Method and system for site selection of wind farm wind measurement tower

Abstract

The application relates to the technical field of digital processing, and discloses a wind farm wind measurement tower site selection method and system, which comprises the following steps: acquiring meteorological live conditions, grid prediction and occultation observation data of each spatial grid point in a candidate area, and calculating a terrain disturbance factor; fusing the multi-source meteorological data by the terrain disturbance factor weighting, forming a comprehensive wind field data set, inputting a numerical weather prediction model, and iteratively optimizing the terrain disturbance factor as a physical constraint to obtain wind field prediction data in space-time distribution; and screening grid points meeting preset conditions from the wind field prediction data as wind measurement tower candidate positions. The wind farm wind measurement tower site selection method and system solve the problem of inaccurate wind measurement tower site selection caused by insufficient consideration of the mechanical disturbance and turbulence effect of terrain on the wind field, and simultaneously improve the meteorological accuracy and reliability of the wind measurement tower site selection.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of digital processing, in particular to a wind farm wind measurement tower site selection method and system. BACKGROUND

[0002] The development and operation of a wind farm highly depend on accurate wind resource assessment, and a wind measurement tower is a core device for obtaining key meteorological parameters such as wind speed and wind direction at a near-ground height layer (usually 10m-100m height layer). The rationality of the wind measurement tower site selection directly affects the representativeness and reliability of the wind resource data - if the site selection location fails to fully reflect the dominant wind field pattern (such as stable wind direction, uniform wind speed distribution) of the wind farm region, it may lead to wind resource assessment deviation, thereby affecting the accuracy of wind farm layout design, unit selection and power generation prediction, and ultimately increasing the development risk and cost.

[0003] The existing wind measurement tower site selection method combines the slope data of a digital elevation model (DEM) and the wind field data predicted by a numerical weather prediction (NWP) model to determine the wind measurement tower site selection according to the wind field prediction data analysis. This method ignores the influence of terrain disturbance on wind direction and wind speed: terrain can affect airflow movement, for example, in complex terrain such as mountains and valleys, airflow can accelerate, flow around, and vortex, etc. The wind speed observed by a ground meteorological station may have a large deviation from the actual wind speed of the large-scale wind field due to local terrain shielding or acceleration effect. Complex terrain can make the wind direction unstable, resulting in local changes in wind direction. For example, the orientation of ridges and valleys can affect airflow direction, leading to large differences in wind direction at different locations and different times. Since the terrain disturbance is not fully considered, the obtained wind direction data cannot represent the true dominant wind direction of the region, thereby affecting the judgment of the wind field pattern; the obtained wind speed data cannot truly reflect the actual wind resource situation of the region, so that the wind direction and wind speed data for site selection do not match the actual development height layer. SUMMARY

[0004] Therefore, the purpose of the present application is to overcome the problem of inaccurate wind measurement tower site selection due to insufficient consideration of the mechanical disturbance and turbulence effect of terrain on the wind field in the prior art, and to provide a wind farm wind measurement tower site selection method and system to simultaneously improve the meteorological accuracy and reliability of the wind measurement tower site selection.

[0005] In a first aspect, to solve the above technical problems, the present application provides a wind farm wind measurement tower site selection method, comprising:

[0006] dividing a candidate region into a plurality of spatial grid points, obtaining meteorological live data, grid prediction data and occultation observation data of the plurality of spatial grid points to form a multi-source meteorological data set;

[0007] obtain terrain data of each of the spatial grid points, and calculate a terrain disturbance factor of the spatial grid point according to the terrain data; wherein the terrain disturbance factor comprises a velocity shear exponent and a turbulence intensity;

[0008] assign a weight based on the terrain disturbance factor, and perform weighted fusion on multi-source meteorological data according to the weight to obtain a comprehensive wind field data set;

[0009] input the comprehensive wind field data set into a numerical weather prediction model, and perform wind field iterative optimization with the velocity shear exponent and the turbulence intensity as physical constraints to obtain wind field prediction data in space-time distribution;

[0010] filter the spatial grid points that meet a preset condition according to the wind field prediction data to obtain candidate locations of a wind measurement tower.

[0011] Preferably, the terrain data comprises an altitude, a slope angle and a slope direction angle; calculating the velocity shear exponent according to the terrain data comprises: constructing a logarithmic law model; the logarithmic law model represents a logarithmic variation relationship between a wind speed and a distance from a ground height in a near-ground boundary layer; determining a terrain modulation factor based on the slope angle and the slope direction angle; correcting the logarithmic law model based on the terrain modulation factor to obtain a wind speed profile model; inputting the altitude into the wind speed profile model to obtain a wind speed at a current altitude; and calculating the velocity shear exponent according to the wind speed at the current altitude and the altitude.

[0012] Preferably, determining the terrain modulation factor based on the slope angle and the slope direction angle comprises:

[0013] ;

[0014] f represents the terrain modulation factor; k represents an empirical coefficient, and the value is 0.15-0.25; represents the slope angle; represents the slope direction angle; represents a dominant wind direction angle of the candidate region.

[0015] Preferably, calculating the turbulence intensity according to the terrain data comprises: obtaining a wind speed in a set time window according to the wind speed profile model to generate a time series wind speed; calculating a wind speed standard deviation and an average wind speed according to the time series wind speed; and calculating a ratio of the wind speed standard deviation to the average wind speed to obtain the turbulence intensity.

[0016] Preferably, assigning weights based on the terrain disturbance factor comprises: if the velocity shear index is greater than 0.3, increasing the weight of the weather live data; if the velocity shear index is less than or equal to 0.3, increasing the weight of the grid forecast data; if the turbulence intensity is greater than 15%, reducing the weight of the occultation observation data; if the turbulence intensity is less than or equal to 15%, increasing the weight of the occultation observation data.

