Wind power plant point location wind speed and direction prediction method and device based on numerical mode

Abstract

The invention relates to the technical field of meteorological algorithm research and development and application, and discloses a wind power plant point location wind speed and direction prediction method and device based on a numerical mode, and the method comprises the steps: obtaining the standard actual observation data and FNL reanalysis data of a wind power plant; fNL reanalysis data is used as an initial field, mode parameter setting and multiple physical parameter scheme setting are carried out on the mesoscale region numerical mode, and a numerical mode prediction model is formed; obtaining a wind power plant point location wind speed and wind direction preliminary prediction result based on the numerical mode prediction model, and debugging and optimizing the numerical mode prediction model based on the wind power plant point location wind speed and wind direction preliminary prediction result and standard actual observation data to obtain a wind power plant point location wind speed and wind direction prediction model; and predicting the wind power plant point location wind speed and wind direction based on the wind power plant point location wind speed and wind direction prediction model and the preset height. According to the method, fine simulation of the micro-scale wind field in the simulation area and the wind speed and direction at the hub height of each fan point can be realized, and the prediction accuracy is improved.

Description

Technical Field

[0001] The present invention relates to the field of meteorological algorithm research and development and application technology, and in particular to a method and device for predicting wind speed and direction at a wind farm point based on a numerical model. Background Art

[0002] The accurate prediction of wind power and output largely comes from the accurate prediction of wind speed at the hub height of each wind turbine point in the wind farm. The current wind speed prediction of wind turbine points in wind farms has the following problems:

[0003] 1. The accuracy of wind resource forecasts at wind farms is generally low, which directly affects the accuracy of wind power forecasts and output forecasts. At present, the accuracy of wind farm power and output forecasts is generally low, among which meteorological factors have the most significant impact (accounting for about 40-50%). Wind farms are affected by large-scale weather background fields superimposed on small-scale circulations (for example, on land, they are affected by complex terrain uplift, local valley wind thermal circulation, and terrain gravity waves; at sea, they are affected by sea-air interactions, typhoons and other extreme marine weather events, etc.), and the local wind field system is very complex, making the wind resources at wind farms extremely intermittent and unstable. It is difficult to accurately predict wind resources, which directly affects the accuracy of wind power forecasts and output forecasts, and the farms face great assessment pressure.

[0004] 2. The distance between wind turbines in the wind farm area is relatively close, while the simulation resolution commonly used in the global / regional numerical prediction models in the existing technology is relatively coarse, which cannot achieve detailed simulation of the wind field conditions of each wind turbine point in the wind farm area, and is prone to cause large errors.

[0005] 3. Wind farm observation data is lacking and the data quality is poor, and data quality control is needed. At present, there is a lack of observation data for onshore and offshore wind farms, and there are a large number of missing measurements and erroneous values ​​in the existing observation data, which makes the observation data availability poor.

[0006] In addition, wind farms are easily affected by micro-scale turbulence and the wake effect between wind turbines, which poses a greater challenge to the accurate prediction of wind speed at wind turbine locations within the site. Summary of the invention

[0007] In view of this, the present invention provides a method and device for predicting wind speed and direction at wind farm locations based on a numerical model, so as to solve the problem of low accuracy in predicting wind speed at wind turbine locations within a wind farm.

[0008] In a first aspect, the present invention provides a method for predicting wind speed and direction at a wind farm point based on a numerical model, the method comprising:

[0009] Obtain standard actual observation data of wind farms and FNL reanalysis data;

[0010] Using FNL reanalysis data as the initial field, the model parameters and various physical parameter schemes are set for the mesoscale regional numerical model to form a numerical model prediction model; the model parameter settings include the horizontal nested grid settings, the vertical hybrid vertical coordinate system settings and the model top height settings;

[0011] Based on the numerical model prediction model, the preliminary prediction results of wind speed and direction at the wind farm point are obtained, and based on the preliminary prediction results of wind speed and direction at the wind farm point and the standard actual observation data, the numerical model prediction model is debugged and optimized to obtain the wind speed and direction prediction model at the wind farm point;

[0012] The wind speed and direction at the wind farm location are predicted based on the wind farm location wind speed and direction prediction model and the preset height.

[0013] The method for predicting wind speed and direction at a wind farm location based on a numerical model provided by the present invention sets up a nested grid ultra-high resolution numerical simulation test with the wind farm station area as the simulation area, takes 1°×1° FNL reanalysis data as the initial field, sets the model parameters so that the innermost layer uses ultra-high resolution terrain data with a resolution of 3 arc seconds (about 90m), and the innermost grid resolution is as high as 333m. Based on the numerical model prediction model, a preliminary prediction result of the wind speed and direction at the wind farm location is obtained, and based on the preliminary prediction result of the wind speed and direction at the wind farm location and the standard actual observation data, the numerical model prediction model is debugged and optimized to obtain the wind speed and direction at the wind farm location. The wind speed and direction prediction model is based on the wind farm point wind speed and direction prediction model and the preset height to predict the wind speed and direction of the wind farm point, which can realize the refined simulation of the micro-scale wind field in the simulation area and the wind speed and direction at the hub height of each wind turbine point. Combined with the standard measured observation data of the wind farm station, it can improve the accuracy of wind speed prediction at the hub height of each wind turbine point in the wind farm station, and provide technical support and decision-making basis for refined point wind speed forecast of wind farm station, regional wind energy resource assessment, extreme meteorological disaster warning of wind farm station, wind power forecast, etc. At the same time, it saves costs and reduces the cost of site assessment, and solves the problem of low accuracy of wind speed prediction at wind turbine points in wind farm stations.

[0014] In an optional implementation, the standard actual observation data of the wind farm is obtained in the following manner:

[0015] Obtain original observation data of wind farms;

[0016] The original observation data of the wind farm is cleaned by data quality control and data normalization to obtain the standard actual observation data of the wind farm.

[0017] The method for predicting wind speed and direction at wind farm locations based on a numerical model provided by the present invention performs quality control and normalization on the original observation data of the wind farm station to obtain high-quality standard measured observation data, can correctly identify and resolve missing values, erroneous values, etc. in the original observation data, thereby improving the quality of the station observation data, and can effectively correct the wind speed simulation results at the hub height of each wind turbine point in the wind farm station area.

[0018] In an optional implementation, the FNL reanalysis data is used as the initial field, and the model parameters and multiple physical parameter schemes are set for the mesoscale regional numerical model to form a numerical model prediction model including:

[0019] Using FNL reanalysis data as the initial field, the mesoscale regional numerical model is set up with nested grids in the horizontal direction, mixed vertical coordinate system and model top height in the vertical direction, as well as multiple physical parameter schemes.

[0020] Randomly select two or more physical parameter schemes from a variety of physical parameter schemes to combine and build a numerical model prediction model based on the set model parameters.

