A method for extracting radial velocity of wind speed using wind profiler radar data

Abstract

The present invention relates to the technical field of radar speed measurement, and in particular to a method for extracting radial velocity of wind speed by using wind profile radar data, comprising the following steps: acquiring original spectrum data of the wind profile radar, and performing zero-frequency removal processing on the original spectrum data; performing standardization processing on the spectrum data after the zero-frequency removal processing to obtain a standardized data set, and drawing a contour coloring map for the standardized data set; converting the contour coloring map into a grayscale map A, setting a threshold, and performing image corrosion on the grayscale map A according to the threshold to obtain a grayscale map B; and obtaining a frequency deviation value by performing image processing on the grayscale map B. The present invention simplifies the analysis steps of extracting wind speed from wind profile radar data, improves recognition efficiency and accuracy, and contributes to the goal of "carbon neutrality".

Description

Technical Field

[0001] The invention relates to the technical field of radar speed measurement, and in particular to a method for extracting radial speed of wind speed by using wind profile radar data. Background Art

[0002] Wind profiler radar is a remote sensing device that detects high-altitude wind fields by emitting electromagnetic beams in different directions to high altitudes, receiving and processing the information returned by these electromagnetic beams due to the uneven vertical structure of the atmosphere. Wind profiler radar uses the Doppler effect to detect the changes in meteorological elements such as wind direction and wind speed above it with altitude, and has the advantages of high spatial and temporal resolution and high degree of automation.

[0003] The detection of low-altitude wind fields plays an important role in airports and wind power plants. When the millimeter-wave radar detects the wind field within the altitude of 0-1km, it is affected by the underlying surface and it is difficult to extract the wind speed. The inventors have found through research that the original spectrum information in the wind profile radar can be normalized after removing the zero frequency. Figure 2 ), and the standardized image is similar to the classic wind profile image ( Figure 3 ) is similar, through Figure 2 and Figure 3 By comparison, it is found that the standardized contour coloring map can accurately show the size of the signal offset, so relying on image analysis to extract wind speed becomes a feasible method; after searching, no technical solutions identical or similar to the present invention are found. Summary of the invention

[0004] The purpose of the present invention is to provide a method for extracting radial velocity of wind speed by using wind profile radar data, so as to solve the problem that when the millimeter wave radar in the prior art detects the wind field within the height of 0-1km, the wind speed extraction is difficult due to the influence of the underlying surface.

[0005] The technical solution of the present invention is: a method for extracting radial velocity of wind speed using wind profile radar data, comprising the following steps:

[0006] Obtain the original spectrum data of the wind profiler radar and perform zero-frequency processing on the original spectrum data;

[0007] The spectrum data after zero-frequency removal is standardized to obtain a standardized data set, and a contour coloring map is drawn for the standardized data set;

[0008] The contour coloring map is converted into a grayscale map A, a threshold is set, and the grayscale map A is eroded according to the threshold to obtain a grayscale map B;

[0009] The frequency deviation value is obtained by performing image processing on the grayscale image B. The specific steps are as follows:

[0010] a. Cut the grayscale image B from top to bottom to obtain several cutting layers, the number of the cutting layers is the same as the number of wind profile radar scanning layers;

[0011] b. Starting from the top layer of the cutting layer, traverse to the bottom layer. Each cutting layer traverses the pixel points on the midline f along its up and down direction from left to right, recorded as Q (i,j) , traverse the gray value of each pixel, recorded as PQ (i,j) , where i is the number of cutting layers from top to bottom, j is the number of pixel arrangement positions from left to right, and both i and j are natural numbers; traverse the pixel with the largest grayscale value on each midline f, recorded as Pmax Q (i,j) ; After completing the traversal of all cutting layers, traverse to find out several continuous and uninterrupted Pmax Q on each midline f (i,j) , denoted as Pmax Q (i,j) Set; then traverse each midline f where the Pmax Q is located on the edge (i,j) The total number of pixels is recorded as Cedge, and Cedge / 2 can be used to obtain Pmax Q on the midline f of the cutting layer. (i,j) The number of sets is denoted as n, where n is a natural number;

