Networking methods and dynamic parameter fusion methods for wind profiler radar and wind lidar

Abstract

This application provides a networking method for wind profiler radar and wind-measuring lidar, as well as a dynamic parameter fusion method, belonging to the field of digital signal processing. The networking method of this application analyzes and evaluates the influence characteristics of the geometric parameters of a triangle on the retrieved key dynamic parameters, the influence characteristics of wind field element observation errors on the retrieved key dynamic parameters, and the influence characteristics of different weather and geographical conditions on the retrieved key dynamic parameters. It employs an area-weighted averaging method to form an optimal triangle networking scheme and calculates the key dynamic parameters, fully utilizing the advantages of both radars to improve the accuracy of retrieved atmospheric dynamic parameters. The dynamic parameter fusion method, based on the retrieved key dynamic parameters, further fuses reanalysis data or numerical model initial fields to obtain multi-source fused data, thereby further improving the accuracy of the data.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of digital signal processing, and more particularly, relates to a wind profile radar and wind lidar networking method and a dynamic parameter fusion method. BACKGROUND

[0002] The accurate detection of atmospheric wind field is the optimal support for strong convective weather monitoring and early warning, mesoscale meteorological research, and applications in aviation, wind energy, and other industries. Currently, wind profile radar and laser wind lidar are the two optimal devices for wind field remote sensing detection. They have different technical principles and form unique detection capabilities, but also have significant limitations in single application.

[0003] Wind Profiler Radar (WPR) takes the inhomogeneous structure of refractive index caused by atmospheric turbulence as the detection target, and retrieves wind field information by receiving electromagnetic wave scattering echoes. Its optimal advantage is that the detection height covers a wide range, which can effectively capture the characteristics of atmospheric wind field at high altitude, and can maintain stable observation under clear and cloudy weather conditions, which is an important supplement to conventional sounding observation. However, the temporal and spatial resolution of this device is relatively low, and it is difficult to capture small-scale and short-time wind field changes, and the measurement accuracy also cannot meet the needs of fine analysis.

[0004] Laser wind lidar is based on the laser Doppler effect, and retrieves wind direction and speed by tracking the motion trajectory of aerosol particles in the atmosphere. As a new type of detection device, its biggest advantage is that it has high temporal and spatial resolution, can accurately capture low-altitude small-scale wind field changes, and has better wind field detection accuracy than wind profile radar, which is outstanding in low-altitude wind shear monitoring and micro-scale wind field analysis. However, the detection height of this device is limited and cannot cover the mid-high altitude wind field; at the same time, it is significantly affected by weather and environment, and its detection performance will be greatly reduced in rainy and foggy weather conditions, and the observation data efficiency may also be reduced in areas with scarce aerosol particles.

[0005] It is this "high-altitude detection and low-altitude detection fine complementation" feature that makes the joint networking of wind profile radar and laser wind lidar an important direction to break through the observation bottleneck of single device. Through joint networking, it can achieve full coverage of high and low altitude wind field, generate more reliable three-dimensional wind field and key dynamic parameters (horizontal divergence, relative vorticity and vertical velocity, same below) products by combining the data advantages of the two, which not only can improve the recognition ability of strong convective weather trigger signal and provide stronger support for early warning work, but also can provide more abundant basic data for mesoscale meteorological dynamics research.

[0006] Taking the Beijing-Tianjin-Hebei region as an example, the region has built a device foundation of 27 wind profile radars and 108 laser wind radars. Through joint networking, not only can the limitations of single equipment be broken through to obtain higher temporal and spatial resolution three-dimensional horizontal divergence, vorticity and vertical velocity field, but also the dynamic field characteristics before convective triggering can be more accurately analyzed, thereby supporting the construction of a quantitative discrimination model of strong convective weather triggering power signals and providing early signal identification capability for strong convective weather monitoring and early warning.

[0007] However, the current joint networking practice faces key bottlenecks: both types of radars are targeted at meeting the detection needs of single equipment in the early stage of construction, without considering the optimal spatial layout principle of joint networking, resulting in that the site location and overall layout cannot meet the optimal requirements of joint detection. If direct conventional methods are used for networking, the complementary effects of the two types of radars cannot be fully brought into play, and the best detection accuracy and high temporal and spatial resolution requirements cannot be met. Therefore, how to optimize the existing radar networking method to improve the joint detection effect has become an optimal problem to be solved. SUMMARY

[0008] Therefore, the embodiments of the present application provide a wind profile radar and wind laser radar networking method and a dynamic parameter fusion method to solve the technical problems of suboptimal wind profile radar and wind laser radar networking scheme and low accuracy of key dynamic parameters obtained by inversion in the prior art.

[0009] To achieve the above-mentioned purpose, the technical scheme adopted by the present application is:

[0010] On the one hand, a wind profile radar and wind laser radar networking method is provided, comprising:

[0011] S100, correcting historical wind profile radar wind field data of a target region according to historical in-situ measured wind field data to obtain historical wind profile radar corrected wind field data;

[0012] S200, correcting historical wind laser radar inversion wind field data of the target region according to the historical in-situ measured wind field data to obtain historical wind laser radar corrected wind field data;

[0013] S300, fusing the historical wind profile radar corrected wind field data and the historical wind laser radar corrected wind field data to obtain historical fusion wind field data;

[0014] S400, dividing the wind profile radar and the wind laser radar of the target region into a plurality of triangles to form an initial triangle set S0; and dividing the triangles in the initial triangle set S0 into multiple types according to the geometric parameters of the triangles;

[0015] S500, for each type of triangle in the initial triangle set S0, according to the historical fusion wind field data, a triangle networking inversion algorithm is used to calculate the key dynamic parameters, a sensitivity test Test 1 is designed, the characteristics of the relative error of the key dynamic parameters changing with the geometric parameters of the triangle are analyzed, and a triangle set S1 considering the networking characteristics is selected;

[0016] S600, adding observation error disturbance to the wind field elements in the historical fusion wind field data to form test wind field data; for the triangle set S1 considering the networking characteristics, according to the test wind field data, a key dynamic parameter networking inversion sensitivity test Test 2 is designed, in which the wind field at the vertex of the triangle changes, and the geometric parameters of the triangle remain unchanged, the characteristics of the relative error of the key dynamic parameters changing with the observation error of the wind field elements are analyzed, and a triangle set S2 considering the networking characteristics and the wind field characteristics is selected;