[0021] The method for predicting wind speed and direction at a wind farm location based on a numerical model provided by the present invention uses 1°×1° FNL reanalysis data as the initial field, thereby achieving the purpose of obtaining a complete reanalysis data set of meteorological elements that includes a wide variety of data with a global wide range and can extend over a long period of time. The model parameters of the mesoscale regional numerical model are set in the horizontal and vertical directions as well as a variety of physical parameter schemes. Two or more physical parameter schemes are arbitrarily selected from the multiple physical parameter schemes to combine and form a numerical model prediction model based on the set model parameters, so that the innermost resolution of the nested grid is as high as 333m, which can well characterize the micro-scale turbulent motion existing in the wind farm station area, thereby realizing the refined simulation of the wind speed and direction at the hub height of the micro-scale wind field and each wind turbine location in the area, which can make up for the shortcomings of observation means, has high research and application value, and provides conditions for subsequent adjustment and optimization of the model. In an optional implementation, the FNL reanalysis data is used as the initial field, and the mesoscale regional numerical model is set up with nested grids in the horizontal direction, a hybrid vertical coordinate system in the vertical direction, and a model top height, as well as multiple physical parameter schemes, including:

[0022] The FNL reanalysis data is used as the initial field, and the model parameters of the mesoscale regional numerical model are set using the first preset number of nested grids in the horizontal direction. The horizontal resolution of each layer of the nested grid is set to a preset distance, and each layer of the grid includes multiple grid points; the innermost grid in the nested grid includes terrain data with a preset resolution;

[0023] In the vertical direction, a hybrid vertical coordinate system with a second preset number of layers is used to set model parameters for the mesoscale regional numerical model, where the hybrid vertical coordinate system includes a terrain following quality coordinate system and an isobaric surface coordinate system;

[0024] Corresponding to each layer of grid in the nested grid, cumulus parameterization scheme, boundary layer scheme, land surface process scheme, longwave radiation scheme, cloud microphysics scheme, surface layer scheme and shortwave radiation scheme are set as multiple physical parameter schemes.

[0025] The method for predicting wind speed and direction at a wind farm location based on a numerical model provided by the present invention adopts a first preset number of nested grids in the horizontal direction to set model parameters for a mesoscale regional numerical model, the horizontal resolution of each layer of grids in the nested grids is set to a preset distance, and each layer of grids includes multiple grid points; the innermost layer of grids in the nested grids includes terrain data of a preset resolution; a second preset number of hybrid vertical coordinate systems are used in the vertical direction to set model parameters for the mesoscale regional numerical model, and the hybrid vertical coordinate system includes a terrain following quality coordinate system and an isobaric surface coordinate system; a cumulus parameterization scheme, a boundary layer scheme, a land process scheme, a long-wave radiation scheme, a cloud microphysics scheme, a surface layer scheme and a short-wave radiation scheme are set correspondingly to each layer of grids in the nested grids as multiple physical parameter schemes, which provides conditions for the subsequent construction of a numerical model prediction model, and adopts the first preset number of nested grids and the second preset number of hybrid vertical coordinate systems to achieve the purpose of fine simulation of the simulation area and improve the prediction accuracy of the prediction model.

[0026] In an optional implementation, the numerical model prediction model is debugged and optimized based on the preliminary prediction results of wind speed and direction at the wind farm location and standard actual observation data, and the wind speed and direction prediction model for the wind farm location is obtained, including:

[0027] Calculate the error between the preliminary wind speed prediction results at the wind farm location and the standard actual observation data;

[0028] When the error does not meet the preset error threshold, multiple physical parameter schemes are recombined to debug and optimize the numerical model prediction model to obtain the wind speed and direction prediction model for the wind farm point.

[0029] The method for predicting wind speed and direction at a wind farm point based on a numerical model provided by the present invention calculates the error between a preliminary prediction result of wind speed at a wind farm point and standard actual observation data; when the error does not meet a preset error threshold, a plurality of physical parameter schemes are recombined to debug and optimize the numerical model prediction model to obtain a wind farm point wind speed and direction prediction model, thereby achieving the purpose of debugging and optimizing the numerical model prediction model based on standard actual observation data and being able to improve the wind speed prediction accuracy at the hub height of each wind turbine point in the wind farm station.

[0030] In an optional embodiment, the preset height includes a plurality of specific heights above each wind turbine point in the wind farm from the ground surface and the height of the wind turbine hub;

[0031] Predicting the wind speed and direction at a wind farm location based on the wind farm location wind speed and direction prediction model and preset height includes:

[0032] The wind speed and direction prediction model of the wind farm point is used to predict the wind speed and direction at multiple specific heights above the ground and at multiple times at the height of the wind turbine hub at each wind turbine point in the wind farm, and the wind speed and direction prediction results are obtained;

[0033] Extract the grid data of wind speed and wind direction values ​​at each moment from the wind speed and wind direction forecast results;

[0034] The interpolation algorithm is used to interpolate the grid data of wind speed and wind direction values ​​at each moment to each wind turbine location to obtain the wind speed and direction values ​​of each wind turbine location in the wind farm.

[0035] The method for predicting wind speed and direction at wind farm points based on a numerical model provided by the present invention adopts a wind farm point wind speed and direction prediction model to predict the wind speed and direction at multiple specific heights above the ground surface and at multiple moments at the height of the wind turbine hub above each wind turbine point in the wind farm, to obtain a wind speed and direction prediction result, and extract grid data of the wind speed value and wind direction value at each moment from the wind speed and direction prediction result; adopts an interpolation algorithm to interpolate the grid data of the wind speed value and wind direction value at each moment to each wind turbine point, to obtain the wind speed and direction value of each wind turbine point in the wind farm, thereby achieving the purpose of fine simulation of the wind speed and direction values ​​of each wind turbine point in the wind farm station, being able to improve the wind speed prediction accuracy at the height of the wind turbine hub at each wind turbine point in the wind farm station, providing technical support and decision-making basis for fine point wind speed forecasting of the station, regional wind energy resource assessment, extreme meteorological disaster warning of the wind farm station, wind power prediction, etc., while saving costs and reducing station assessment costs.

[0036] In an optional implementation, an interpolation algorithm is used to interpolate the grid data of wind speed and wind direction values ​​at each moment to each wind turbine location, and the wind speed and wind direction values ​​of each wind turbine location in the wind farm are obtained, including:

[0037] The horizontal interpolation algorithm is used to interpolate the grid data of wind speed and wind direction at each moment to each wind turbine point, and the one-dimensional height coordinate data from a single specific height above the ground surface to the top of the atmosphere at each moment of each wind turbine point in the wind farm is obtained;

[0038] Based on the one-dimensional height coordinate data at each moment, a vertical interpolation algorithm is used to interpolate the grid data of wind speed and wind direction values ​​at each moment to each wind turbine location to obtain the wind speed and direction values ​​of each wind turbine location in the wind farm.

[0039] The method for predicting wind speed and direction at wind farm locations based on a numerical model provided by the present invention obtains the wind speed and direction values ​​at each wind turbine location in the wind farm through horizontal interpolation and vertical interpolation algorithms. The obtained wind speed and direction values ​​at each wind turbine location in the wind farm are close to the standard measured observation values, thereby improving the prediction accuracy of the wind speed and direction values ​​at each wind turbine location in the wind farm.

[0040] In a second aspect, the present invention provides a device for predicting wind speed and direction at a wind farm point based on a numerical model, the device comprising:

[0041] Data acquisition module, used to obtain standard actual observation data of wind farms and FNL reanalysis data;

[0042] The model building module is used to set the model parameters and various physical parameter schemes of the mesoscale regional numerical model using FNL reanalysis data as input data to form a numerical model prediction model; the model parameter settings include the horizontal nested grid settings, the vertical hybrid vertical coordinate system settings and the model top height settings;

[0043] The model optimization module is used to obtain the preliminary prediction results of wind speed and direction at wind farm locations based on the numerical model prediction model, and to debug and optimize the numerical model prediction model based on the preliminary prediction results of wind speed and direction at wind farm locations and standard actual observation data to obtain the wind speed and direction prediction model at wind farm locations;

[0044] The prediction module is used to predict the wind speed and direction at the wind farm point based on the wind farm point wind speed and direction prediction model and the preset height.

[0045] In a third aspect, the present invention provides a computer device, comprising: a memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, and the processor executing the method for predicting wind speed and direction at a wind farm point based on a numerical model according to the first aspect or any corresponding embodiment thereof by executing the computer instructions.