[0017] Preferably, inputting the comprehensive wind field data set into a numerical weather prediction model to perform wind field iterative optimization with the velocity shear index and the turbulence intensity as physical constraints to obtain wind field prediction data with spatial and temporal distribution, comprising:

[0018] Step one: embedding the velocity shear index as a correction term of mixing length in the boundary layer parameterization of the numerical weather prediction model to obtain a corrected numerical weather prediction model;

[0019] Step two: running the corrected numerical weather prediction model to generate first-round wind speed and direction prediction values;

[0020] Step three: comparing the first-round wind speed and direction prediction values with observation values, if the error exceeds a threshold, adjusting the parameterization coefficients of the corrected numerical weather prediction model, and returning to step one;

[0021] Step four: until the wind field prediction data is obtained through iterative convergence.

[0022] Preferably, in step one, further comprising: introducing a turbulence intensity driven turbulent kinetic energy term in the turbulence parameterization of the numerical weather prediction model.

[0023] Preferably, further comprising clustering the wind field prediction data, and screening to obtain the wind tower candidate position according to the clustering result, which comprises: extracting statistical features and temporal variability features of each spatial grid point from the wind field prediction data to construct a wind field feature vector; wherein the statistical features include maximum wind speed, minimum wind speed, average wind speed, wind speed variation coefficient and wind direction standard deviation; the temporal variability features include periodicity features or trend change indicators; clustering the wind field feature vectors of all spatial grid points to divide them into multiple wind field characteristic clusters according to wind field characteristic similarity.

[0024] Preferably, screening the spatial grid points meeting the preset conditions as the wind tower candidate positions comprises: extracting the average wind speed, the wind speed variation coefficient, the wind direction standard deviation and the position information of the spatial grid points in the wind field characteristic cluster from the plurality of wind field characteristic clusters; and screening the spatial grid points meeting the conditions of the average wind speed being not less than 5 m / s, the wind direction standard deviation being not greater than 30°, the wind speed variation coefficient being not greater than 0.2 and being in the central region of the wind field characteristic cluster as the wind tower candidate positions.

[0025] In a second aspect, to solve the above technical problems, the present application provides a wind farm wind tower site selection system, comprising:

[0026] A data acquisition module acquires meteorological real-time data, grid forecast data and occultation observation data of a plurality of spatial grid points in a candidate area to form a multi-source meteorological data set;

[0027] A terrain processing module calculates a terrain disturbance factor based on terrain data of the spatial grid points; the terrain disturbance factor comprises a velocity shear index and a turbulence intensity;

[0028] A data fusion module assigns a weight to the multi-source meteorological data according to the terrain disturbance factor, and performs weighted fusion on the multi-source meteorological data according to the weight to obtain a comprehensive wind field data set;

[0029] A wind field prediction module inputs the comprehensive wind field data set into a numerical weather prediction model to perform wind field iterative optimization with the velocity shear index and the turbulence intensity as physical constraints to obtain wind field prediction data with spatial and temporal distribution;

[0030] A site selection decision module screens the spatial grid points meeting the preset conditions as wind tower candidate positions according to the wind field prediction data.

[0031] The above technical solutions of the present application have the following beneficial effects compared with the prior art:

[0032] The wind farm wind tower site selection method and system of the present application significantly improve the wind field data accuracy, multi-source data fusion reliability and wind field prediction accuracy by quantifying the terrain disturbance factor and integrating multi-source data fusion and model optimization, accurately screen the wind tower positions with rich wind resources, stable wind field mode and strong data representativeness, effectively solve the data deviation and site selection deviation problems of the traditional site selection method under complex terrain, simultaneously improve the meteorological accuracy and reliability of the wind tower site selection, and are especially suitable for complex terrain areas such as mountains and coastal belts, providing reliable data support for wind power project approval and power grid dispatching.

[0033] Wherein the influence of the terrain on the wind speed vertical shear and airflow irregularity is quantified by the velocity shear index and the turbulence intensity, so that the comprehensive wind field data set can more truly reflect the actual wind conditions of the candidate area, and solve the problems of near-surface wind speed measurement deviation and wind direction stability misjudgment caused by ignoring terrain disturbance.

[0034] Based on the terrain disturbance factor, the fusion weight of meteorological live data, grid forecast and occultation observation data is dynamically adjusted, the weight of occultation data is increased and the weight of ground live data and grid forecast data is reduced in the complex terrain area, so that the adaptability of multi-source meteorological data in different terrain conditions and the accuracy of fusion results are significantly improved.

[0035] The velocity shear index and the turbulence intensity are embedded as physical constraints into the iterative optimization process of the numerical weather prediction model, so that the wind field prediction data output by the model is more in line with the real wind field characteristics under the influence of the terrain, and the problem of insufficient simulation accuracy of the vertical shear of wind speed and the turbulence structure of the traditional model under complex terrain is solved. BRIEF DESCRIPTION OF DRAWINGS

[0036] In order to make the content of the application easier to be clearly understood, the application will be further described in detail below according to the specific embodiments of the application and in combination with the drawings, in which:

[0037] Figure 1 The flowchart of the wind farm wind measurement tower site selection method in the preferred embodiment of the application;

[0038] Figure 2 The flowchart of obtaining the velocity shear index in the preferred embodiment of the application;

[0039] Figure 3 The flowchart of obtaining the wind field prediction data in the preferred embodiment of the application;

[0040] Figure 4 The structure block diagram of the wind farm wind measurement tower site selection system in the preferred embodiment of the application. DETAILED DESCRIPTION

[0041] The application will be further described below in combination with the drawings and specific embodiments, so that those skilled in the art can better understand the application and implement it, but the embodiments are not as a limitation on the application.