[0017] S700, under different weather and geographical conditions, for the triangle set S2 considering the networking characteristics and the wind field characteristics, according to the historical fusion wind field data, a key dynamic parameter networking inversion sensitivity test Test 3 is designed, in which the wind field at the vertex of the triangle remains unchanged, the geometric parameters of the triangle remain unchanged, and the weather conditions and geographical conditions change, the characteristics of the relative error of the key dynamic parameters changing with the weather conditions and geographical conditions are analyzed, and a triangle set S3 considering the networking characteristics, the wind field characteristics, different weather conditions and geographical conditions is selected;

[0018] S800, in the triangle set S3 considering the networking characteristics, the wind field characteristics, different weather conditions and geographical conditions, for the four triangles composed of four adjacent wind profile radars or wind measuring laser radars, the adjacent four wind profile radars or wind measuring laser radars are encrypted at the diagonal and the midpoint of each side to construct eight triangles, forming an encrypted triangle networking set S4, which is used as the optimal triangle networking scheme;

[0019] S900, using an area weighting method, the key dynamic parameters of the encrypted triangle networking set S4 are calculated based on the triangle networking inversion algorithm.

[0020] In some embodiments, the step S100 comprises:

[0021] S101, the historical in-situ measured wind field data and the historical wind profile radar inversion wind field data are matched in space and time;

[0022] S102, a cubic spline relationship model of the historical in-situ measured wind field data and the historical wind profile radar inversion wind field data is established, and the historical wind profile radar inversion wind field data is corrected according to the cubic spline relationship model to obtain historical wind profile radar corrected wind field data;

[0023] The step S200 includes:

[0024] S201, spatially and temporally matching the historical in-situ measured wind field data and the historical wind lidar retrieved wind field data;

[0025] S202, establishing a cubic spline relationship model of the historical in-situ measured wind field data and the historical wind lidar retrieved wind field data, and correcting the historical wind lidar retrieved wind field data according to the cubic spline relationship model to obtain the historical wind lidar corrected wind field data.

[0026] In some embodiments, the step S300 includes:

[0027] Fusing the historical wind profiler corrected wind field data and the historical wind lidar corrected wind field data through a vertical layer feature matching fusion algorithm based on density clustering;

[0028]

[0029]

[0030]

[0031] wherein, is the historical fused wind field data, is the historical wind profiler corrected wind field data, is the historical wind lidar corrected wind field data, is the spatiotemporally encrypted historical wind profiler corrected wind field data, is the variance of the historical wind profiler corrected wind field data, is the variance of the historical wind lidar corrected wind field data, is a weight coefficient for fusion of the wind profiler, is a weight coefficient for fusion of the wind lidar.

[0032] In some embodiments, the step S400 includes:

[0033] S401, combining all wind profilers and all wind lidars in the target area into all possible triangles through a spatial combination algorithm to form an initial triangle set S0;

[0034] S402, dividing the triangles in the initial triangle set S0 into multiple types from three dimensions of the internal angle, the side length and / or the area of the triangle.

[0035] In some embodiments, the step S500 includes:

[0036] For each type of triangle in the initial triangle set S0, according to the multi-year average value of the historical fusion wind field, based on the control variable idea, a key dynamic parameter networking inversion sensitivity test Test 1 of the triangle vertex wind field invariable and the triangle geometric parameter changing is designed, the characteristics of the relative error of the key dynamic parameters with the change of the triangle geometric parameters are analyzed, the triangles with too large inversion error are removed, and a triangle set S1 considering the networking characteristics is screened.

[0037] In some embodiments, the step S600 comprises:

[0038] S601, adding observation error disturbance to the wind field elements of different height layers in the historical fusion wind field data to form test wind field data;

[0039] S602, adding observation error disturbance to the wind field elements in the historical fusion wind field data to form test wind field data; based on the control variable idea, for the triangle set S1 considering the networking characteristics, a key dynamic parameter networking inversion sensitivity test Test 2 of the triangle vertex wind field changing and the triangle geometric parameter invariable is designed, the characteristics of the relative error of the key dynamic parameters with the change of the wind field element observation error are analyzed, the triangles with too large inversion error accumulation are removed, and a triangle set S2 considering the networking characteristics and the wind field characteristics is screened.

[0040] In some embodiments, the step S700 comprises:

[0041] Classifying different weathers of the target area, combining the longitude, latitude and altitude geographical conditions, based on the control variable idea, for the triangle set S2 considering the networking characteristics and the wind field characteristics, a key dynamic parameter networking inversion sensitivity test Test 3 of the triangle vertex wind field invariable, the triangle geometric parameter invariable, the different weather conditions and the geographical conditions changing is designed, the characteristics of the relative error of the key dynamic parameters with the change of the weather type and the geographical conditions are analyzed, the triangles with too large inversion error accumulation are removed, and a triangle set S3 considering the networking characteristics, the wind field characteristics, the different weather conditions and the geographical conditions characteristics is screened.

[0042] In some embodiments, the step S900 comprises:

[0043] S901, quality control is performed on the real-time wind profile radar wind field detection data of the target area to obtain real-time wind profile radar inversion wind field data; quality control is performed on the real-time wind detection laser radar wind field detection data of the target area to obtain real-time wind detection laser radar inversion wind field data;

[0044] S902, fuse the real-time wind profile radar inversion wind field data and the real-time wind lidar inversion wind field data to obtain real-time fusion wind field data;

[0045] S903, for each triangle in the optimal triangular network scheme, calculate the horizontal components of the wind speed of each vertex in the u direction and the v direction on different height layers according to the real-time fusion wind field data, , wherein the u direction and the v direction are horizontal directions and perpendicular to each other, is the vertex serial number, ;

[0046] S904, for each triangle in the optimal triangular network scheme, calculate the horizontal components of the edge length of the triangle in the u direction and the v direction according to the longitude and latitude of each vertex, , and the earth radius R; , ;

[0047] S905a, according to the horizontal components of the wind speed of each vertex on different height layers of each triangle, , and the horizontal components of the edge length of each edge, , , the horizontal divergence of each triangle on different height layers in the optimal triangular network scheme is calculated ,

[0048] ;

[0049] S905b, according to the horizontal components of the wind speed of each vertex on different height layers of each triangle, , and the horizontal components of the edge length of each edge, , , the vorticity of each triangle on different height layers in the optimal triangular network scheme is calculated ,

[0050] ;

[0051] S905c, the horizontal divergence of each triangle on different height layers in the optimal triangular network scheme is integrated from the ground upwards to obtain the vertical velocity of each triangle on different height layers , ,

[0052]

[0053] wherein, is the target height layer, is a height layer variable.