[0046] In a fourth aspect, the present invention provides a computer-readable storage medium having computer instructions stored thereon, the computer instructions being used to enable a computer to execute the method for predicting wind speed and direction at a wind farm point based on a numerical model according to the first aspect or any corresponding embodiment thereof. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0048] Figure 1 is a flow chart of a method for predicting wind speed and direction at a wind farm location based on a numerical model according to an embodiment of the present invention;

[0049] Figure 2 is a flow chart of another method for predicting wind speed and direction at a wind farm location based on a numerical model according to an embodiment of the present invention;

[0050] Figure 3 is a flow chart of another method for predicting wind speed and direction at a wind farm location based on a numerical model according to an embodiment of the present invention;

[0051] Figure 4 is a schematic diagram of a process of data cleaning according to an embodiment of the present invention;

[0052] Figure 5 is a schematic diagram of jump data according to an embodiment of the present invention;

[0053] Figure 6 is an enlarged schematic diagram of the third to fifth layers of grids in the nested grids used in the numerical model simulation according to an embodiment of the present invention;

[0054] Figure 7 is a schematic diagram of a horizontal interpolation algorithm process according to an embodiment of the present invention;

[0055] Figure 8 is a graph of wind resource prediction results in a pilot application at a first wind farm station according to an embodiment of the present invention;

[0056] Fig. 9 is a graph of wind resource prediction results in a pilot application at a second wind farm according to an embodiment of the present invention;

[0057] Fig.10 is a graph of wind resource prediction results in a pilot application at a third wind farm station according to an embodiment of the present invention;

[0058] Fig.11 is a graph of wind resource prediction results in a pilot application of the fourth wind farm station according to an embodiment of the present invention;

[0059] Fig.12 is a comparison diagram of wind resource prediction results using the method of the present invention and other methods according to an embodiment of the present invention;

[0060] Fig.13 is a structural block diagram of a device for predicting wind speed and direction at a wind farm location based on a numerical model according to an embodiment of the present invention;

[0061] Fig.14 It is a schematic diagram of the hardware structure of a computer device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0062] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are 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 those skilled in the art without creative work are within the scope of protection of the present invention.

[0063] In the description of the present invention, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc., indicating the orientation or positional relationship, are based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first", "second", and "third" are used for descriptive purposes only, and cannot be understood as indicating or implying relative importance.

[0064] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, it can also be the internal connection of two components, it can be a wireless connection, or it can be a wired connection. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0065] In addition, the technical features involved in the different embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0066] According to an embodiment of the present invention, an embodiment of a method for predicting wind speed and direction at a wind farm point based on a numerical model is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in an order different from that shown here.

[0067] In this embodiment, a method for predicting wind speed and direction at a wind farm location based on a numerical model is provided, which can be used in mobile terminals, such as central processing units, servers, etc. Figure 1 is a flow chart of a method for predicting wind speed and direction at a wind farm location based on a numerical model according to an embodiment of the present invention. Figure 1 As shown, the process includes the following steps:

[0068] Step S101, obtaining standard actual observation data of the wind farm and FNL reanalysis data.

[0069] Specifically, standard measured observation data can be obtained through data preprocessing of raw observation data, which can be actual wind speed and direction data. FNL (Final Reanalysis Data) reanalysis data is reanalysis data with a spatial resolution of 1°×1° produced every 6 hours by the National Center for Meteorological and Environmental Prediction (NCEP) and the National Center for Atmospheric Research (NCAR) of the United States. FNL reanalysis data uses the most advanced global data assimilation system and the most complete database to perform quality control and assimilation processing on observation data from various sources, thereby obtaining a complete set of reanalysis data sets, which not only contains a wide variety of meteorological elements and has global data of a large range, but also can extend for a long period of time. Therefore, it is a comprehensive data set and is widely used in analysis and research.

[0070] Step S102, using the FNL reanalysis data as the initial field, setting the model parameters and various physical parameter schemes for the mesoscale regional numerical model to form a numerical model prediction model; the model parameter setting includes the horizontal nested grid setting, the vertical hybrid vertical coordinate system setting and the model top height setting.

[0071] Specifically, using the FNL reanalysis data as the initial field, the model parameters of the mesoscale regional numerical model, including the horizontal nested grid setting, the vertical hybrid vertical coordinate system setting, and the model top height setting, as well as the setting of multiple physical parameter schemes, are set. The physical parameter schemes can be cumulus parameterization schemes for simulating and predicting the formation and development of cumulus clouds, boundary layer schemes for dealing with surface flow fluxes (heat, humidity, momentum) and vertical diffusion in air columns, and land surface process schemes for all physical, chemical, and biological processes occurring on the land surface and their interactions with the atmosphere and the ocean.

[0072] Since there are many physical parameter schemes, different numerical model prediction models will be formed by combining the set model parameters and different physical parameter schemes.

[0073] Step S103, obtaining preliminary prediction results of wind speed and direction at wind farm locations based on the numerical model prediction model, and debugging and optimizing the numerical model prediction model based on the preliminary prediction results of wind speed and direction at wind farm locations and standard actual observation data to obtain a wind speed and direction prediction model at wind farm locations.

[0074] Specifically, different numerical model prediction models formed by different physical parameter scheme combinations are used to predict the wind speed and direction at wind farm locations, and preliminary prediction results of wind speed and direction at multiple wind farm locations are obtained. The preliminary prediction results of wind speed and direction at each wind farm location are compared with standard actual observation data to obtain the optimal physical parameterization scheme combination suitable for short-term wind resource prediction at wind farms. The optimal physical parameterization scheme combination is the obtained wind farm location wind speed and direction prediction model.

[0075] Step S104: predicting the wind speed and direction at the wind farm location based on the wind farm location wind speed and direction prediction model and the preset height.

[0076] Specifically, the wind speed and direction data of each wind turbine location in the target wind farm at different specific heights above the ground and at the hub height are calculated based on the output results of the numerical model prediction model.

[0077] The method for predicting wind speed and direction at a wind farm location based on a numerical model provided in this embodiment is to set up a nested grid ultra-high resolution numerical simulation test with the wind farm station area as the simulation area, use 1°×1° FNL reanalysis data as the initial field, set the model parameters so that the innermost layer uses ultra-high resolution terrain data with a resolution of 3 arc seconds (about 90m), and the innermost grid resolution is as high as 333m. Based on the numerical model prediction model, a preliminary prediction result of the wind speed and direction at the wind farm location is obtained, and based on the preliminary prediction result of the wind speed and direction at the wind farm location and the standard actual observation data, the numerical model prediction model is debugged and optimized to obtain the wind speed and direction at the wind farm location. The wind direction prediction model predicts the wind speed and direction at the wind farm point based on the wind farm point wind speed and direction prediction model and the preset height. It can realize the refined simulation of the micro-scale wind field in the simulation area and the wind speed and direction at the hub height of each wind turbine point. Combined with the standard measured observation data of the wind farm station, it can improve the accuracy of wind speed prediction at the hub height of each wind turbine point in the wind farm station, and provide technical support and decision-making basis for refined point wind speed forecasting of wind farm stations, regional wind energy resource assessment, extreme meteorological disaster warning of wind farm stations, wind power prediction, etc. At the same time, it saves costs and reduces the cost of site assessment, and solves the problem of low accuracy of wind speed prediction at wind turbine points in wind farm stations.

[0078] In this embodiment, a method for predicting wind speed and direction at a wind farm location based on a numerical model is provided, which can be used in mobile terminals, such as central processing units, servers, etc. Figure 2 is a flow chart of a method for predicting wind speed and direction at a wind farm location based on a numerical model according to an embodiment of the present invention. Figure 2 As shown, the process includes the following steps:

[0079] Step S201, obtaining standard actual observation data of the wind farm and FNL reanalysis data.

[0080] Specifically, in the above step S201, the standard actual observation data of the wind farm is obtained in the following manner:

[0081] Step S2011, obtaining original observation data of the wind farm.

[0082] Specifically, the original observation data of the wind farm may be the measured wind speed and direction values ​​of wind turbine locations in the wind farm.