[0042] Terrain is one of the core factors affecting the distribution of near-surface wind field - complex terrain such as mountains, hills and valleys will significantly affect the wind speed, wind direction stability and spatial distribution characteristics by changing the airflow movement path (such as acceleration effect, flow around, vortex generation), inducing wind speed vertical shear and airflow irregularity. Specifically, (1) velocity shear effect: when the airflow passes through the terrain with slope change (such as the acceleration of the windward slope of the mountain and the deceleration of the leeward slope), the wind speed difference of different height layers (i.e. velocity shear index, reflecting the rate of change of wind speed with height) significantly increases, and the near-surface layer (such as 10m-30m height) wind speed may be significantly weakened due to friction, while the high altitude wind speed is relatively stable. (2) Turbulence effect: complex terrain (such as broken ridges and canyon exits) will cause airflow separation, vortex shedding and other phenomena, which will significantly increase the fluctuation amplitude of wind speed (i.e. turbulence intensity), affecting the stable operation of wind turbine generators.

[0043] The purpose of the embodiment of the present application is to solve the problem that the mechanical disturbance (i.e. velocity shear effect) and turbulence effect of terrain on wind field are not fully considered, resulting in inaccurate wind tower site selection.

[0044] Embodiment one: refer to Figure 1 The embodiment of the present application discloses a wind farm wind tower site selection method, comprising:

[0045] S100, divide the candidate area into a plurality of spatial grid points, obtain the meteorological real-time data, grid forecast data and occultation observation data of the plurality of spatial grid points, and construct a multi-source meteorological data set;

[0046] S200, obtain the terrain data of each spatial grid point, and calculate the terrain disturbance factor of the spatial grid point according to the terrain data; wherein the terrain disturbance factor includes the velocity shear index and the turbulence intensity;

[0047] S300, assign a weight based on the terrain disturbance factor, and perform weighted fusion on the multi-source meteorological data according to the weight to obtain a comprehensive wind field data set;

[0048] S400, input the comprehensive wind field data set into a numerical weather prediction model, and perform wind field iterative optimization with the velocity shear index and the turbulence intensity as physical constraints to obtain wind field prediction data with spatial and temporal distribution;

[0049] S500, according to the wind field prediction data, screen the spatial grid points meeting the preset condition as wind tower candidate positions.

[0050] In specific application scenarios, based on a geographic information system (GIS) tool, using digital elevation model (DEM) data or vector boundary files, the candidate area is discretized into a plurality of uniformly distributed spatial grid points through a grid algorithm (such as regular rectangular grid division), and each spatial grid point has a clear latitude and longitude coordinate and a spatial range. For each spatial grid point, collect the actual observed meteorological live data, the model predicted grid forecast data, and the satellite inversion occultation observation data. The three types of data, meteorological live data, grid forecast data, and occultation observation data, together constitute a multi-source meteorological data set describing the wind field characteristics of the area.

[0051] The meteorological live data includes the historical or real-time observation data obtained from the ground meteorological stations (such as national meteorological stations and wind measurement stations around wind farms), including wind speed (at different height layers, such as 10m, 30m, and 50m), wind direction, air temperature, air pressure, and other parameters. The gridded forecast data for a future period (such as 7 days or 1 month) is obtained from numerical weather prediction models (such as ECMWF, GFS, WRF, etc.), with a resolution of 1km-10km, directly including wind speed, wind direction, and other parameters for each grid point. The atmospheric wind field data inverted using global navigation satellite system occultation technology (such as GPS / MET, COSMIC) extracts the wind speed and wind direction parameters at the corresponding height layer (such as 50m-200m) for each spatial grid point through satellite orbit coverage information, supplementing the high-altitude wind field information. Align the three types of data according to the grid point coordinates to form a three-dimensional data structure of "grid point- meteorological parameter-time".

[0052] Extract the terrain parameters of each spatial grid point through high-resolution digital elevation model (DEM), such as ASTER GDEM and SRTM 30m / 90m data, including elevation, slope angle, and slope direction angle. The elevation is the absolute altitude of the grid point. The slope angle is the surface inclination angle calculated based on the elevation difference of adjacent grid points. The slope direction angle is the angle between the slope direction and the dominant wind direction. Calculate the velocity shear index and turbulence intensity based on the elevation, slope angle, and slope direction angle of each spatial grid point. The velocity shear index represents the vertical wind speed profile gradient, i.e., the nonlinear change rate of wind speed with height, reflecting the acceleration / deceleration characteristics of near-surface layer wind speed. The turbulence intensity represents the irregular pulsation degree of airflow, reflecting the stability of the wind field. Quantify the influence of terrain on wind speed vertical shear and airflow irregularity through velocity shear index and turbulence intensity, so that the comprehensive wind field data set can more realistically reflect the actual wind conditions of the candidate area, solving the problems of near-surface wind speed measurement bias and wind direction stability misjudgment caused by ignoring terrain disturbance.

[0053] Different meteorological data sources have different response capabilities to terrain disturbance (such as occultation data being less affected by terrain obstruction and being more reliable, and ground data being easily affected by local interference and being less reliable), and the weights are dynamically assigned according to the differences in response capabilities to terrain disturbance, the influence of terrain disturbance intensity on data reliability is reflected, and more accurate and reliable comprehensive wind field data set is obtained by weighting and fusing three types of meteorological data according to the weights, so that the adaptability of multi-source meteorological data under different terrain conditions and the accuracy of fusion results are significantly improved.