[0054] The wind profile radar and the wind lidar networking method provided by the embodiments of the present application have the beneficial effects that, compared with the prior art, the wind profile radar and the wind lidar networking method of the embodiments of the present application correct the historical wind profile radar wind field data by historical in-situ measured wind field data, correct the historical wind lidar inversion wind field data by historical in-situ measured wind field data, and then fuse to obtain historical fusion wind field data, so that the accuracy of the historical fusion wind field data is higher.

[0055] According to the historical fusion wind field data, the key dynamic parameters are inverted, the influence characteristics of the geometric parameters of the triangle on the inversion of the key dynamic parameters, the influence characteristics of the observation error of the wind field elements on the inversion of the key dynamic parameters of the triangle with different geometric parameters, and the influence characteristics of different weather and geographical conditions on the inversion of the key dynamic parameters of the triangle with different geometric parameters are analyzed and evaluated, and finally the three influence characteristics are comprehensively selected from the initial triangle set S0 to obtain the encrypted triangle networking set S4, and then the optimal triangle networking scheme is formed. Finally, the key dynamic parameters are calculated by using the triangle networking inversion algorithm through the optimal triangle networking scheme, which can fully utilize the advantages of the two kinds of radars, greatly improve the accuracy of the inversion of the atmospheric dynamic parameters, and achieve better joint networking effect.

[0056] Another technical solution adopted by the present application is to provide a dynamic parameter fusion method, comprising:

[0057] T100, taking the key dynamic parameters obtained by the wind profile radar and the wind lidar networking method according to any one of the above as the inversion key dynamic parameters;

[0058] T200, obtaining reanalysis data or numerical model initial field of a target area and a plurality of corresponding grids;

[0059] T300, determining one or more overlapping triangles overlapping with the same grid in the optimal triangle networking scheme, and calculating the proportion of the overlapping area of each overlapping triangle to the total area of the corresponding grid;

[0060]

[0061] wherein, is the proportion of the overlapping area of the i-th overlapping triangle to the total area of the grid, is the serial number of the overlapping triangle, is the overlapping area of the i-th overlapping triangle and the grid, is the total area of the grid;

[0062] ​​T400 fuses reanalysis data or numerical model initial fields with inverted key dynamic parameters to obtain multi-source fused data. ,

[0063]

[0064] in, For the first The network inversion feature variables obtained from the overlapping triangles, This refers to the reanalysis data or initial field data of the numerical model corresponding to the grid.

[0065] In some embodiments, in step T100, the key dynamic parameters include: horizontal divergence, vorticity, and / or vertical velocity; the key dynamic parameters are used as the inversion key dynamic parameters after undergoing time-consistent average and median tests.

[0066] The beneficial effects of the dynamic parameter fusion method provided in this application embodiment are as follows: Compared with the prior art, the dynamic parameter fusion method in this application embodiment uses the key dynamic parameters obtained by the networking method of wind profiler radar and wind measuring lidar as the inversion key dynamic parameters, makes full use of the high-precision key dynamic parameters obtained after the encrypted networking of wind profiler radar and wind measuring lidar, and on this basis, fuses reanalysis data or numerical model initial fields to obtain multi-source fusion data, thereby further improving the data accuracy. Attached Figure Description

[0067] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0068] Figure 1 A flowchart illustrating the networking method of wind profiler radar and wind-measuring lidar provided in the embodiments of this application;

[0069] Figure 2 A schematic diagram showing the encryption of four adjacent wind profiler radars or wind-measuring lidars forming a triangle;

[0070] Figure 3 A schematic diagram showing the distribution of wind profiler radar, wind lidar, and sounding stations in the target area;

[0071] Figure 4 This is a schematic diagram of the initial triangular network formation for wind profiler radars in the target area;

[0072] Figure 5A schematic diagram of a triangle network for a wind lidar in a target area;

[0073] Figure 6 A schematic diagram of an encrypted joint network for a wind profile radar and a wind lidar in a target area;

[0074] Figure 7 A flowchart of the power parameter fusion method provided by the embodiments of the present application. DETAILED DESCRIPTION

[0075] In order to make the technical problems, technical solutions and beneficial effects of the present application clearer, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and do not limit the present application.

[0076] In addition, in the description of the present application, the meaning of "multiple" or "several" is two or more than two, unless otherwise explicitly and specifically limited.

[0077] Please refer to Figures 1 to 6 , the wind profile radar and wind lidar networking method provided by the embodiments of the present application will be described. A wind profile radar and wind lidar networking method, comprising:

[0078] S100, correcting the historical wind profile radar wind field data of a target area according to the historical in-situ measured wind field data, to obtain historical wind profile radar corrected wind field data;

[0079] S200, correcting the historical wind lidar inversion wind field data of a target area according to the historical in-situ measured wind field data, to obtain historical wind lidar corrected wind field data;

[0080] S300, fusing the historical wind profile radar corrected wind field data and the historical wind lidar corrected wind field data, to obtain historical fusion wind field data;

[0081] S400, dividing the wind profile radar and wind lidar of the target area into a plurality of triangles to form an initial triangle set S0; according to the geometric parameters of the triangles, the triangles in the initial triangle set S0 are divided into multiple types;

[0082] S500, for each type of triangle in the initial triangle set S0, according to the historical fusion wind field data, a triangle networking inversion algorithm is used to calculate the key dynamic parameters, a sensitivity test Test 1 is designed, the characteristics of the relative error of the key dynamic parameters with the change of the triangle geometric parameters are analyzed, and a triangle set S1 considering the networking characteristics is selected;

[0083] S600, adding observation error disturbance to the wind field elements in the historical fusion wind field data to form test wind field data; for the triangle set S1 considering the networking characteristics, according to the test wind field data, designing the key dynamic parameter networking inversion sensitivity test Test 2 of the triangle vertex wind field change and the unchanged triangle geometric parameters, analyzing the characteristics of the relative error of the key dynamic parameters with the change of the wind field element observation error, and screening to obtain the triangle set S2 considering the networking characteristics and the wind field characteristics;

[0084] S700, under different weather and geographical conditions, for the triangle set S2 considering the networking characteristics and the wind field characteristics, according to the historical fusion wind field data, designing the key dynamic parameter networking inversion sensitivity test Test 3 of the unchanged triangle vertex wind field, the unchanged triangle geometric parameters, the changed weather conditions and the changed geographical conditions, analyzing the characteristics of the relative error of the key dynamic parameters with the change of the weather conditions and the geographical conditions, and screening to obtain the triangle set S3 considering the networking characteristics, the wind field characteristics, the different weather conditions and the geographical conditions characteristics;

[0085] S800, in the triangle set S3 considering the networking characteristics, the wind field characteristics, the different weather conditions and the geographical conditions characteristics, for the four triangles composed of the adjacent four wind profile radars or wind measuring laser radars, encrypting at the opposite angles and the midpoints of the sides of the adjacent four wind profile radars or wind measuring laser radars to construct eight triangles, forming the encrypted triangle networking set S4, and taking it as the optimal triangle networking scheme;

[0086] S900, using the area weighting method, the key dynamic parameters of the encrypted triangle networking set S4 are calculated based on the triangle networking inversion algorithm.