[0083] Step S2012, performing data cleaning processing such as data quality control and data normalization on the original observation data of the wind farm to obtain standard actual observation data of the wind farm.

[0084] Specifically, the original observation data has problems such as missing values ​​and erroneous values. In order to improve the quality of data, enhance data value, reduce redundancy, and accelerate data analysis, the present invention first cleans the original observation data of the wind farm. The specific process includes data quality control and data normalization, such as Figure 4 shown.

[0085] Meteorological data usually produce errors during the observation process. According to the nature of the errors and the causes of the errors, there may be three types of errors in meteorological observation data, namely: random errors (mean is 0), systematic errors (mean is not 0, mainly habitual errors, instrument deviations, etc.) and negligent errors (observation, calculation, transmission, etc.). Random errors are mainly eliminated through multiple sampling, and data quality control is mainly to reduce the impact of the latter two errors on the data.

[0086] Data quality control includes:

[0087] 1) Boundary value check: Boundary value checks are divided into climatological boundary value checks according to their characteristics. Data that does not fall within the boundary value range will be treated as abnormal data. Whether the upper limit of the boundary value range is set reasonably determines the effect of the boundary value check. If the upper limit is set too high, erroneous data may be treated as correct data, resulting in the use of erroneous data in the application; if the upper limit is set too low, individual correct data with large values ​​may be treated as erroneous data, which is very helpful for studying extreme weather and meteorological warnings.

[0088] For example, a wind farm uses the highest and lowest wind speeds in the same period of the past several years as the limit values ​​for wind speed, and locates the wind speed measurement values ​​that exceed this range as erroneous data. The limit value check eliminates the measurement error data caused by other external factors. As shown in Table 1 below, the number 999.0 represents an abnormal value.

[0089] Table 1

[0090]

[0091]

[0092] 2) Time consistency check:

[0093] The time consistency check includes time continuity check and meteorological element characteristic check. The results of missing time point data are shown in Table 2 below.

[0094] Table 2

[0095]

[0096] 2) Meteorological element characteristic inspection is a method of testing whether the change of meteorological element values ​​conforms to specific rules, using the fact that meteorological elements change with time and have certain rules, and will not have large jumps in a relatively short period of time. The sampling value of each meteorological element cannot exceed the range of change within a certain period of time, and the data exceeding the range is suspicious. Figure 5 As shown in the figure, the data circled in the figure are suspicious data and need to be eliminated. Figure 5 The horizontal axis is the time coordinate, and the vertical axis is the climate element data when the wind speed is 100m / s.

[0097] Since meteorological variables have a wide range of fluctuations within a year, using these data directly without correction will cause large errors in calculation and evaluation results. Normalization of data is to scale the data according to the proportion, put the data with different distribution ranges into a common range, eliminate the inconsistency of data unit magnitude, and convert the data into dimensionless pure values, so that indicators of different units and magnitudes can be compared and weighted.

[0098] Data normalization can not only unify the dimensions and smooth the gradients between different batches and layers of data, but also prevent the model gradient from exploding or spreading. After the data is normalized, the "gradient gap" of the original data will be reduced as a whole, effectively preventing a certain layer of data in the model from experiencing large fluctuations, and having the effect of smoothing the gradient. The most typical data normalization operation is to perform data normalization, that is, mapping the data to the interval of [0.1].

[0099] This embodiment uses minimum and maximum normalization to map data values ​​to between [0, 1]. The conversion formula is as follows:

[0100]

[0101] Step S202, using the FNL reanalysis data as the initial field, setting the model parameters and multiple physical parameter schemes for the mesoscale regional numerical model to form a numerical model prediction model; the model parameter setting includes the horizontal nested grid setting, the vertical hybrid vertical coordinate system setting and the model top height setting.

[0102] Specifically, the above step S202 includes:

[0103] Step S2021, using the FNL reanalysis data as the initial field, setting the nested grids in the horizontal direction, the mixed vertical coordinate system and the model top height in the vertical direction, and setting various physical parameter schemes for the mesoscale regional numerical model.

[0104] In some optional implementations, the above step S2021 includes:

[0105] Step a1, using the FNL reanalysis data as the initial field, and using a first preset number of nested grids in the horizontal direction to set the model parameters for the mesoscale regional numerical model, the horizontal resolution of each layer of the nested grids is set to a preset distance, and each layer of the grid includes multiple grid points; the innermost layer of the nested grids includes terrain data of a preset resolution.

[0106] Specifically, by using the 1°×1° FNL reanalysis data as the initial field, the purpose of obtaining a complete reanalysis data set containing a wide variety of meteorological elements with global wide data and extending over a long period of time is achieved. In this embodiment, the first preset number of layers is selected as 5 layers.

[0107] In this embodiment, a numerical model prediction model is built based on the mesoscale regional numerical model WRF (Weather Research and Forecasting, referred to as WRF model), and a detailed simulation study is conducted on the wind speed at the wind turbine locations of four target onshore wind farms in a certain area.

[0108] In this embodiment, 5 layers of nested grids are used, and appropriate wind turbine locations are selected as the center of the simulation area, so that the innermost simulation area contains 4 target onshore wind farms. Arakawa C grids (staggered grids) are used in the horizontal direction. The horizontal resolution of the first layer of grids is 27 km (the distance between any two adjacent grid points is 27 km), and the number of grid points is 150×150; the horizontal resolution of the second layer of grids is 9 km (the distance between any two adjacent grid points is 9 km), and the number of grid points is 250×250; the horizontal resolution of the third layer of grids is 3 km (the distance between any two adjacent grid points is 3 km), and the number of grid points is 150×150; the horizontal resolution of the fourth layer of grids is 1 km (the distance between any two adjacent grid points is 1 km), and the number of grid points is 100×100; the horizontal resolution of the innermost layer, that is, the fifth layer of grids, is 333 m (the distance between any two adjacent grid points is 333 m), and the number of grid points is 211×211. The innermost grid also includes terrain data of preset resolution. The schematic diagram of nested grid distribution is as follows Figure 6 As shown, it is an enlarged schematic diagram of the third to fifth grid layers.

[0109] Due to the complex terrain in the simulation area, the default terrain data of the WRF model cannot clearly display the complete terrain in the simulation area, which will increase the calculation error. Therefore, in addition to the MODIS terrain data (MODIS land cover classification data) with a resolution of 30 arc seconds (about 900m) that comes with the WRF model, the numerical model prediction model uses SRTM (Shuttle Radar Topography Mission) high-resolution terrain data with a resolution of 3 arc seconds (about 90m) in the innermost nested simulation. Adding high-resolution terrain to the WRF model can more accurately depict the terrain features in the simulation area, match the high-resolution numerical model, improve the accuracy of the simulation, and be more reasonable for the simulation of meteorological elements under high-resolution conditions.

[0110] The model top height is set to 50 hPa (hPa is the unit of pressure, hectopascals) in each layer of the grid.

[0111] Step a2, using a hybrid vertical coordinate system with a second preset number of layers in the vertical direction to set model parameters for the mesoscale regional numerical model, the hybrid vertical coordinate system including a terrain following quality coordinate system and an isobaric surface coordinate system.

[0112] Specifically, in this embodiment, the second preset number of layers is selected as 71 layers, and a mixed vertical coordinate system is used in the vertical direction. The coordinate system is a terrain-following mass coordinate in the lower layer, and is converted to an isobaric surface coordinate in the higher layer. A total of 71 layers are set in the vertical direction (that is, the distance between the ground and the top of the atmosphere is evenly divided into 71 parts). The definition of the lower layer and the higher layer is set according to the actual situation. For example, the lower layer can be set as a plane with a large elevation change where the terrain changes greatly, and the higher layer can be set as a plane with a small terrain change and almost equal elevation.