[0012] c. A Pmax Q for each pixel on the midline f of each cutting layer (i,j) The set is recorded as wind profile bright body D, that is, there are n wind profile bright bodies D in each cutting layer; when n = 1, it means that there is only one wind profile bright body D in the cutting layer, that is, there is only one Pmax Q (i,j) The set is recorded as a single wind profile bright body; when n ≥ 2, it means that there are multiple wind profile bright bodies D in the cutting layer, that is, there are multiple Pmax Q (i,j) The collection is recorded as the bright body of the windy profile;

[0013] d. Calculate the frequency deviation value fd in each cutting layer by the value of j at the midpoint of the single wind profile bright body or the midpoint of the multi-wind profile bright body in each cutting layer. The calculation formula is: p is the total number of pixels in each row, floor means rounding down, the midpoint pixel of the multi-wind profile bright body is the average value of the midpoint pixel of several wind profile bright bodies D in the multi-wind profile bright body; suppose there are m pixels Q in a single wind profile bright body (i,j) , each pixel from left to right is recorded as Then the pixel point at the midpoint of the bright body of the single wind profile is If m is an even number, The value is an integer. If m is an odd number, The value of is rounded up; that is, the value of j at the midpoint of the bright body of the single wind profile is The value of j at which The j value at is known; suppose there are y wind profile bright bodies D in the multi-wind profile bright body, the pixel point judgment method at the midpoint of each wind profile bright body D is the same as that of the single wind profile bright body, and the j of the pixel point at the midpoint of each wind profile bright body D is recorded as j1, j2, j3...j from left to right. y , the pixel at the midpoint of the bright body of the windy profile

[0014] e. The radial velocity of the wind speed in each cutting layer is further calculated through the frequency deviation value fd in each cutting layer.

[0015] Preferably, the process of normalizing the original spectrum data to obtain a standardized data set is:

[0016] Assume that the total number of original spectrum points is k. If k is an odd number, remove 2 points for rounding up and down; if k is an even number, remove and 2 points at ;

[0017] Traverse the data of each layer of the original spectrum from top to bottom, and find the maximum and minimum values ​​of each layer, which are recorded as Sp_max and Sp_min respectively;

[0018] Each data Sp_data[t] of each layer is normalized: Sp_normal[t]=(Sp_data[t]-Sp_min) / (Sp_max-Sp_min), where Sp_normal is the data set after the normalization process.

[0019] Preferably, the size of the normalized data set Sp_normal is set to R×s, where R is the number of layers and s is the number of spectrum points in each layer; the classic R-layer one-dimensional line graph is constructed as a two-dimensional contour coloring graph of size R×s;

[0020] The setting in drawing the contour coloring map is: the higher the signal value, the closer the grayscale is to 225, and the lower the signal value, the closer the grayscale is to 0.

[0021] Preferably, in the grayscale image A, the grayscale value of white is 255, and the grayscale value of black is 0;

[0022] During the erosion of the grayscale image A, the number of erosion iterations of the grayscale image A is controlled by a threshold value, and the threshold value is I. Here, SNR is the signal-to-noise ratio.

[0023] Preferably, the cutting layers are sequentially recorded as cutting layers L1, L2, L3, ... L from the top layer to the bottom layer. i ; Traverse and find out several continuous Pmax Q on the line f in each cutting layer(i,j) The steps are: traverse and find each Pmax Q (i,j) The left and right sides, namely Pmax Q (i,j-1) 、Pmax Q (i,j+1) Is there a Pmax Q in all (i,j) , if it exists, it means that Pmax Q (i,j) Not at the edge; if not, it means that Pmax Q (i,j) Located on the edge.

[0024] Preferably, the calculation formula of the radial velocity is: Wherein λ represents the wavelength, which is a known value in wind profiler radar.

[0025] Compared with the prior art, the advantages of the present invention are:

[0026] (1) The present invention performs zero-frequency processing on the original spectrum data of wind profiler radar data; performs standardization processing on the spectrum data after zero-frequency processing to obtain a standardized data set, and draws a contour coloring map for the standardized data set; converts the contour coloring map into a grayscale map A, sets a threshold, and performs image corrosion on the grayscale map A according to the threshold to obtain a grayscale map B; performs image processing on the grayscale map B to obtain a frequency deviation value, thereby extracting the radial velocity of the beam, and using the data of different beams to finally calculate the wind speed and wind direction.