[0087] Compared with the prior art, the wind profile radar and wind measuring laser radar networking method of the embodiment of the application corrects the historical wind profile radar inversion wind field data through historical in-situ measured wind field (such as in-situ wind field of sounding, tower, etc.) data, corrects the historical wind measuring laser radar inversion wind field data through historical in-situ measured wind field data, and then obtains the historical fusion wind field data of the wind profile radar and the wind measuring laser radar, so that the accuracy of the historical fusion wind field data is higher.

[0088] According to the historical fusion wind field data, the key dynamic parameters are inversed, the influence characteristics of the geometric parameters of the triangle on the inversion of the key dynamic parameters are analyzed and evaluated, the influence characteristics of the observation error of the wind field elements on the inversion of the key dynamic parameters of the triangle with different geometric parameters are analyzed and evaluated, and the influence characteristics of different weather and geographical conditions on the inversion of the key dynamic parameters of the triangle with different geometric parameters are analyzed and evaluated, and then the optimal triangle networking set is formed through the area weighted average method of the encrypted sites; finally, the key dynamic parameters are calculated through the triangle networking inversion algorithm; the method can fully utilize the advantages of the two radars, improve the accuracy of the inversion of the key dynamic parameters, and achieve better joint networking effect.

[0089] The specific implementation details of the above embodiments are further described below.

[0090] The historical in-situ measured wind field data used in steps S100 and S200 can be in-situ sounding, tower observation data, or other observation data with higher accuracy than wind profile radar and wind lidar observation. The historical in-situ wind field data, historical wind profile radar inversed wind field data, historical wind lidar inversed wind field data, and other wind field data generally include horizontal wind speed, wind direction, and vertical speed three types of data without special instructions.

[0091] The historical wind profile radar wind field data in step S100 can be data processed by quality control. The quality control processing method can be any known quality control method. As an example, after obtaining the original power spectrum data of the wind profile radar, the noise threshold is determined by combining the Kalman filter algorithm and improving the objective analysis method to calculate the noise level of each distance bin. According to the distribution characteristics of the power spectrum density diagram, the non-turbulent echo characteristics such as ground clutter peak area, cloud peak area, and precipitation peak area are judged, the chain test method is used to suppress interference signals and ground clutter, and the target echo is extracted. Based on the spectrum moment calculation, the quality-controlled RAD radial data and the quality-controlled ROBS data are obtained.

[0092] The historical wind lidar inversed wind field data in step S200 can be data processed by quality control. The quality control processing method can be any known quality control method. As an example, the original.BIN data of the wind lidar is quality controlled to remove ground clutter and non-meteorological echoes, and the quality-controlled RAD radial data and the quality-controlled ROBS data are obtained.

[0093] The fusion method of the historical wind profile radar corrected wind field data and the historical wind lidar corrected wind field data in step S300 can adopt known time and space fusion methods, or the fusion methods adopted in other embodiments of the present application.

[0094] In step S400, for the division of the wind profile radar and the wind lidar of the target area, as many triangles as possible should be divided to form the initial triangle set S0, so as to increase the number of triangles in the initial triangle set S0, improve the accuracy of the evaluation result, and each wind profile radar and wind lidar can be repeatedly divided into different triangles. The geometric parameters of the triangles for classification can be any one, two or three of the internal angle, the side length and the area. As an example, in the internal angle dimension, it can be classified into categories of every 5°, in the side length dimension, it can be classified into categories of every 1 km, and in the area dimension, it can be classified into categories of every 1 km2, and any two or three of them can be combined for classification.

[0095] In step S500, the key dynamic parameters are respectively inverted for each type of triangle, so that the key dynamic parameters corresponding to different types of triangles can be obtained, the characteristics of the relative error of the key dynamic parameters changing with the geometric parameters of the triangle are analyzed, the triangles with too large inversion error are removed, and the triangle set S1 considering the networking characteristics is screened. The key dynamic parameters include horizontal divergence, relative vorticity and vertical velocity. As an example, when the geometric parameters used for classification include the internal angle, the characteristics of the relative error of the key dynamic parameters changing with the size of the internal angle of the triangle can be analyzed, so as to help select the size of the internal angle of the triangle; when the geometric parameters used for classification include the side length, the characteristics of the relative error of the key dynamic parameters changing with the size of the side length of the triangle can be analyzed, so as to help select the size of the side length of the triangle; when the geometric parameters used for classification include the area, the characteristics of the relative error of the key dynamic parameters changing with the size of the area of the triangle can be analyzed, so as to help select the size of the area of the triangle.

[0096] In step S600, by adding observation error disturbance to the wind field elements in the historical fusion wind field data, the characteristics of the relative error of the key dynamic parameters changing with the observation error of the wind field elements can be analyzed, the triangles with too large inversion error accumulation are removed, and the triangle set S2 considering the networking characteristics and the wind field characteristics is screened. Taking the most common wind speed and wind direction as an example, the error disturbance is added to the wind speed, and then the characteristics of the relative error of the key dynamic parameters changing with the observation error of the wind speed can be analyzed; similarly, the error disturbance is added to the wind direction, and then the characteristics of the relative error of the key dynamic parameters changing with the observation error of the wind direction can be analyzed. So as to be used for screening the triangle set S3 considering the networking characteristics, the wind field characteristics, different weather conditions and geographical conditions in subsequent step S700.

