Forecasts are optimized using Kalman filtering, machine learning algorithms (such as LSTM neural networks), or weighted averaging algorithms. Uncertainty quantification is used to assess the impact of input data (such as terrain roughness and initial meteorological conditions) on the forecast, and model parameters are dynamically adjusted to enhance robustness. For example, historical data regression analysis can be used to optimize the weight parameters of the fusion algorithm and reduce model bias.

[0092] 5. Result verification and visualization output

[0093] By comparing simulation results with actual observational data (such as wind speed and direction time series), statistical metrics (such as root mean square error (RMSE) and mean absolute error (MAE)) are used to evaluate model accuracy. A feedback mechanism is established to iteratively optimize model parameters (such as turbulence model coefficients and grid resolution) based on the verification results. Finally, forecast results are output through visualization tools (such as web-based platforms), including wind speed vector diagrams, wind direction distribution maps, wind field animations, and multi-timescale forecast products (hourly, daily, and weekly), meeting application needs in fields such as agricultural disaster prevention, urban management, and shipping safety.

[0094] In this embodiment, by combining CFD technology with traditional meteorological models, the accuracy of shallow wind forecasts is improved, demonstrating particular advantages in complex terrain or urban environments. The use of high-precision ground roughness and building impact models can more realistically reflect wind field changes in actual environments, improving the accuracy of simulation results. By introducing data assimilation techniques (such as Kalman filtering), real-time observation data is combined with CFD simulation results, effectively improving the accuracy and timeliness of shallow wind forecasts. Through optimized grid division and calculation methods, the computational complexity of CFD simulations is reduced, simulation efficiency is improved, and the real-time forecasting needs of large-scale areas are met.

[0095] and Figure 1 Corresponding to the above method, the embodiment of the present invention further provides a wind field prediction device for Figure 1 The specific implementation of the method is that the wind field prediction device is applied to electronic equipment. The structural diagram of the wind field prediction device is as follows Figure 3 As shown, specifically including:

[0096] The determining unit 301 is configured to determine, in response to a prediction instruction, a target area corresponding to the prediction instruction;

[0097] An acquisition unit 302 is configured to acquire simulation initial conditions, simulation boundary conditions, and a three-dimensional fluid model of the target area, wherein the simulation boundary conditions include inlet and outlet boundary conditions and surface boundary conditions;

[0098] The simulation unit 303 is configured to perform simulation processing based on the simulation initial conditions, simulation boundary conditions, and three-dimensional fluid model of the target area to obtain wind field forecast data of the target area; the wind field forecast data includes wind direction distribution forecast information and wind speed forecast information;

[0099] The execution unit 304 is configured to optimize the wind farm forecast data based on the real-time meteorological data of the target area using a preset fusion algorithm to obtain optimized wind farm forecast data.

[0100] In an embodiment provided in the present application, based on the above solution, optionally, the acquiring unit 302 includes:

[0101] An acquisition subunit, configured to acquire meteorological data, terrain data, and grid configuration information of the target area; the grid configuration information includes a grid type and a grid refinement strategy;

[0102] The first execution subunit is configured to use the meteorological data and terrain data and grid configuration information of the target area as initial conditions for simulation of the target area.

[0103] In an embodiment provided in the present application, based on the above solution, optionally, the simulation unit 303 includes:

[0104] A first determining subunit is used to determine a simulation time step corresponding to the target area;

[0105] The generation subunit is used to call preset HPC computing resources, perform parallel calculations on the simulation initial conditions, simulation boundary conditions and three-dimensional fluid model of the target area according to the simulation time step, and generate wind field forecast data for the target area.

[0106] In an embodiment provided in this application, based on the above solution, optionally, the execution unit 304 includes:

[0107] The second execution subunit is used to optimize the wind field forecast data based on the real-time meteorological data of the target area by using a fusion algorithm selected from the group consisting of a Kalman filter algorithm, a weighted average algorithm, and a machine learning algorithm to obtain optimized wind field forecast data.

[0108] In an embodiment provided in this application, based on the above solution, optionally, the following is further included:

[0109] An acquisition unit, used to obtain actual wind field observation results;

[0110] A parameter optimization unit is used to optimize the parameters of the fusion algorithm according to the actual observation results.

[0111] In an embodiment provided in this application, based on the above solution, optionally, the following is further included:

[0112] a processing unit for determining a user-selected forecast time scale;

[0113] The output unit is used to output the optimized wind field forecast data on a preset display interface according to the forecast time scale.

[0114] An embodiment of the present application further provides a storage medium, which includes stored instructions, wherein when the instructions are executed, the device where the storage medium is located is controlled to execute the above-mentioned wind field prediction method.

[0115] The present application also provides an electronic device, the structure of which is shown in FIG. Figure 4 As shown, it specifically includes a memory 401 and one or more instructions 402, wherein the one or more instructions 402 are stored in the memory 401 and are configured to be executed by one or more processors 403 to execute the one or more instructions 402 to perform the above-mentioned wind field prediction method.

[0116] It should be noted that the various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same and similar parts between the various embodiments can be referenced to each other.

[0117] 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.