[0054] The comprehensive wind field data set is used as the initial wind speed / direction of the numerical weather prediction model, and the terrain data, land use data, ground roughness and other basic geographic information are input at the same time, in the model calculation process, the target function is constructed to quantify the deviation of the predicted wind field and the terrain disturbance factor, the key parameters (such as the friction coefficient of the near-surface layer and the turbulent kinetic energy parameter) of the model or the initial wind field distribution are adjusted through gradient descent method or genetic algorithm, the target function value is minimized, the model runs multiple iterations, after each round of wind field prediction result is output, the wind speed shear index and the turbulent intensity of the current prediction are recalculated, compared with the actual terrain disturbance factor and the weight is updated, until the deviation between the prediction result and the constraint condition is less than the set threshold, and finally the optimized spatiotemporal distribution wind field prediction data is obtained, including the wind speed and direction of each spatial grid point at different height layers / time points. By introducing the velocity shear index and the turbulent intensity as physical constraint conditions in the model calculation process, the initial field, boundary conditions or parameterization scheme of the model are adjusted through iterative optimization, so that the spatiotemporal distribution wind field data output by the model is more consistent with the characteristics of the real wind field under the influence of the actual terrain.

[0055] Based on the optimized wind field prediction data, the grid points with rich wind resources, stable wind conditions and regional representativeness are selected as the candidate positions of the wind tower through the preset wind resource evaluation index (such as average wind speed threshold and wind direction stability requirement).

[0056] The wind farm wind tower site selection method of the application significantly improves the accuracy of wind field data, the reliability of multi-source data fusion and the accuracy of wind field prediction by quantifying the terrain disturbance factor and integrating multi-source data fusion and model optimization, accurately selects the wind tower position with rich wind resources, stable wind field mode and strong data representativeness, effectively solves the data deviation and site selection deviation problem of traditional site selection method under complex terrain, simultaneously improves the meteorological accuracy and reliability of wind tower site selection, and is especially suitable for complex terrain areas such as mountains and coastal belts, providing reliable data support for wind power project approval and power grid scheduling.

[0057] On the basis of the above embodiment, the terrain data includes altitude, slope angle and slope direction angle; refer to Figure 2As shown, the speed shear index is calculated according to the terrain data, including: constructing a logarithmic law model; the logarithmic law model represents the logarithmic change relationship between the wind speed and the distance from the ground height in the near-surface boundary layer; determining a terrain modulation factor based on the slope angle and the aspect angle; correcting the logarithmic law model based on the terrain modulation factor to obtain a wind speed profile model; inputting the altitude into the wind speed profile model to obtain the wind speed at the current altitude; and calculating the speed shear index according to the wind speed at the current altitude and the altitude.

[0058] In a specific application scenario, in the near-surface boundary layer, when the atmosphere is in a neutral stable state, the wind speed changes with the height in accordance with the logarithmic law, the core assumption of which is that the airflow forms a turbulent boundary layer under the action of surface friction, and the vertical distribution of the wind speed satisfies the logarithmic relationship, and the logarithmic law model is:

[0059]

[0060] wherein u(z) represents the wind speed at the height z; and z represents the altitude; wherein u(z0) represents the von Karman constant, and the value is 0.4; z0 represents the virtual plane height at which the wind speed is zero, and is related to the type of the ground surface, and the value of the smooth water surface is 0.0001-0.001 m; the value of the flat grassland is 0.01-0.05 m; the value of the shrub / crop is 0.1-0.3 m; and the value of the forest / city is 0.5-2.0 m.

[0061] wherein u(z) represents the friction velocity, representing the surface friction strength; and the friction velocity is calculated and determined in the following manner: wherein τ(z) represents the surface shear stress; wherein ρ represents the air density.

[0062] The terrain modulation factor is determined based on the slope angle and the aspect angle, including:

[0063]

[0064] wherein f represents the terrain modulation factor; and k represents an empirical coefficient, and the value is 0.15-0.25; wherein θ represents the slope angle; wherein φ represents the aspect angle; wherein θ0 represents the dominant wind direction angle of the candidate area, and is determined by a historical wind rose diagram, for example, the wind direction frequency distribution of the candidate area is counted for many years, and the wind direction with the highest frequency is selected as the dominant wind direction angle.

[0065] The slope angle wherein θ represents the inclination angle of a point, quantifies the inclination degree of the slope surface--the steeper the slope, ​​​The larger the height, the stronger the vertical movement component of the airflow forced by the terrain, and the more significant the disturbance to the vertical distribution of wind speed. For example, at steep ridges The larger the height, the more obvious the airflow acceleration effect, resulting in an increase in the near-surface wind speed gradient (speed shear index).

[0066] slope aspect represents the actual orientation of the terrain slope, and determines the spatial relationship between the slope and the airflow direction; dominant wind direction represents the direction of the prevailing wind in the area, and is the basis for evaluating the consistency of the airflow and terrain interaction direction; is the cosine similarity measure of the slope aspect and the dominant wind direction angle The smaller the angle difference between the two, the closer to 1, the stronger the synergy between the airflow and the slope, and the more significant the terrain enhancement effect on wind speed; the larger the angle difference, tends to 0 or negative, the stronger the airflow disturbance by the terrain, which may inhibit wind speed growth or even cause local low-speed areas.

[0067] Based on the terrain modulation factor correction logarithmic law model, a wind speed profile model is obtained:

[0068] ;

[0069] represents the wind speed at height z after being modulated by the terrain modulation factor; f represents the terrain modulation factor;

[0070] Input the altitude into the wind speed profile model to obtain the wind speed at the current altitude.

[0071] Calculate the speed shear index WSE according to the wind speed at the current altitude and the altitude:

[0072] .