[0097] In step S700, the weather conditions usually include clear sky, precipitation, gale, etc., and the geographical conditions usually include common conditions such as longitude, latitude, altitude, etc. By respectively inverting the key dynamic parameters of different types of triangles under different weather and geographical conditions, the characteristics of the relative error of the key dynamic parameters changing with the weather conditions and the geographical conditions are analyzed, and the triangles with accumulated inversion error that is too large are removed, so as to screen the triangle set S3 considering the networking characteristics, the wind field characteristics, the different weather conditions and the geographical conditions characteristics, which is used to help the optimal triangle networking set S4 in the subsequent step S900. Taking the Beijing-Tianjin-Hebei region as an example, different weathers in the Beijing-Tianjin-Hebei region are classified, combined with the geographical conditions such as longitude, latitude, altitude, etc., based on the control variable idea, for the triangle set S2 considering the networking characteristics and the wind field characteristics, a key dynamic parameter networking inversion sensitivity test Test3 of the triangle vertex wind field being unchanged, the triangle geometric parameters being unchanged, and the different weather conditions and the geographical conditions changing is designed, the characteristics of the relative error of the key dynamic parameters changing with the weather conditions and the geographical conditions are analyzed, and the triangles with accumulated inversion error that is too large are removed, so as to screen the triangle set S3 considering the networking characteristics, the wind field characteristics, the different weather conditions and the geographical conditions characteristics.

[0098] In step S800, in the triangle set S3 considering the networking characteristics, the wind field characteristics, the different weather conditions and the geographical conditions characteristics, for the 4 triangle networks composed of 4 adjacent wind profile radars or wind measuring laser radars satisfying the conditions, the 8 triangles are constructed by encrypting at the diagonal and the midpoint of each side of the 4 adjacent wind profile radars or wind measuring laser radars, to form the encrypted optimal triangle networking set S4. Referring to FIG. 4, in the figure, the 4 vertices are 4 adjacent radars, T1_old, T2_old, T3_old, T4_old are the 4 triangles before encryption, and T1, T2, T3, T4, T5, T6, T7, T8 are the 8 triangles after encryption. Figure 2

[0099] In step S900, the measured wind profile radar and wind measuring laser radar data are obtained, and the area weighted method is used to obtain the key dynamic parameter data by inversion calculation through the optimal triangle networking set.

[0100] It should be noted that the method of inverting the key dynamic parameters in steps S500, S600 and S700 can be the same as that in step S900, and specific embodiments are described in other embodiments.

[0101] As a specific embodiment of the wind profile radar and wind measuring laser radar networking method provided in the present application, step S100 includes:

[0102] S101, spatially and temporally matching the historical in-situ measured wind field data and the historical wind profile radar inversion wind field data; ​

[0103] S102, a cubic spline relationship model of the historical in-situ measured wind field data and the historical wind profile radar inversion wind field data is established, and the historical wind profile radar inversion wind field data is corrected according to the cubic spline relationship model to obtain historical wind profile radar corrected wind field data;

[0104] Step S200 includes:

[0105] S201, the historical in-situ measured wind field data and the historical wind lidar inversion wind field data are spatially and temporally matched;

[0106] S202, a cubic spline relationship model of the historical in-situ measured wind field data and the historical wind lidar inversion wind field data is established, and the historical wind lidar inversion wind field data is corrected according to the cubic spline relationship model to obtain historical wind lidar corrected wind field data.

[0107] In steps S101 and S201, the historical in-situ measured wind field data can adopt L-band sounding data or tower wind field. The L-band sounding second-level data observation frequency is twice routine observation at 07:15 (Beijing time) and 19:15 (Beijing time) every day, the wind profile radar product data observation frequency is 2-3 minutes, the vertical resolution is 120 m and 240 m, the wind lidar product data is 1 minute, and the vertical resolution is 30 m. The L-band sounding data height and time are matched with the wind profile radar and wind lidar detection closest height and time data, and the linear interpolation method is used to interpolate the sounding second-level data to 30 m and 120 m vertical resolution in height, respectively.

[0108] In steps S102 and S202, a cubic spline fitting is used to establish a cubic spline relationship model of the L-band sounding second-level data and the wind profile radar and wind lidar, respectively. The fitting formula is:

[0109]

[0110] wherein, is the L-band sounding second-level data in each interval, , , , are correction coefficients, is the local variable in the wind profile radar and wind lidar interval.

[0111] As a specific embodiment of the wind profile radar and wind lidar networking method provided by the present application, step S300 includes:

[0112] The historical wind profile radar corrected wind field data and the historical wind lidar corrected wind field data are fused by a vertical layer feature matching fusion algorithm based on density clustering.

[0113]

[0114]

[0115]

[0116] wherein is the historical fused wind field data, is the historical wind profile radar corrected wind field data, is the historical wind lidar corrected wind field data, is the historical wind profile radar corrected wind field data after time and space encryption, is the variance of the historical wind profile radar corrected wind field data, is the variance of the historical wind lidar corrected wind field data, is the weight coefficient of the wind profile radar for fusion, is the weight coefficient of the wind lidar for fusion. More specifically, is the historical wind profile radar corrected wind field data after time and space encryption with a time resolution of 1 minute and a vertical height resolution of 30 m, and the Kalman filter (KF) and cubic spline based on the probability model are used for dynamic time and space interpolation encryption in time and height, respectively.

[0117] In a specific implementation, a vertical layer feature matching fusion algorithm based on density clustering (DBSCAN) is used to fuse the corrected historical wind profile radar wind field data and the historical wind lidar retrieved wind field data. The fusion is performed in the time and space dimensions, respectively, to obtain the historical fused wind field data with a time resolution of 1 minute and a vertical height resolution of 30 m.

[0118] As a specific implementation of the wind profile radar and wind lidar networking method provided in the present application, step S400 includes:

[0119] S401, by a spatial combination algorithm, combining all wind profile radars and all wind lidars in a target region into all possible triangles to form an initial triangle set S0.

[0120] S402, dividing the triangles in the initial triangle set S0 into multiple types from three dimensions of the internal angle, the side length and / or the area of the triangle.

[0121] In step S401 of the embodiment, all wind profile radars and all wind lidars in the target area are combined into as many triangles as possible, except for the case where they cannot form a triangle due to being located on the same straight line or for other reasons that cannot be overcome. Taking the number of wind profile radars and wind lidars as an example, which is 5 in total (the actual number is much larger than this), 10 triangles are formed by selecting any 3 of them, and the number of combinations is represented as C(5, 3).

[0122] In step S402 of the embodiment, classification is performed in one, two, or three dimensions of the internal angle, the side length, and the area.

[0123] As a specific implementation of the wind profile radar and wind lidar networking method provided in the present application, step S500 includes:

[0124] For each type of triangle in the initial triangle set S0, according to the multi-year average value of the historical fusion wind field, based on the control variable idea, a key dynamic parameter networking inversion sensitivity test Test 1 is designed, in which the wind field at the vertex of the triangle is unchanged and the geometric parameters of the triangle are changed, the characteristics of the relative error of the key dynamic parameters with the change of the geometric parameters of the triangle are analyzed, triangles with too large inversion error are removed, and a triangle set S1 considering the networking characteristics is screened.