[0118] For the convenience of description, the above device is described as being divided into various units according to their functions. Of course, when implementing the present invention, the functions of each unit can be implemented in the same or multiple software and / or hardware.

[0119] From the above description of the embodiments, it is clear that those skilled in the art will clearly understand that the present invention can be implemented using software and a necessary general-purpose hardware platform. Based on this understanding, the technical solution of the present invention, or the portion that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a storage medium such as ROM / RAM, a magnetic disk, or an optical disk, and includes instructions for enabling a computer device (such as a personal computer, server, or network device) to execute the methods described in various embodiments of the present invention, or portions thereof.

[0120] The above is a detailed introduction to a wind field prediction method provided by the present invention. Specific examples are used in this article to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core ideas. At the same time, for general technical personnel in this field, according to the ideas of the present invention, there will be changes in the specific implementation methods and application scopes. In summary, the content of this specification should not be understood as limiting the present invention.

Claims

1. A wind field prediction method, characterized in that: include: In response to the prediction instruction, determining a target area corresponding to the prediction instruction; Acquiring simulation initial conditions, simulation boundary conditions, and a three-dimensional fluid model of the target area, wherein the simulation boundary conditions include inlet and outlet boundary conditions and surface boundary conditions; Performing simulation processing based on the simulation initial conditions, simulation boundary conditions, and three-dimensional fluid model of the target area to obtain wind field forecast data for the target area; the wind field forecast data includes wind direction distribution forecast information and wind speed forecast information; The wind farm forecast data is optimized based on the real-time meteorological data of the target area using a preset fusion algorithm to obtain optimized wind farm forecast data.

2. The method according to claim 1, characterized in that The process of obtaining the simulation initial conditions of the target area includes: Acquiring meteorological data, terrain data, and grid configuration information of the target area; the grid configuration information includes a grid type and a grid refinement strategy; The meteorological data, terrain data and grid configuration information of the target area are used as initial conditions for simulation of the target area.

3. The method according to claim 1, characterized in that The performing simulation processing according to the simulation initial conditions, simulation boundary conditions and three-dimensional fluid model of the target area to obtain wind field forecast data of the target area includes: Determining a simulation time step corresponding to the target area; The preset HPC computing resources are called to perform parallel calculations on the simulation initial conditions, simulation boundary conditions, and three-dimensional fluid model of the target area according to the simulation time step to generate wind field forecast data for the target area.

4. The method according to claim 1, wherein The optimizing the wind farm forecast data based on the real-time meteorological data of the target area using a preset fusion algorithm to obtain the optimized wind farm forecast data includes: A fusion algorithm selected from a Kalman filter algorithm, a weighted average algorithm, and a machine learning algorithm is used to optimize the wind farm forecast data based on the real-time meteorological data of the target area to obtain optimized wind farm forecast data.

5. The method according to claim 4, characterized in that After optimizing the wind farm forecast data based on the real-time meteorological data of the target area using a preset fusion algorithm to obtain the optimized wind farm forecast data, the method further includes: Obtain actual wind field observation results; Optimize the parameters of the fusion algorithm according to the actual observation results.

6. The method according to claim 1, characterized in that After optimizing the wind farm forecast data based on the real-time meteorological data of the target area using a preset fusion algorithm to obtain the optimized wind farm forecast data, the method further includes: Determine the forecast time scale selected by the user; The optimized wind field forecast data is output on a preset display interface according to the forecast time scale.

7. A wind field prediction device, characterized in that: include: a determining unit, configured to determine, in response to the prediction instruction, a target area corresponding to the prediction instruction; an acquisition unit, configured to acquire simulation initial conditions, simulation boundary conditions, and a three-dimensional fluid model of the target area, wherein the simulation boundary conditions include inlet and outlet boundary conditions and surface boundary conditions; a simulation unit, configured to perform simulation processing based on the simulation initial conditions, simulation boundary conditions, and three-dimensional fluid model of the target area to obtain wind field forecast data for the target area; the wind field forecast data includes wind direction distribution forecast information and wind speed forecast information; An execution unit is configured to optimize the wind farm forecast data based on the real-time meteorological data of the target area using a preset fusion algorithm to obtain optimized wind farm forecast data.

8. The device according to claim 7, characterized in that The acquisition unit includes: An acquisition subunit, configured to acquire meteorological data, terrain data, and grid configuration information of the target area; the grid configuration information includes a grid type and a grid refinement strategy; The first execution subunit is configured to use the meteorological data and terrain data and grid configuration information of the target area as initial conditions for simulation of the target area.

9. The device according to claim 7, characterized in that The simulation unit comprises: A first determining subunit is used to determine a simulation time step corresponding to the target area; The generation subunit is used to call preset HPC computing resources, perform parallel calculations on the simulation initial conditions, simulation boundary conditions and three-dimensional fluid model of the target area according to the simulation time step, and generate wind field forecast data for the target area.

10. The device according to claim 7, characterized in that The execution unit includes: The second execution subunit is used to optimize the wind field forecast data based on the real-time meteorological data of the target area by using a fusion algorithm selected from the group consisting of a Kalman filter algorithm, a weighted average algorithm, and a machine learning algorithm to obtain optimized wind field forecast data.

Images