[0113] Step a3, corresponding to each layer of the nested grid, a cumulus parameterization scheme, a boundary layer scheme, a land process scheme, a longwave radiation scheme, a cloud microphysics scheme, a surface layer scheme, and a shortwave radiation scheme are set as multiple physical parameter schemes. Specifically, the cumulus parameterization scheme can adopt the Grell-Devenyi ensemble scheme. This Grell-Devenyi ensemble scheme combines multiple cumulus schemes, performs calculations in each grid point, averages the calculation results, and feeds them back to the model to optimize the model results. In this embodiment, the Grell-Devenyi ensemble scheme is used in the first layer of the nested grid, and the cumulus parameterization scheme is used in the second to fifth layers of the grid.

[0114] The boundary layer schemes include the BMJ (Betts-Miller-Janjic) boundary scheme and the YSU (Yonsei University) boundary scheme. In this embodiment, the BMJ boundary scheme is used in the first to fourth grid layers, and the YSU boundary scheme is used in the fifth grid layer.

[0115] The land surface process scheme adopts the Noah land surface process scheme. In this embodiment, the Noah land surface process scheme is adopted in the first to fifth grid layers.

[0116] The long-wave radiation scheme includes the RRTM (Rapid Radiative Transfer Model) long-wave radiation scheme. In this embodiment, the RRTM long-wave radiation scheme is adopted in the first to fifth grid layers.

[0117] The cloud microphysics scheme includes the Thompson cloud microphysics scheme (Thompson scheme). In this embodiment, the Thompson cloud microphysics scheme is adopted in the first to fifth grid layers.

[0118] The surface layer schemes include the Eta surface layer scheme and the MM5 surface layer scheme. In this embodiment, the first to fourth grid layers adopt the ETA surface layer scheme, and the fifth grid layer adopts the MM5 surface layer scheme.

[0119] The short-wave radiation scheme includes the Dudhia short-wave radiation scheme. In this embodiment, the Dudhia short-wave radiation scheme is adopted in the first to fifth grid layers.

[0120] The specific parameter settings of the model and the selection of physical parameterization schemes are shown in Table 3.

[0121] Table 3

[0122]

[0123] Step S2022, arbitrarily select two or more physical parameter schemes from a plurality of physical parameter schemes to combine the physical parameter schemes and construct a numerical model prediction model based on the set model parameters.

[0124] Specifically, based on the model parameters such as the grid spacing, number of grid points, vertical layer and model top height in Table 1, any two or more physical parameter schemes are selected from the multiple physical parameter schemes in Table 1 to combine the physical parameter schemes to form a numerical model prediction model.

[0125] Step S203: obtain preliminary prediction results of wind speed and direction at wind farm locations based on the numerical model prediction model, and debug and optimize the numerical model prediction model based on the preliminary prediction results of wind speed and direction at wind farm locations and standard actual observation data to obtain a prediction model of wind speed and direction at wind farm locations. Figure 1Step S103 of the illustrated embodiment will not be described in detail here.

[0126] Step S204: predict the wind speed and direction of the wind farm point based on the wind farm point wind speed and direction prediction model and the preset height. Figure 1 Step S103 of the illustrated embodiment will not be described in detail here.

[0127] The method for predicting wind speed and direction at a wind farm location based on a numerical model provided in the present embodiment uses a first preset number of nested grids in the horizontal direction to set model parameters for a mesoscale regional numerical model, the horizontal resolution of each layer of grids in the nested grids is set to a preset distance, and each layer of grids includes multiple grid points; the innermost grid in the nested grids includes terrain data of a preset resolution; a second preset number of hybrid vertical coordinate systems are used in the vertical direction to set model parameters for the mesoscale regional numerical model, and the hybrid vertical coordinate system includes a terrain following mass coordinate system and an isobaric surface coordinate system; a cumulus parameterization scheme, a boundary layer scheme, a land process scheme, a long-wave radiation scheme, a cloud microphysics scheme, a surface layer scheme, and a short-wave radiation scheme are set corresponding to each layer of grids in the nested grids as a variety of physical parameter schemes, which provides conditions for the subsequent construction of a numerical model prediction model, and the innermost resolution is as high as 333m, which can well characterize the micro-scale turbulent motion existing in the wind farm station area, thereby realizing the refined simulation of the wind speed and direction at the hub height of the micro-scale wind field and each wind turbine location in the area, which can make up for the shortcomings of observation means and has high research and application value. The prediction accuracy of the prediction model is improved.

[0128] In this embodiment, a method for predicting wind speed and direction at a wind farm location based on a numerical model is provided, which can be used in mobile terminals, such as central processing units, servers, etc. Figure 3 is a flow chart of a method for predicting wind speed and direction at a wind farm location based on a numerical model according to an embodiment of the present invention. Figure 3 As shown, the process includes the following steps:

[0129] Step S301, obtain standard actual observation data of wind farms and FNL reanalysis data. Figure 2 Step S201 of the illustrated embodiment will not be described in detail here.

[0130] Step S302: Using the FNL reanalysis data as the initial field, the model parameters and various physical parameter schemes are set for the mesoscale regional numerical model to form a numerical model prediction model; the model parameter settings include the horizontal nested grid settings, the vertical hybrid vertical coordinate system settings and the model top height settings. For details, please refer to Figure 2 Step S202 of the illustrated embodiment will not be described in detail here.

[0131] Step S303, obtaining preliminary prediction results of wind speed and direction at wind farm locations based on the numerical model prediction model, and debugging and optimizing the numerical model prediction model based on the preliminary prediction results of wind speed and direction at wind farm locations and standard actual observation data to obtain a wind speed and direction prediction model at wind farm locations.

[0132] Specifically, the constructed numerical model prediction model is used to output preliminary prediction results of wind speed and direction at wind farm locations at hourly intervals.

[0133] In the above step S303, the numerical model prediction model is debugged and optimized based on the preliminary prediction results of wind speed and wind direction at the wind farm point and the standard actual observation data, and the wind speed and wind direction prediction model at the wind farm point is obtained, which includes:

[0134] Step S3031, calculating the error between the preliminary prediction result of the wind speed at the wind farm point and the standard actual observation data.

[0135] Specifically, the difference between the preliminary prediction result of the wind speed at the wind farm point and the standard actual observation data is calculated, and this difference is taken as the error.

[0136] Step S3032, when the error does not meet the preset error threshold, a plurality of physical parameter schemes are recombined to debug and optimize the numerical model prediction model to obtain a wind speed and direction prediction model for a wind farm point.

[0137] Specifically, the preset error threshold is set according to actual conditions and is not specifically limited here. In this embodiment, the smaller the preset error threshold is, the better, that is, the closer the preliminary prediction result of the wind speed at the wind farm point is to the standard actual observation data, the better.

[0138] When the error does not meet the preset error threshold, multiple physical parameter schemes are recombined to form a new combination scheme, and the new combination scheme is used as a new numerical model prediction model. The new numerical model prediction model is then used to predict the wind speed at the wind farm point, and the error between the preliminary prediction result of the wind speed at the wind farm point and the standard actual observation data is continued to be calculated. The error is compared with the preset error threshold until the preset error threshold is met. When there are multiple numerical model prediction models that meet the preset error threshold, the one with the smallest error is selected as the optimal physical parameter scheme combination, that is, the wind speed and direction prediction model for the wind farm point.

[0139] Step S304: predicting the wind speed and direction at the wind farm location based on the wind farm location wind speed and direction prediction model and the preset height.

[0140] Specifically, the preset height includes a plurality of specific heights above each wind turbine point in the wind farm from the ground surface and the height of the wind turbine hub; the above step S304 includes:

[0141] Step S3041, using the wind farm point wind speed and direction prediction model to predict the wind speed and direction at multiple specific heights above the ground surface and at multiple moments at the wind turbine hub height of each wind turbine point in the wind farm, to obtain wind speed and direction prediction results.