[0125] In step S500 of the embodiment, for each type of triangle in the initial triangle set S0, according to the historical fusion wind field data, the key dynamic parameters are inverted using a triangle networking inversion algorithm, and the characteristics of the relative error of the key dynamic parameters with the change of the geometric parameters of the triangle are analyzed.

[0126] The inverted key dynamic parameters can be one, two, or three of the horizontal divergence, the relative vorticity, and the vertical velocity, and then the characteristics of the relative error of the horizontal divergence, the relative vorticity, and the vertical velocity with the change of the geometric parameters of the triangle are analyzed respectively.

[0127] As a specific implementation of the wind profile radar and wind lidar networking method provided in the present application, step S600 includes:

[0128] S601, adding observation error disturbance to the wind field elements at different height layers in the historical fusion wind field data to form test wind field data;

[0129] S602, adding observation error disturbance to the wind field elements in the historical fusion wind field data to form test wind field data; based on the control variable idea, for the triangle set S1 considering the networking characteristics, designing the key dynamic parameter networking inversion sensitivity test Test 2 of the triangle vertex wind field change and the triangle geometric parameter invariable, analyzing the characteristics of the relative error of the key dynamic parameters with the change of the observation error of the wind field elements, removing the triangles with too large inversion error accumulation, and screening to obtain the triangle set S2 considering the networking characteristics and the wind field characteristics.

[0130] In step S601 of the embodiment, observation error disturbance is added to the wind field elements at different height layers in the historical fusion wind field data to form test wind field data. The size of the added observation error disturbance can be set according to common radar detection error.

[0131] In step S602 of the embodiment, based on the control variable idea, for the triangle set S1 considering the networking characteristics, the key dynamic parameter networking inversion sensitivity test Test 2 of the triangle vertex wind field change and the triangle geometric parameter invariable is designed, the characteristics of the relative error of the key dynamic parameters with the change of the observation error of the wind field elements are analyzed, the triangles with too large inversion error accumulation are removed, and the triangle set S2 considering the networking characteristics and the wind field characteristics is screened.

[0132] For each type of triangle in the initial triangle set S0, the horizontal divergence, vorticity and / or vertical velocity at different height layers are inverted by using the triangle networking inversion algorithm according to the wind field data at different height layers in the test wind field data, and the characteristics of the relative error of the horizontal divergence, vorticity and / or vertical velocity with the change of the observation error of the wind field elements are analyzed respectively. The key dynamic parameters inverted can be one, two or three of the horizontal divergence, vorticity and vertical velocity, and then the characteristics of the relative error of the horizontal divergence, vorticity and / or vertical velocity with the change of the observation error of the wind field elements are analyzed respectively.

[0133] As a specific implementation of the wind profile radar and wind measuring laser radar networking method provided in the application, the wind field elements in step S600 are wind speed and / or wind direction.

[0134] In the embodiment, the wind field elements can be one or both of the wind speed and wind direction.

[0135] As a specific implementation of the wind profile radar and wind measuring laser radar networking method provided in the application, step S700;

[0136] The different weathers of the target area are classified, the longitude, latitude, and altitude geographical conditions are combined, and based on the control variable thought, for the triangle set S2 considering the networking characteristics and the wind field characteristics, a key dynamic parameter networking inversion sensitivity test Test3 of the triangle vertex wind field invariable, the triangle geometric parameter invariable, and the different weather conditions and the geographical condition change is designed, the characteristics of the relative error of the key dynamic parameter with the weather type and the geographical condition change are analyzed, the triangle with the accumulated inversion error too large is removed, and the triangle set S3 considering the networking characteristics, the wind field characteristics, the different weather conditions, and the geographical condition characteristics is screened.

[0137] Taking the Beijing-Tianjin-Hebei region as an example, the different weathers of the Beijing-Tianjin-Hebei region are classified, the longitude, latitude, and altitude geographical conditions are combined, and based on the control variable thought, for the triangle set S2 considering the networking characteristics and the wind field characteristics, a key dynamic parameter networking inversion sensitivity test Test3 of the triangle vertex wind field invariable, the triangle geometric parameter invariable, and the different weather conditions and the geographical condition change is designed, the influence characteristics of the different weathers and the geographical conditions on the triangle inversion key dynamic parameter of different geometric parameters are evaluated, the triangle with the accumulated inversion error too large is removed, and the triangle set S3 considering the networking characteristics, the wind field characteristics, the different weather conditions, and the geographical condition characteristics is screened.

[0138] As a specific embodiment of the wind profile radar and the wind measurement laser radar networking method provided in the application, the step S900 comprises:

[0139] S901, the real-time wind profile radar wind field observation data of a target area is quality controlled to obtain real-time wind profile radar inversion wind field data; the real-time wind measurement laser radar wind field observation data of the target area is quality controlled to obtain real-time wind measurement laser radar inversion wind field data;

[0140] S902, the real-time wind profile radar inversion wind field data and the real-time wind measurement laser radar inversion wind field data are fused to obtain real-time fusion wind field data;

[0141] S903, for each triangle in the optimal triangle networking scheme, the horizontal components of the wind speed of each vertex at different height layers in the u direction and the v direction are calculated according to the real-time fusion wind field data, wherein the u direction and the v direction are horizontal directions and perpendicular to each other, , wherein, u direction and v direction are horizontal directions and perpendicular to each other, is a vertex serial number, ;

[0142] S904, for each triangle in the optimal triangle networking scheme, the longitude and latitude of each vertex are calculated according to the real-time fusion wind field data, ​​) and the earth radius R to calculate the horizontal component of the side length of the triangle in the u direction and the v direction , ) ;

[0143] S905a, according to the horizontal component of the wind speed of each vertex on each height layer of each triangle ( , ) and the horizontal component of the side length of each edge ( , ), the horizontal divergence of each height layer of each triangle in the optimal triangular networking scheme is calculated ,

[0144] ;

[0145] S905b, according to the horizontal component of the wind speed of each vertex on each height layer of each triangle ( , ) and the horizontal component of the side length of each edge ( , ), the vorticity of each height layer of each triangle in the optimal triangular networking scheme is calculated ,

[0146] ;

[0147] S905c, the horizontal divergence of each height layer of each triangle in the optimal triangular networking scheme is integrated from the ground upwards, and the vertical velocity of each height layer of each triangle is inversely calculated ,

[0148]

[0149] wherein, is the target height layer, is the height layer variable.