[0073] The wind speed profile model obtained by the above-mentioned terrain modulation factor correction logarithmic law model can significantly improve the accuracy of wind speed prediction in complex terrain areas and improve the physical consistency of wind speed prediction. The calculated speed shear index is one of the core indicators for quantifying terrain disturbance intensity, and is directly used for multi-source meteorological data fusion weight distribution. In areas with strong terrain disturbance, the weight of occultation observation data is increased. As a physical constraint condition, the simulation accuracy of the model for the near-surface wind speed cutting edge is adjusted, solving the problem of low wind speed prediction accuracy in non-flat areas in traditional methods.

[0074] On the basis of the above embodiments, the turbulence intensity is calculated according to the terrain data, including: obtaining the wind speed in a set time window according to the wind speed profile model to generate a time series wind speed; calculating the wind speed standard deviation and the average wind speed according to the time series wind speed; calculating the ratio of the wind speed standard deviation to the average wind speed to obtain the turbulence intensity. In the embodiment of the application, the time series wind speed containing the influence of terrain disturbance is obtained through the wind speed profile model, and then the turbulence intensity is calculated, which can more truly reflect the irregularity of airflow in a complex terrain area (such as a mountainous area or a canyon).

[0075] On the basis of the above embodiments, the weight is assigned based on the terrain disturbance factor, including: if the velocity shear index is greater than 0.3, the weight of the meteorological live data is increased; if the velocity shear index is less than or equal to 0.3, the weight of the grid forecast data is increased; if the turbulence intensity is greater than 15%, the weight of the occultation observation data is reduced; and if the turbulence intensity is less than or equal to 15%, the weight of the occultation observation data is increased.

[0076] In a specific application scenario, the adjustment logic of the velocity shear index is as follows:

[0077] The velocity shear index quantifies the nonlinear change rate of wind speed with height, and the greater the velocity shear index, the more significant the influence of terrain friction, slope and other factors on the near-surface wind speed, and the more intense the vertical shear. The meteorological live data directly measures the near-surface wind speed, but is easily disturbed by local obstacles (such as mountains and forests) in complex terrain, especially when the velocity shear index is large, the measured value is increased in error due to "near-surface turbulence mixing" or "terrain shielding". However, the meteorological live data can still reflect the "instantaneous fluctuations" of the local real wind speed in the extreme shear area (WSE>0.3), and has irreplaceability in capturing strong shear characteristics, so the weight of the meteorological live data is increased.

[0078] The grid forecast data is obtained based on a large-scale physical model, and has certain modeling capability for the vertical distribution of wind speed in flat or moderately complex terrain, but in the strong shear area of WSE>0.3, the model underestimates the shear intensity due to insufficient resolution or simplification of the parameterization scheme (such as not finely describing the terrain friction), resulting in prediction bias. In the weak shear area of WSE≤0.3, the grid forecast data is more reliable than the discrete live data due to its wide coverage and high calculation consistency, so the weight of the grid forecast data is increased.

[0079] Therefore, when WSE>0.3, the weight of the live data is increased to correct the underestimation of the model for extreme shear; and when WSE≤0.3, the weight of the grid forecast data is increased, at this time the modeling capability of the model for weak shear wind field is more stable, and the model has a wide coverage and can provide more continuous and consistent wind speed prediction results.

[0080] The adjustment logic of the turbulence intensity is as follows:

[0081] The turbulence intensity represents the ratio of the fluctuation amplitude of the wind speed to the average wind speed, and a turbulence intensity greater than 15% indicates that the airflow irregularity is significant; a turbulence intensity less than or equal to 15% indicates that the wind field is relatively stable.

[0082] The occultation observation data is atmospheric wind field information retrieved by satellite remote sensing, has the advantages of global coverage and being not affected by terrain obstruction, and can reflect the average characteristics of large-scale wind field, but in the high turbulence area where the turbulence intensity is greater than 15%, the retrieval result deviates greatly from the actual wind speed fluctuation because the local small-scale vortex cannot be resolved; in the stable area where the turbulence intensity is less than or equal to 15%, the occultation data can more accurately capture the characteristics of the large-scale stable wind field because it avoids the interference of ground obstacles, and its reliability is higher than that of the near-surface observation affected by local turbulence.

[0083] Therefore, when the turbulence intensity is greater than 15%, the local fluctuation in the high turbulence area will interfere with the retrieval accuracy of the occultation data, reduce the weight of the occultation observation data, and reduce its contribution to the fusion result; if the turbulence intensity is less than or equal to 15%, the occultation data can stably reflect the average state of the large-scale wind field and is not affected by terrain obstruction, thereby improving the weight of the occultation observation data.

[0084] In the embodiments of the present application, by dynamically adjusting the weight, the fusion process is adapted to the local conditions, relying on the local authenticity of the live weather data in the strong shear area, trusting the overall consistency of the model in the weak shear area, suppressing the local error of the occultation data in the high turbulence area, and taking advantage of the large-scale observation in the stable area. This adaptive mechanism significantly improves the accuracy and reliability of the fusion result.

[0085] The traditional numerical weather prediction model simulates the wind field based on a large-scale dynamic framework and a parameterization scheme. For the near-surface layer, the local wind speed vertical shear and airflow irregular disturbance caused by the surface roughness and terrain undulation cannot be accurately described, resulting in a deviation between the model output wind speed and wind direction and the actual observation. In order to solve this problem, as shown in FIG. 1, the comprehensive wind field data set is input into the numerical weather prediction model, and the wind field is iteratively optimized with the speed shear index and the turbulence intensity as physical constraints to obtain the spatiotemporal distribution of the wind field prediction data, including: Figure 3

[0086] Step one: embedding the speed shear index as a correction term of the mixing length in the boundary layer parameterization of the numerical weather prediction model, and introducing a turbulent energy source term driven by the turbulence intensity in the turbulence parameterization of the numerical weather prediction model to obtain a corrected numerical weather prediction model;

[0087] Step two: running the corrected numerical weather prediction model to generate the first round of wind speed and wind direction prediction values;

[0088] ​Step three: Compare the first round of wind speed and direction prediction values with the observed values. If the error exceeds the threshold, adjust the parameterization coefficients of the corrected numerical weather prediction model, and return to step one;

[0089] Step four: Until the iteration converges to obtain the wind field prediction data.