[0142] Specifically, multiple specific heights above the ground above each wind turbine point in the wind farm include specific heights of 10m, 30m, 50m, and 70m above the ground above each wind turbine point in the target wind farm. The wind turbine hub height is related to the specific wind turbine model and ranges from 80 to 120m.

[0143] The wind speed and direction prediction model for wind farm points is used to predict the wind speed and direction at specific heights of 10m, 30m, 50m, and 70m above the ground above each wind turbine point in the wind farm and at multiple moments at the height of the wind turbine hub to obtain the wind speed and direction prediction results.

[0144] Step S3042: extracting gridded data of wind speed and wind direction values ​​at each moment from the wind speed and wind direction prediction results.

[0145] Specifically, since the wind field data output by the numerical model prediction model is the three-dimensional U (latitudinal wind component) and V (meridional wind component) grid data at each moment in the simulation area, [horizontal coordinate: longitude (151) vertical coordinate: latitude (151) vertical coordinate: altitude (71)], in order to calculate the simulation results of wind speed and wind direction at a specific height of each wind turbine point in the target wind farm, it is first necessary to vector synthesize the U and V wind components and calculate the wind speed and wind direction values ​​at each grid point in the simulation area.

[0146] Step S3043, using an interpolation algorithm to interpolate the grid data of wind speed and wind direction values ​​at each moment to each wind turbine location, to obtain the wind speed and wind direction values ​​of each wind turbine location in the wind farm.

[0147] In an optional implementation, step S3043 includes:

[0148] Step b1, using a horizontal interpolation algorithm to interpolate the grid data of wind speed and wind direction values ​​at each moment to each wind turbine position, to obtain one-dimensional height coordinate data from a single specific height above the ground surface to the top of the atmosphere at each moment for each wind turbine position in the wind farm.

[0149] Specifically, the horizontal interpolation algorithm is also called the distance weight difference algorithm. The specific process of the horizontal interpolation algorithm is as follows:

[0150] First, the wind turbine point location coordinates O (lat_O, lon_O) of the target wind farm station and the longitude and latitude grid coordinates (lat, lon) [151×151, a total of 22801 points] of the innermost nested area in the numerical model prediction model are input into the wind farm point wind speed and direction prediction model, and the (lat-lat_O) at each grid point is calculated.2 and (lon_lon_O) 2 value, and get (lat-lat_O) 2 and (lon_lon_O) 2 The minimum value of △lat min and △lon min , and then calculate it in the following three cases ( Figure 7 shown):

[0151] A. If △lat min =0 and △lon min = 0, then point O coincides with a grid point (such as Figure 7 In case (2)), no interpolation is required and the point (△lat min =0 and △lon min =0 corresponding grid point) is taken as the simulation result of the wind turbine point of the target wind farm.

[0152] B. If △lat min ≠0 and △lon min ≠0, then point O does not coincide with any grid point (such as Figure 7 In case (1)), we need to calculate (lat-lat_O) 2 and (lon_lon_O) 2 The four smallest values ​​are found, and based on these four smallest values, the four grid points closest to the target wind farm station wind turbine are found, which are ABCD. Then, the weight values ​​are calculated according to the distances AO (the distance from point A to point O), BO (the distance from point B to point O), CO (the distance from point C to point O), and DO (the distance from point D to point O). Taking AO as an example, the distance and weight between A and O are calculated according to the following formula:

[0153]

[0154] Therefore, after horizontal interpolation, the wind speed value y at point O is O for:

[0155]

[0156] Among them, y A ,y B ,y C ,y D They are the wind speed values ​​at the four grid points A, B, C, and D respectively.

[0157] C. If △lat min =0 or △lon min = 0, then point O is on a grid line (such as Figure 7In case (3), the nearest four points are A, B, C, D and E, B, F, D. However, since the distances AO = EO and CO = FO, the weights of the two options are the same, but the wind speed y at point A is A and the wind speed y at point E E 、Wind speed y at point C C and the wind speed y at point F F Different, so we get two kinds of y O The calculation result. At this time, (lat-lat_O) should be calculated 2 and (lon_lon_O) 2 The two smallest values ​​are selected and the two grid points closest to the target wind farm wind turbine are B and D. The weight values ​​are calculated according to the distances BO (the distance from point B to point O) and DO (the distance from point D to point O). At this time, the wind speed value y at point O after horizontal interpolation is O for:

[0158]

[0159] Among them, y B、 y D are the wind speed values ​​at the two grid points B and D respectively.

[0160] Step b2, based on the one-dimensional height coordinate data at each moment, a vertical interpolation algorithm is used to interpolate the grid data of wind speed and wind direction values ​​at each moment to each wind turbine position, so as to obtain the wind speed and wind direction values ​​of each wind turbine position in the wind farm.

[0161] Specifically, after the horizontal interpolation of step b1, the one-dimensional coordinate data [vertical coordinate: height (71)] in the vertical direction from 10m above the ground to the top of the atmosphere (about 50hpa pressure layer height) at each moment above each wind turbine point in the target wind farm station can be obtained. Since the coordinate system in the vertical direction uses a mixed vertical coordinate system, the coordinate system is a terrain-following mass coordinate in the lower layer and is converted to an isobaric surface coordinate in the higher layer, resulting in uneven height distribution in the vertical direction. Therefore, in order to obtain the wind speed and wind direction simulation results at specific heights above each wind turbine point in the target wind farm station, such as 10m, 30m, 50m, 70m and the hub height, it is necessary to further interpolate the one-dimensional wind field data in the vertical direction above each wind turbine point in the target wind farm station.

[0162] Near the ground (sea) surface (within the atmospheric boundary layer), the wind speed changes with height in a logarithmic manner; from 100m above the ground (sea) surface to the top of the troposphere (about 10,000m), the change law is close to power growth. Therefore, the change of wind speed with height is approximately linear growth. In this embodiment, the vertical interpolation algorithm adopts a linear interpolation algorithm for processing. The principle of the linear interpolation algorithm is shown in the following formula:

[0163]

[0164] From formula (6), we can know that we can use the known data (x0, y0) and (x1, y1) to fit a curve, and calculate the y value on the straight line corresponding to a certain position x in the interval [x0, x1]. Among them, x, x0, x1 represent different vertical height data, and y, y0, y1 represent different wind speed data. In this embodiment, there are 71 point coordinates (x0, y1) in the vertical direction above each wind turbine point of the target wind farm station. i ,y i )i=1,2,3...71, which can cover the full height layer distribution from 10m above the ground to the top of the atmosphere (about 50hpa pressure layer height), so as to realize the accurate fitting of height-wind speed in the vertical direction of each point. Therefore, firstly, 71 points in the vertical direction above each wind turbine point of the target wind farm are fitted to form a height-wind speed fitting curve y=f(x), where x represents the height from the ground, and y represents the wind speed value. Then, x=10m, 30m, 50m, 70m and the hub height value of the wind turbine point (80~120m) are substituted into the height-wind speed fitting curve y=f(x), so as to obtain the wind speed and direction values ​​at each specific height of the wind turbine point of each wind farm.

[0165] The wind speed prediction results of the simulation area of ​​the present invention including four target onshore wind farms (the first wind farm pilot, the second wind farm pilot, the third wind farm pilot and the fourth wind farm pilot) are analyzed. The specific analysis results are as follows: Figure 8 , Fig. 9 , Fig.10 and Fig.11 As shown in the figure, the wind resource forecast results of the European Centre for Medium-Range Weather Forecasts (ECMWF) (ERA5 product) are compared. The comparison method is to calculate the root mean square error (RMSE) between the wind resource forecast results of the two methods and the measured wind speed and direction. The smaller the RMSE, the closer the forecast results are to the actual observation results. The specific comparison results are shown in the figure. Fig.12 As shown, it can be seen that the refined wind farm point wind speed calculation method based on numerical model proposed in the present invention can well simulate the wind speed change trend and extreme value distribution of the four target onshore wind farms, and the RMSE index is improved compared with the prediction results of the ECMWF method, with the improvement rate up to 37%.