[0150] In this embodiment, the u direction and the v direction can be the latitude direction and the longitude direction respectively. The horizontal divergence, vorticity and vertical velocity calculation method of this embodiment is also applicable to steps S500, S600 and S700.

[0151] Please refer to Figure 7 , the application embodiment also provides a dynamic parameter fusion method, and the dynamic parameter fusion method comprises:

[0152] T100, the key dynamic parameters obtained through the wind profile radar and the wind measurement laser radar networking method of any one of the above are used as the inverse key dynamic parameters;

[0153] T200, obtaining reanalysis data or numerical model initial field of a target area and a plurality of grids corresponding to the reanalysis data or the numerical model initial field;

[0154] T300, determining one or more overlapping triangles in the optimal triangular meshing scheme that overlap with a same grid, and calculating a proportion of an overlapping area of each overlapping triangle to a total area of the grid;

[0155]

[0156] wherein, is a proportion of an overlapping area of the i-th overlapping triangle to a total area of the grid, is a serial number of the overlapping triangle, is an overlapping area of the i-th overlapping triangle and the grid, is a total area of the grid; T400, fusing the reanalysis data or the numerical model initial field and the inversion key dynamic parameter to obtain multi-source fusion data

[0157]

[0158]

[0159] wherein, is an obtained meshing inversion characteristic variable of the i-th overlapping triangle, is grid corresponding reanalysis data or numerical model initial field data. Compared with the prior art, the dynamic parameter fusion method of the embodiment of the application takes the key dynamic parameter obtained by the wind profile radar and the wind measurement laser radar meshing method as the inversion key dynamic parameter, fully utilizes the high-precision key dynamic parameter obtained after the wind profile radar and the wind measurement laser radar are meshed, and fuses the reanalysis data or the numerical model initial field on this basis to obtain multi-source fusion data, so that the data accuracy is further improved.

[0160] The common reanalysis data or numerical model initial field usually corresponds to a grid, while the key dynamic parameter obtained by the wind profile radar and the wind measurement laser radar meshing method corresponds to a triangular grid. The dynamic parameter fusion method of the application fuses the reanalysis data or the numerical model initial field and the inversion key dynamic parameter, realizes the fusion of the wind profile radar, the wind measurement laser radar and the reanalysis data or the numerical model initial field, and further improves the data accuracy.

[0161]

[0162] ​​​​In step T100, the key dynamic parameters can be compared with the reanalysis data (such as ERA5) or the initial field of the numerical model of the same time of the target area for multi-level and multi-index comparison and evaluation. The evaluation indexes include mean deviation, root mean square error, correlation coefficient and spatial distribution consistency, so as to quantify the accuracy of the key dynamic parameters and analyze the error sources.

[0163] In step T400, the obtained multi-source fusion data can also be organized according to the time sequence (target time resolution 1 minute) and vertical level (target vertical resolution 30 m) to form a long time sequence, high space-time resolution grid divergence, vorticity and vertical velocity product data set, and a corresponding metadata and storage format are designed for meteorological research and business application.

[0164] As a specific embodiment of the dynamic parameter fusion method provided in the application, in step T100, the key dynamic parameters include horizontal divergence, vorticity and / or vertical velocity; and the key dynamic parameters are subjected to time consistency average value and median value test to serve as the inversion key dynamic parameters.

[0165] In step T100, the key dynamic parameters can be one, two or three of horizontal divergence, vorticity and vertical velocity; and correspondingly, the multi-source fusion data obtained in the final step T400 can be one, two or three of horizontal divergence, vorticity and vertical velocity.

[0166] The above only describes the preferred embodiments of the application and is not intended to limit the application. Any modification, equivalent replacement and improvement made within the spirit and principle of the application shall be included in the protection scope of the application.

Claims

1. A method for networking wind profile radars and wind lidars, characterized in that, The method comprises the following steps: S100, correcting historical wind profile radar wind field data of a target area according to historical in-situ measured wind field data, to obtain historical wind profile radar corrected wind field data; S200, correcting historical wind lidar inversion wind field data of the target area according to the historical in-situ measured wind field data, to obtain historical wind lidar corrected wind field data; S300, fusing the historical wind profile radar corrected wind field data and the historical wind lidar corrected wind field data, to obtain historical fused wind field data; S400, dividing the wind profile radar and the wind lidar of the target area into a plurality of triangles to form an initial triangle set S0; and dividing the triangles in the initial triangle set S0 into a plurality of types according to geometric parameters of the triangles; S500, for each type of triangle in the initial triangle set S0, calculating key dynamic parameters according to the historical fused wind field data by using a triangle networking inversion algorithm, designing a sensitivity test Test 1, analyzing characteristics of relative errors of the key dynamic parameters changing with geometric parameters of the triangles, and screening to obtain a triangle set S1 considering networking characteristics; S600, adding observation error disturbances to wind field elements in the historical fused wind field data to form test wind field data; for the triangle set S1 considering the networking characteristics, designing a key dynamic parameter networking inversion sensitivity test Test 2 of changing wind field at a vertex of a triangle and keeping geometric parameters of the triangle unchanged according to the test wind field data, analyzing characteristics of relative errors of the key dynamic parameters changing with observation errors of the wind field elements, and screening to obtain a triangle set S2 considering the networking characteristics and wind field characteristics; S700, for the triangle set S2 considering the networking characteristics and the wind field characteristics, designing a key dynamic parameter networking inversion sensitivity test Test 3 of keeping wind field at a vertex of a triangle unchanged, keeping geometric parameters of the triangle unchanged, changing weather conditions and geographical conditions according to the historical fused wind field data under different weather and geographical conditions, analyzing characteristics of relative errors of the key dynamic parameters changing with the weather conditions and the geographical conditions, and screening to obtain a triangle set S3 considering the networking characteristics, the wind field characteristics, different weather conditions and geographical conditions; S800, in the triangle set S3 considering the networking characteristics, the wind field characteristics, different weather conditions and geographical conditions, for four triangles composed of four adjacent wind profile radars or wind lidars, encrypting at diagonal lines and midpoints of each side of the four adjacent wind profile radars or wind lidars to construct eight triangles, forming an encrypted triangle networking set S4, and taking the encrypted triangle networking set S4 as an optimal triangle networking scheme; S900, calculating key dynamic parameters of the encrypted triangle networking set S4 by using an area weighting method based on the triangle networking inversion algorithm. 2.The wind profile radar and wind lidar networking method according to claim 1, wherein, The step S100 comprises: S101, performing spatial and temporal matching on the historical in-situ measured wind field data and the historical wind profile radar inversion wind field data. S102, a cubic spline relationship model of the historical in-situ measured wind field data and the historical wind profile radar inversion wind field data is established, and the historical wind profile radar inversion wind field data is corrected according to the cubic spline relationship model to obtain historical wind profile radar corrected wind field data; The step S200 comprises: S201, the historical in-situ measured wind field data and the historical wind lidar inversion wind field data are matched in space and time; S202, a cubic spline relationship model of the historical in-situ measured wind field data and the historical wind lidar inversion wind field data is established, and the historical wind lidar inversion wind field data is corrected according to the cubic spline relationship model to obtain the historical wind lidar corrected wind field data. 3.The wind profile radar and wind lidar networking method according to claim 1, wherein, The step S300 comprises: The historical wind profile radar corrected wind field data and the historical wind lidar corrected wind field data are fused through a vertical layer feature matching fusion algorithm based on density clustering; wherein, is historical fusion wind farm data, is historical wind profiler corrected wind farm data, is historical wind lidar corrected wind farm data, is spatio-temporally encrypted historical wind profiler corrected wind farm data, is variance of historical wind profiler corrected wind farm data, is variance of historical wind lidar corrected wind farm data, is a weight coefficient for wind profiler fusion, is a weight coefficient for wind lidar fusion. 4.The wind profile radar and wind lidar networking method according to claim 1, wherein, The step S400 comprises: S401, all wind profile radars and all wind lidars in the target area are combined into all possible triangles through a spatial combination algorithm to form an initial triangle set S0; S402, the triangles in the initial triangle set S0 are divided into multiple types from three dimensions of the internal angle, the side length and / or the area of the triangle. 5.The wind profile radar and wind lidar networking method according to claim 1, wherein, The step S500 comprises: For each type of triangle in the initial triangle set S0, based on the multi-year average value of the historical fused wind field, a key dynamic parameter networking inversion sensitivity test Test 1 of the triangle vertex wind field being unchanged and the triangle geometric parameter changing is designed based on the control variable idea, the characteristics of the relative error of the key dynamic parameter changing with the triangle geometric parameter are analyzed, the triangles with too large inversion error are removed, and a triangle set S1 considering the networking characteristics is screened. 6.The wind profile radar and wind lidar networking method according to claim 1, wherein, The step S600 comprises: S601, observation error disturbance is added to the wind field elements at different height layers in the historical fused wind field data to form test wind field data; S602, observation error disturbance is added to the wind field elements in the historical fused wind field data to form test wind field data; based on the control variable idea, for the triangle set S1 considering the networking characteristics, a key dynamic parameter networking inversion sensitivity test Test 2 of the triangle vertex wind field changing and the triangle geometric parameter being unchanged is designed, the characteristics of the relative error of the key dynamic parameter changing with the observation error of the wind field element are analyzed, the triangles with too large accumulated inversion error are removed, and a triangle set S2 considering the networking characteristics and the wind field characteristics is screened.