[0090] In specific application scenarios, the numerical weather prediction model NWP (Numerical Weather Prediction) is a scientific method and technical means based on the basic physical laws of atmospheric motion, which uses high-performance computers to solve partial differential equations describing atmospheric motion, and through the given initial atmospheric state, it quantitatively predicts the atmospheric state in the future. Key elements of the numerical weather prediction model include numerical model, initial condition, boundary condition, and physical process parameterization.

[0091] In the boundary layer parameterization scheme of the numerical weather prediction model, the mixing length is a key parameter that describes the vertical motion scale of air microclusters, directly affecting the shear of wind speed with height. The traditional numerical weather prediction model assumes that the mixing length is a stability function, without considering the local vertical variation of wind speed caused by terrain disturbance. This scheme embeds the velocity shear index as a correction term in the calculation of mixing length, dynamically adjusting its distribution with height and terrain. The embedding process can be: in the boundary layer parameterization module of the NWP model, locate the mixing length calculation subroutine, introduce the velocity shear index in the original formula, and pass the real-time velocity shear index of each spatial grid point through the input field.

[0092] In addition, the traditional numerical weather prediction model's turbulence parameterization scheme estimates the turbulent kinetic energy based on the stability function, without explicitly considering the local turbulence enhancement caused by terrain disturbance; this scheme introduces a turbulent energy term driven by turbulent intensity in the turbulence parameterization equation, directly relating the turbulent intensity and the generation rate of turbulent kinetic energy. The embedding process can be: in the turbulence parameterization module of the NWP model, modify the source term part of the turbulent kinetic energy equation, add a calculation subroutine, and input the real-time turbulent intensity of each spatial grid point.

[0093] Run the corrected NWP model to generate the first round of wind speed and direction prediction values; compare the prediction values with the observed values in the multi-source meteorological data set. If the error exceeds the threshold, adjust the parameterization coefficients of the model, and return to step one to run the corrected NWP model again until the iteration converges to obtain the wind field prediction data.

[0094] In the embodiment of the present application, the velocity shear exponent is embedded in the mixing length correction term to dynamically increase the mixing length in steep terrain areas, so that the NWP model can more accurately simulate the physical process of airflow acceleration along the slope. For example, in the windward slope area, the corrected mixing length allows the vertical movement of air parcels to be stronger, so that the output vertical profile of wind speed is closer to the measured logarithmic distribution, and the prediction error of the velocity shear exponent is reduced by 20%-30%. By introducing a turbulent energy source term driven by turbulent intensity, the intensity of airflow fluctuations is explicitly related to the generation rate of turbulent kinetic energy, which significantly increases the simulation value of turbulent kinetic energy in high turbulence areas, making the output wind speed fluctuation amplitude of the NWP model more matched with the measured value. For example, the prediction error of the turbulent intensity in the leeward slope area is reduced by 15%-25%, and the stability prediction of wind direction fluctuation is more consistent with the actual situation.

[0095] There are significant differences in wind field patterns at different spatial grid points in the wind farm area. The traditional method does not systematically classify these differences, which may lead to the placement of a wind measurement tower at an isolated point with a special wind field pattern, which cannot represent the overall characteristics of the region. To solve this problem, the embodiment of the present application also includes clustering wind field prediction data and selecting candidate locations for the wind measurement tower based on the clustering results, which includes: extracting statistical features and temporal variability features of each spatial grid point from the wind field prediction data to construct a wind field feature vector; wherein the statistical features include maximum wind speed, minimum wind speed, average wind speed, wind speed variability coefficient and wind direction standard deviation; the temporal variability features include periodicity features or trend change indicators; clustering the wind field feature vectors of all spatial grid points and dividing them into multiple wind field characteristic clusters according to the similarity of wind field characteristics.

[0096] By clustering analysis of the wind field feature vectors of all spatial grid points, the grid points are divided into multiple wind field characteristic clusters according to the similarity of wind field characteristics, and each cluster represents a type of wind field pattern with similar spatial distribution of wind speed and direction and temporal variability features. This grouping method reveals the spatial heterogeneity of the regional wind field, enabling site selection personnel to clearly identify the distribution range of different wind field patterns and avoid placing the wind measurement tower at isolated points with abnormal wind field patterns, significantly improving the scientificity of site selection. For example, in a mountain wind farm, clustering can distinguish between a mountain top stable wind field cluster and a mountain valley vortex wind field cluster, and the wind measurement tower is preferentially selected in the mountain top cluster to represent stable wind resources.

[0097] On the basis of the above embodiment, the spatial grid points that meet the preset conditions are selected as candidate locations for the wind measurement tower, which includes: from the multiple wind field characteristic clusters, extracting the average wind speed, wind speed variability coefficient, wind direction standard deviation and location information of each spatial grid point in its wind field characteristic cluster; selecting the spatial grid points that simultaneously satisfy the conditions of average wind speed not less than 5 m / s, wind direction standard deviation not greater than 30°, wind speed variability coefficient not greater than 0.2, and being in the central region of the wind field characteristic cluster as candidate locations for the wind measurement tower.