[0166] The method for predicting wind speed and direction at wind farm locations based on a numerical model provided in this embodiment aims to solve the problem that the accuracy of wind speed at the hub height of wind turbine locations in wind farms is generally low. A refined method for calculating wind speed at wind farm locations based on a numerical model is proposed. The method can achieve refined simulation of wind speed and direction at the hub height of micro-scale wind fields and each wind turbine location in the simulation area. Combined with the high-quality observation data obtained by quality control and homogenization of the original observation data of the wind farm, the accuracy of wind speed prediction at the hub height of each wind turbine location in the wind farm can be improved. This provides technical support and decision-making basis for refined wind speed forecasting at the site, regional wind energy resource assessment, extreme meteorological disaster warning at the wind farm, wind power prediction, etc., while saving costs and reducing site assessment fees.

[0167] In this embodiment, a device for predicting wind speed and direction at a wind farm point based on a numerical model is also provided, which is used to implement the above-mentioned embodiments and preferred implementation modes, and the descriptions that have been made will not be repeated. As used below, the term "module" can be a combination of software and / or hardware that implements a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0168] This embodiment provides a wind farm point wind speed and direction prediction device based on a numerical model, such as Fig.13 As shown, including:

[0169] The data acquisition module 1401 is used to acquire standard actual observation data of the wind farm and FNL reanalysis data.

[0170] The model building module 1402 is used to use the FNL reanalysis data as input data to set model parameters and multiple physical parameter schemes for the mesoscale regional numerical model to form a numerical model prediction model; the model parameter setting includes the nested grid setting in the horizontal direction, the mixed vertical coordinate system setting in the vertical direction, and the model top height setting.

[0171] The model optimization module 1403 is used to obtain preliminary prediction results of wind speed and direction at wind farm locations based on the numerical model prediction model, and to debug and optimize the numerical model prediction model based on the preliminary prediction results of wind speed and direction at wind farm locations and standard actual observation data to obtain a prediction model for wind speed and direction at wind farm locations.

[0172] The prediction module 1404 is used to predict the wind speed and direction at a wind farm location based on the wind farm location wind speed and direction prediction model and a preset height.

[0173] In some optional implementations, the data acquisition module 1401 includes:

[0174] The data acquisition unit is used to acquire the original observation data of the wind farm. The preprocessing unit is used to perform data cleaning processing on the original observation data of the wind farm for data quality control and data normalization to obtain the standard actual observation data of the wind farm.

[0175] In some optional implementations, the model building module 1402 includes:

[0176] The parameter setting unit is used to use the FNL reanalysis data as the initial field, to set the horizontal nested grids, the vertical mixed vertical coordinate system and the model top height, as well as various physical parameter schemes for the mesoscale regional numerical model.

[0177] A model building unit is used to select two or more physical parameter schemes from a plurality of physical parameter schemes to combine the physical parameter schemes and to build a numerical model prediction model based on the set model parameters. In some optional embodiments, the parameter setting unit includes:

[0178] The horizontal parameter setting subunit is used to use the FNL reanalysis data as the initial field, and to set the model parameters of the mesoscale regional numerical model using the first preset number of nested grids in the horizontal direction. The horizontal resolution of each layer of grids in the nested grids is set to a preset distance, and each layer of grids includes multiple grid points; the innermost grid in the nested grids includes terrain data of preset resolution.

[0179] The vertical parameter setting subunit is used to set the model parameters of the mesoscale regional numerical model in the vertical direction using a hybrid vertical coordinate system with a second preset number of layers. The hybrid vertical coordinate system includes a terrain following quality coordinate system and an isobaric surface coordinate system.

[0180] The physical parameter scheme setting subunit is used to set cumulus parameterization scheme, boundary layer scheme, land process scheme, longwave radiation scheme, cloud microphysics scheme, surface layer scheme and shortwave radiation scheme as multiple physical parameter schemes corresponding to each layer of grid in the nested grid.

[0181] In some optional implementations, the model optimization module 1403 includes:

[0182] The error calculation unit is used to calculate the error between the preliminary prediction result of the wind speed at the wind farm point and the standard actual observation data.

[0183] The model optimization unit is used to re-combine multiple physical parameter schemes when the error does not meet the preset error threshold, so as to debug and optimize the numerical model prediction model and obtain the wind speed and direction prediction model for the wind farm point.

[0184] In some optional implementations, the preset height includes a plurality of specific heights above each wind turbine point in the wind farm from the ground surface and the height of the wind turbine hub; the prediction module 1404 includes:

[0185] The wind speed and direction prediction unit is used to predict the wind speed and direction at multiple specific heights above the ground surface and at multiple moments at the wind turbine hub height of each wind turbine point in the wind farm using a wind farm point wind speed and direction prediction model to obtain wind speed and direction prediction results.

[0186] The grid data extraction unit is used to extract the grid data of the wind speed value and wind direction value at each moment from the wind speed and wind direction prediction results.

[0187] The interpolation unit is used to interpolate the grid data of wind speed and wind direction values ​​at each moment to each wind turbine position by using an interpolation algorithm to obtain the wind speed and wind direction values ​​of each wind turbine position in the wind farm.

[0188] In some optional implementations, the interpolation unit includes:

[0189] The horizontal interpolation subunit is used to interpolate the grid data of wind speed and wind direction values ​​at each moment to each wind turbine position using a horizontal interpolation algorithm, and obtain the one-dimensional height coordinate data from a single specific height above the ground surface to the top of the atmosphere at each moment above each wind turbine position in the wind farm.

[0190] The vertical interpolation subunit is used to interpolate the grid data of wind speed and wind direction values ​​at each moment to each wind turbine position based on the one-dimensional height coordinate data at each moment using a vertical interpolation algorithm to obtain the wind speed and wind direction values ​​of each wind turbine position in the wind farm.

[0191] The further functional description of each of the above modules and units is the same as that of the above corresponding embodiments and will not be repeated here.

[0192] The wind speed and direction prediction device for a wind farm point based on a numerical model in this embodiment is presented in the form of a functional unit, where the unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that executes one or more software or fixed programs, and / or other devices that can provide the above functions.

[0193] The embodiment of the present invention also provides a computer device having the above Fig.13 The device shown is a wind farm point wind speed and direction prediction device based on a numerical model.

[0194] See also Fig.14 , Fig.14 is a schematic diagram of the structure of a computer device provided by an optional embodiment of the present invention, such as Fig.14As shown, the computer device includes: one or more processors 10, a memory 20, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. Various components are connected to each other using different buses for communication, and can be installed on a common mainboard or installed in other ways as needed. The processor can process the instructions executed in the computer device, including instructions stored in or on the memory to display the graphical information of the GUI on an external input / output device (such as, a display device coupled to the interface). In some optional embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Similarly, multiple computer devices can be connected, and each device provides some necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system). Fig.14 A processor 10 is taken as an example.

[0195] The processor 10 may be a central processing unit, a network processor or a combination thereof. The processor 10 may further include a hardware chip. The hardware chip may be a dedicated integrated circuit, a programmable logic device or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general purpose array logic or any combination thereof.

[0196] The memory 20 stores instructions executable by at least one processor 10, so that the at least one processor 10 executes the method shown in the above embodiment.

[0197] The memory 20 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created according to the use of the computer device, etc. In addition, the memory 20 may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some optional embodiments, the memory 20 may optionally include a memory remotely arranged relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0198] The memory 20 may include a volatile memory, such as a random access memory; the memory may also include a non-volatile memory, such as a flash memory, a hard disk or a solid state drive; the memory 20 may also include a combination of the above types of memory.