7. The wind profile radar and wind lidar networking method of claim 1, wherein, The step S700 comprises: The different weather of the target area is classified, combined with longitude, latitude, altitude geographical conditions, based on the control variable idea, for the triangle set S2 considering the networking characteristics and wind field characteristics, a key dynamic parameter networking inversion sensitivity test Test 3 is designed, in which the wind field of the triangle vertex is invariable, the triangle geometric parameter is invariable, the different weather conditions and the geographical conditions are changed, the characteristics of the relative error of the key dynamic parameter with the change of the weather type and the geographical condition are analyzed, the triangles with too large inversion error accumulation are removed, and a triangle set S3 considering the networking characteristics, the wind field characteristics, the different weather conditions and the geographical condition characteristics is screened. 8.The wind profile radar and wind lidar networking method according to claim 1, wherein, The step S900 comprises: S901, quality control is performed on real-time wind profile radar wind field detection data of a target area to obtain real-time wind profile radar inversion wind field data; real-time wind measurement laser radar wind field detection data of the target area is quality controlled to obtain real-time wind measurement laser radar inversion wind field data; S902, the real-time wind profile radar inversion wind field data and the real-time wind measurement laser radar inversion wind field data are fused to obtain real-time fusion wind field data; S903, for each triangle in the optimal triangular meshing scheme, according to the real-time fused wind field data, the horizontal components of the wind speed at each vertex on different height layers in the u direction and the v direction are calculated (u, v), wherein the u direction and the v direction are horizontal directions and perpendicular to each other, , , is a vertex number, ; S904, for each triangle in the optimal triangle networking scheme, based on the latitude and longitude of each vertex ( , ) and the Earth's radius R, calculate the horizontal components of the side lengths of the three sides of the triangle in the u and v directions. , ); S905a, calculating the horizontal divergence of each triangle in the optimal triangle networking scheme according to the horizontal component of the wind speed at each vertex of each different height layer of the triangle ( , ) and the horizontal component of the edge length of each edge ( , ) , ； S905b, based on the horizontal component of wind speed at each vertex of each triangle at different height levels ( , ) and the horizontal component of the side length of each side ( , The vorticity at different height levels of each triangle in the optimal triangular network scheme was calculated. , ； S905c, horizontal divergence of different height layers of each triangle in the optimal triangular meshing scheme Integrating upward from the ground, the inversion obtains the vertical velocity of each triangle at different height layers , wherein is the target height level, is the height level variable.

9. A power parameter fusion method, characterized in that, Comprise: T100, the key dynamic parameters obtained by the wind profile radar and wind measurement laser radar networking method according to any one of claims 1-8 are used as the inversion key dynamic parameters; T200, obtaining reanalysis data or numerical model initial field of a target area and a plurality of corresponding grids; T300, determining one or more overlapping triangles overlapping with the same grid in the optimal triangle networking scheme, and calculating the proportion of the overlapping area of each overlapping triangle in the total area of the corresponding grid; wherein, is the proportion of the total area of the grid that is occupied by the th overlapping triangle, is the number of overlapping triangles, is the proportion of the total area of the grid that is occupied by the th overlapping triangle, is the total area of the grid; T400, the reanalysis data or numerical model initial field is fused with the inversion key dynamic parameters to obtain multi-source fusion data , wherein, is the acquired characteristic variable of the network inversion of the th overlapping triangle, is the reanalysis data or numerical model initial field data corresponding to the grid.

10. The power parameter fusion method of claim 9, wherein, In the step T100, the key dynamic parameters include: horizontal divergence, vorticity and / or vertical velocity; after time consistency average value and median value test on the key dynamic parameters, the key dynamic parameters are used as the inversion key dynamic parameters.

Images