[0098] The threshold of the average wind speed not less than 5 m / s ensures that the screened wind tower position has the basic wind resource condition for developing wind power and focuses on the area with high wind energy potential; the limitation of the wind direction standard deviation not greater than 30° ensures that the wind direction of the area where the wind tower position is located has good stability, which helps to obtain reliable wind resource data and ensure the long-term stable operation of the wind power plant; the threshold of the wind speed variation coefficient not greater than 0.2 ensures that the wind speed of the area where the wind tower position is located has good stability, which provides guarantee for the long-term reliable operation of the wind power plant. The wind field characteristic cluster is obtained by clustering analysis on the wind field characteristic vectors of all spatial grid points, and each cluster represents a kind of wind field mode with similar wind speed and wind direction spatial distribution and time variability characteristics. The cluster center area is a typical representative area of this kind of wind field mode, and the wind field characteristics can reflect the common characteristics of the cluster, and have high representativeness and typicality.

[0099] The embodiment scheme of the present application can efficiently lock the optimal position with rich wind resources, stable wind conditions and regional representativeness from a large number of spatial grid points through multi-dimensional index screening, and significantly improves the accuracy of the wind tower site selection.

[0100] Embodiment two: refer to Figure 4 The embodiment of the present application discloses a wind power plant wind tower site selection system, which comprises:

[0101] The data acquisition module acquires the meteorological real-time data, grid forecast data and occultation observation data of a plurality of spatial grid points in the candidate area to form a multi-source meteorological data set;

[0102] The terrain processing module calculates the terrain disturbance factor based on the terrain data of the spatial grid points; the terrain disturbance factor includes the velocity shear index and the turbulence intensity;

[0103] The data fusion module values the multi-source meteorological data according to the terrain disturbance factor, and performs weighted fusion on the multi-source meteorological data according to the weight to obtain a comprehensive wind field data set;

[0104] The wind field prediction module inputs the comprehensive wind field data set into a numerical weather prediction model to perform wind field iterative optimization with the velocity shear index and the turbulence intensity as physical constraints to obtain the wind field prediction data with spatial and temporal distribution;

[0105] The site selection decision module screens the spatial grid points meeting the preset condition as the wind tower candidate position according to the wind field prediction data.

[0106] The embodiment of the present application and the embodiment one are based on the same inventive concept and have the same technical effect, which will not be described here.

[0107] In summary, the wind farm wind measurement tower site selection method and system provided by the present application significantly improves the wind farm data accuracy, multi-source data fusion reliability and wind farm prediction accuracy by quantifying the terrain disturbance factor and integrating multi-source data fusion and model optimization, accurately selects the wind measurement tower position with rich wind resources, stable wind farm mode and strong data representativeness, effectively solves the data deviation and site deviation problems of the traditional site selection method under complex terrain, simultaneously improves the meteorological accuracy and reliability of the wind measurement tower site selection, and is especially suitable for complex terrain areas such as mountains and coastal belts, and provides reliable data support for wind power project approval and power grid dispatching.

[0108] Those skilled in the art will understand that embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROMs, optical storage, etc.) containing computer-usable program code.

[0109] The present application is described with reference to flowcharts and / or block diagrams according to the methods, devices (systems), and computer program products of the embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of flows and / or blocks in the flowcharts and / or block diagrams can be implemented by computer program instructions. These computer program instructions can be provided to a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing apparatus to produce a machine, so that the instructions executed by the computer or other programmable data processing apparatus produce a device that implements the functions specified in the flowcharts and / or block diagrams. Figure 1 one or more flows and / or blocks Figure 1 an apparatus that carries out the functions specified in one or more flows and / or blocks.

[0110] These computer program instructions can also be stored in a computer-readable memory that can direct the computer or other programmable data processing apparatus to work in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured product including instruction apparatus, which implements the functions specified in the flowcharts and / or block diagrams. Figure 1 one or more flows and / or blocks Figure 1 an apparatus that carries out the functions specified in one or more flows and / or blocks.

[0111] These computer program instructions can also be loaded into a computer or other programmable data processing apparatus, so that a series of operation steps are performed on the computer or other programmable data processing apparatus to produce a computer-implemented process, so that the instructions executed on the computer or other programmable data processing apparatus provide a device for implementing the functions specified in the flowcharts and / or block diagrams. Figure 1 one or more flows and / or blocks Figure 1steps of the functions specified in the one or more blocks.

[0112] Obviously, the above-mentioned embodiments are only examples for clearly illustrating the present application, and are not intended to limit the present application. Based on the above-mentioned embodiments, other different forms of changes or variations can be made by those skilled in the art. Here, it is not necessary or possible to enumerate all the embodiments. The obvious changes or variations derived therefrom are still within the protection scope of the present application.

Claims

1. A method for selecting the site of a wind measurement tower in a wind farm, characterized in that, include: The candidate region is divided into multiple spatial grid points, and meteorological real-time data, grid forecast data and occultation observation data of the multiple spatial grid points are obtained to form a multi-source meteorological dataset. The terrain data of each spatial grid point is acquired, and the terrain disturbance factor of the spatial grid point is calculated based on the terrain data; wherein, the terrain disturbance factor includes the velocity shear index and the turbulence intensity; Based on the topographic disturbance factor, a weight is assigned, and multi-source meteorological data are weighted and fused according to the weight to obtain a comprehensive wind field dataset. The comprehensive wind field dataset is input into the numerical weather prediction model, and the wind field is iteratively optimized using the velocity shear index and turbulence intensity as physical constraints to obtain spatiotemporal wind field prediction data. Based on the wind field prediction data, the spatial grid points that meet the preset conditions are selected as candidate locations for the wind measurement tower; The terrain data includes altitude, slope angle, and aspect angle. Calculating the velocity shear index based on the terrain data includes: constructing a logarithmic law model; the logarithmic law model characterizes the logarithmic relationship between wind speed and altitude in the near-surface boundary layer; determining a terrain modulation factor based on the slope angle and aspect angle; modifying the logarithmic law model based on the terrain modulation factor to obtain a wind speed profile model; inputting the altitude into the wind speed profile model to obtain the wind speed at the current altitude; and calculating the velocity shear index based on the wind speed at the current altitude and the altitude. Calculating turbulence intensity based on the terrain data includes: obtaining wind speed within a set time window based on the wind speed profile model, generating a time series wind speed; calculating the standard deviation and average wind speed based on the time series wind speed; and calculating the ratio of the standard deviation to the average wind speed to obtain the turbulence intensity.