[0199] The computer device also includes an input device 30 and an output device 40. The processor 10, the memory 20, the input device 30 and the output device 40 may be connected via a bus or other means. Fig.14 The example of connecting through bus is taken in the following.

[0200] The input device 30 can receive input digital or character information, and generate key signal input related to the user settings and function control of the computer device, such as a touch screen, a keypad, a mouse, a track pad, a touch pad, an indicator bar, one or more mouse buttons, a trackball, a joystick, etc. The output device 40 may include a display device, an auxiliary lighting device (e.g., an LED) and a tactile feedback device (e.g., a vibration motor), etc. The above-mentioned display device includes but is not limited to a liquid crystal display, a light emitting diode, a display and a plasma display. In some optional embodiments, the display device can be a touch screen.

[0201] The embodiment of the present invention also provides a computer-readable storage medium. The method according to the embodiment of the present invention can be implemented in hardware, firmware, or can be implemented as a computer code that can be recorded in a storage medium, or can be implemented as a computer code that is originally stored in a remote storage medium or a non-temporary machine-readable storage medium and will be stored in a local storage medium through a network download, so that the method described herein can be stored in such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only storage memory, a random access memory, a flash memory, a hard disk or a solid-state hard disk, etc.; further, the storage medium can also include a combination of the above types of memories. It can be understood that a computer, a processor, a microprocessor controller, or programmable hardware includes a storage component that can store or receive software or computer code. When the software or computer code is accessed and executed by a computer, a processor, or hardware, the method shown in the above embodiment is implemented.

[0202] Although the embodiments of the present invention have been described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present invention, and such modifications and variations are all within the scope defined by the appended claims.

Claims

1. A method for predicting wind speed and direction at a wind farm location based on a numerical model, characterized in that: The method comprises: Obtain standard actual observation data of wind farms and FNL reanalysis data; The FNL reanalysis data is used as the initial field to set the model parameters and various physical parameter schemes for the mesoscale regional numerical model to form a numerical model prediction model; the model parameter setting includes the nested grid setting in the horizontal direction, the mixed vertical coordinate system setting in the vertical direction and the model top height setting; Based on the numerical model prediction model, a preliminary prediction result of wind speed and direction at the wind farm point is obtained, and based on the preliminary prediction result of wind speed and direction at the wind farm point and standard actual observation data, the numerical model prediction model is debugged and optimized to obtain a wind speed and direction prediction model for the wind farm point; The wind speed and direction at the wind farm point are predicted based on the wind farm point wind speed and direction prediction model and the preset height.

2. The method according to claim 1, characterized in that Obtain standard actual observation data of wind farms in the following ways: Obtain original observation data of wind farms; The original observation data of the wind farm is subjected to data quality control and data normalization data cleaning processing to obtain standard actual observation data of the wind farm.

3. The method according to claim 1, characterized in that The FNL reanalysis data is used as the initial field to set the model parameters and various physical parameter schemes for the mesoscale regional numerical model, forming a numerical model prediction model including: Using the FNL reanalysis data as the initial field, the mesoscale regional numerical model is subjected to horizontal nested grid settings, vertical hybrid vertical coordinate system settings, model top height settings, and multiple physical parameter scheme settings; Randomly select two or more physical parameter schemes from a variety of physical parameter schemes to combine and build a numerical model prediction model based on the set model parameters.

4. The method according to claim 3, characterized in that: The FNL reanalysis data is used as the initial field, and the mesoscale regional numerical model is subjected to horizontal nested grid setting, vertical hybrid vertical coordinate system setting, model top height setting, and multiple physical parameter scheme setting, including: The FNL reanalysis data is used as an initial field, and a first preset number of nested grids are used in the horizontal direction to set model parameters for the preset mesoscale regional numerical model, wherein the horizontal resolution of each layer of grids in the nested grids is set to a preset distance, and each layer of grids includes a plurality of grid points; and the innermost layer of grids in the nested grids includes terrain data of a preset resolution; In the vertical direction, a hybrid vertical coordinate system with a second preset number of layers is used to set model parameters for the preset mesoscale regional numerical model, wherein the hybrid vertical coordinate system includes a terrain following quality coordinate system and an isobaric surface coordinate system; Corresponding to each layer of grid in the nested grid, cumulus parameterization scheme, boundary layer scheme, land surface process scheme, longwave radiation scheme, cloud microphysics scheme, surface layer scheme and shortwave radiation scheme are set as multiple physical parameter schemes.

5. The method according to claim 1, characterized in that Based on the preliminary prediction results of wind speed and direction at wind farm locations and standard actual observation data, the numerical model prediction model is debugged and optimized to obtain the wind speed and direction prediction model for wind farm locations, including: Calculating the error between the preliminary prediction result of the wind speed at the wind farm point and the standard actual observation data; When the error does not meet the preset error threshold, a plurality of physical parameter schemes are recombined to debug and optimize the numerical model prediction model to obtain a wind speed and direction prediction model for a wind farm point.

6. The method according to claim 4, characterized in that The preset height includes a plurality of specific heights above each wind turbine point in the wind farm from the ground surface and the height of the wind turbine hub; Predicting the wind speed and direction at a wind farm point based on the wind farm point wind speed and direction prediction model and the preset height includes: The wind speed and direction prediction model for wind farm points is used to predict the wind speed and direction at multiple specific heights above the ground surface and at multiple moments at the height of the wind turbine hub at each wind turbine point in the wind farm, to obtain wind speed and direction prediction results; Extracting gridded data of wind speed and wind direction values ​​at each moment from the wind speed and wind direction prediction results; The grid data of the wind speed and wind direction values ​​at each moment are interpolated to each wind turbine location using an interpolation algorithm to obtain the wind speed and wind direction values ​​of each wind turbine location in the wind farm.

7. The method according to claim 6, characterized in that The interpolation algorithm is used to interpolate the grid data of the wind speed value and wind direction value at each moment to each wind turbine point, and the wind speed and direction value of each wind turbine point in the wind farm is obtained, which includes: The grid data of the wind speed value and wind direction value at each moment are interpolated to each wind turbine point using a horizontal interpolation algorithm to obtain one-dimensional height coordinate data from a single specific height above the ground surface to the top of the atmosphere at each moment of each wind turbine point in the wind farm; Based on the one-dimensional height coordinate data at each moment, a vertical interpolation algorithm is used to interpolate the grid data of the wind speed value and wind direction value at each moment to each wind turbine point, so as to obtain the wind speed and direction value of each wind turbine point in the wind farm.

8. A wind speed and direction prediction device for wind farm points based on numerical model, characterized in that: The device comprises: Data acquisition module, used to obtain standard actual observation data of wind farms and FNL reanalysis data; A model building module is used to set model parameters and multiple physical parameter schemes for the mesoscale regional numerical model using the FNL reanalysis data as input data to form a numerical model prediction model; the model parameter setting includes a nested grid setting in the horizontal direction and a mixed vertical coordinate system setting in the vertical direction; A model optimization module is used for the prediction model to obtain preliminary prediction results of wind speed and direction at wind farm locations based on the numerical model prediction model, and to debug and optimize the numerical model prediction model based on the preliminary prediction results of wind speed and direction at wind farm locations and standard actual observation data to obtain a prediction model for wind speed and direction at wind farm locations; The prediction module is used to predict the wind speed and direction of the wind farm point based on the wind farm point wind speed and direction prediction model and the preset height.

9. A computer device, characterized in that: include: A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the method for predicting wind speed and direction at a wind farm point based on a numerical model as described in any one of claims 1 to 7 by executing the computer instructions.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute the method for predicting wind speed and direction at a wind farm point based on a numerical model according to any one of claims 1 to 7.

Images