2. The method for selecting the location of a wind farm meteorological tower according to claim 1, characterized in that, Determining the terrain modulation factor based on the slope angle and aspect angle includes: ； f represents the terrain modulation factor; k represents the empirical coefficient, with a value ranging from 0.15 to 0.25; Indicates the slope angle; Indicates the slope angle; This indicates the prevailing wind direction angle for the candidate area.

3. The method for selecting a wind farm anemometer tower according to any one of claims 1-2, characterized in that, The weights are assigned based on the terrain disturbance factor, including: If the velocity shear index is greater than 0.3, the weight of the meteorological data will be increased. If the velocity shear index is less than or equal to 0.3, then the weight of the grid forecast data is increased; If the turbulence intensity is greater than 15%, the weight of the occultation observation data will be reduced. If the turbulence intensity is less than or equal to 15%, the weight of the occultation observation data is increased.

4. The method for selecting the location of a wind farm meteorological tower according to claim 1, characterized in that, The comprehensive wind field dataset is input into a numerical weather prediction model, and the wind field is iteratively optimized using the velocity shear index and turbulence intensity as physical constraints to obtain spatiotemporally distributed wind field prediction data, including: Step 1: Embed the velocity shear index as a correction term for the mixing length in the boundary layer parameterization of the numerical weather prediction model to obtain the corrected numerical weather prediction model. Step 2: Run the modified numerical weather prediction model to generate the first round of wind speed and direction predictions; Step 3: Compare the predicted wind speed and direction values ​​with the observed values ​​in the first round. If the error exceeds the threshold, adjust the parameterization coefficients of the corrected numerical weather prediction model and return to Step 1. Step 4: Continue iteratively until convergence is achieved to obtain the wind field prediction data.

5. The method for selecting the location of a wind farm meteorological tower according to claim 4, characterized in that, Step one further includes: introducing the turbulence intensity-driven turbulent energy term into the turbulence parameterization of the numerical weather prediction model.

6. The method for selecting the location of a wind farm meteorological tower according to claim 1, characterized in that, It also includes clustering the wind field prediction data and filtering candidate locations for the meteorological towers based on the clustering results, which includes: Statistical features and temporal variability features of each spatial grid point are extracted from the wind field prediction data to construct a wind field feature vector; wherein, the statistical features include maximum wind speed, minimum wind speed, average wind speed, wind speed coefficient of variation, and wind direction standard deviation; the temporal variability features include periodic features or trend change indicators; The wind field feature vectors of all the spatial grid points are clustered and divided into multiple wind field feature clusters based on the similarity of wind field characteristics.

7. The method for selecting a wind farm meteorological tower according to claim 6, characterized in that, Selecting spatial grid points that meet preset conditions as candidate locations for wind measurement towers includes: From the multiple wind field characteristic clusters, extract the average wind speed, wind speed variation coefficient, wind direction standard deviation, and the location information of the spatial grid point in its respective wind field characteristic cluster for each spatial grid point; Spatial grid points that simultaneously meet the following criteria—average wind speed not less than 5 m / s, wind direction standard deviation not greater than 30°, wind speed variation coefficient not greater than 0.2, and located in the central region of their respective wind field characteristic cluster—are selected as candidate locations for the wind measurement tower.

8. A wind farm anemometer tower site selection system, characterized in that, include: The data acquisition module acquires real-time meteorological data, grid forecast data, and occultation observation data from multiple spatial grid points within the candidate area, forming a multi-source meteorological dataset; The terrain processing module calculates the terrain disturbance factor based on the terrain data of the spatial grid points; the terrain disturbance factor includes the velocity shear index and turbulence intensity; The data fusion module assigns weights to multi-source meteorological data according to the terrain disturbance factor, and performs weighted fusion of the multi-source meteorological data according to the weights to obtain a comprehensive wind field dataset. The wind field prediction module inputs the comprehensive wind field dataset into the numerical weather prediction model, and performs iterative optimization of the wind field using the velocity shear index and turbulence intensity as physical constraints to obtain spatiotemporal wind field prediction data. The site selection decision module filters the spatial grid points that meet preset conditions as candidate locations for the wind measurement tower based on the wind field prediction data. The terrain data includes altitude, slope angle, and aspect angle. Calculating the velocity shear index based on the terrain data includes: constructing a logarithmic law model; the logarithmic law model characterizes the logarithmic relationship between wind speed and altitude in the near-surface boundary layer; determining a terrain modulation factor based on the slope angle and aspect angle; modifying the logarithmic law model based on the terrain modulation factor to obtain a wind speed profile model; inputting the altitude into the wind speed profile model to obtain the wind speed at the current altitude; and calculating the velocity shear index based on the wind speed at the current altitude and the altitude. Calculating turbulence intensity based on the terrain data includes: obtaining wind speed within a set time window based on the wind speed profile model, generating a time series wind speed; calculating the standard deviation and average wind speed based on the time series wind speed; and calculating the ratio of the standard deviation to the average wind speed to obtain the turbulence intensity.

Images