[0027] (2) The present invention improves the recognition efficiency and accuracy, simplifies the analysis steps, and improves the speed resolution and data availability of wind profile radar in mountainous areas and foggy days without changing the hardware conditions. It can accurately output key information such as wind speed, wind direction and turbulence intensity to wind power plants, assist power plants in predicting wind power generation, and contribute to the "carbon neutrality" goal. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] The present invention will be further described below in conjunction with the accompanying drawings and embodiments:

[0029] Figure 1 This is an image of the data described in this embodiment that has not been standardized.

[0030] Figure 2 This is the image after the data standardization processing described in this embodiment.

[0031] Figure 3 It is the classic wind profile diagram described in the background technology.

[0032] Figure 4 This is the grayscale image A described in this embodiment.

[0033] Figure 5 This is the grayscale image B described in this embodiment.

[0034] Figure 6 This is a schematic diagram of performing average cutting from top to bottom on the grayscale image B according to this embodiment.

[0035] Figure 7 Schematic diagram of the pixel point at the midpoint of the bright body of a single wind profile in each cutting layer described in this embodiment.

[0036] Figure 8 Schematic diagram of the pixel point at the midpoint of the bright body of the multi-wind profile described in this embodiment.

[0037] Fig. 9 This is a flowchart of a method for extracting radial velocity of wind speed using wind profile radar data described in this embodiment. DETAILED DESCRIPTION

[0038] The present invention is further described in detail below in conjunction with specific embodiments:

[0039] In the description of the invention, it is necessary to understand that the terms “center”, “longitudinal”, “lateral”, “length”, “width”, “thickness”, “up”, “down”, “front”, “back”, “left”, “right”, “vertical”, “horizontal”, “top”, “bottom”, “inside”, “outside”, “clockwise”, “counterclockwise”, etc., indicating the orientation or position relationship, are based on the orientation or position relationship shown in the drawings, and are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the invention.

[0040] like Figure 2 , 4 , 5, and 9, a method for extracting radial velocity of wind speed using wind profile radar data comprises the following steps: obtaining original spectrum data of wind profile radar, removing zero frequency from the original spectrum data; performing standardization on the spectrum data after zero frequency removal ( Figure 2 ), obtain a standardized data set, draw a contour color map for the standardized data set; convert the contour color map into a grayscale map A( Figure 4 ), set the threshold, and perform image erosion on the grayscale image A according to the threshold to obtain the grayscale image B ( Figure 5 ); The frequency deviation value is obtained by performing image processing on the grayscale image B.

[0041] The process of normalizing the original spectrum data to obtain a standardized data set is as follows: let the total number of original spectrum points be k. If k is an odd number, remove 2 points for rounding up and down; if k is an even number, remove and In order to avoid zero-frequency interference, the point at the midpoint of the spectrum needs to be removed; traverse the original spectrum data from top to bottom, find the maximum and minimum values ​​for each layer, and record them as Sp_max and Sp_min respectively; standardize each data Sp_data[t] of each layer: Sp_normal[t] = (Sp_data[t]-Sp_min) / (Sp_max-Sp_min), where Sp_normal is the standardized data set. Due to the difference in signal strength between radar scanning layers, the magnitude of the data may be different. If the data is not standardized, the subsequent images will be inaccurate and the wind speed cannot be accurately extracted; Figure 1 For images where the data is not standardized, Figure 1 and Figure 2 It can be seen from the comparison that when the data is not standardized, much information in the image cannot be displayed, which increases the difficulty of extracting the wind profile.

[0042] Assume that the size of the standardized dataset Sp_normal is R×s, where R is the number of layers and s is the number of spectrum points in each layer; the classic R-layer one-dimensional line graph is constructed as a two-dimensional contour coloring graph of size R×s; when drawing the contour coloring graph, it is set that the higher the signal value, the closer the grayscale is to 225, and the lower the signal value, the closer the value is to 0. Since the contour coloring graph is not conducive to subsequent image processing, it is converted into a grayscale graph A, in which the grayscale value of white is 255 and the grayscale value of black is 0.

[0043] In the process of grayscale image A corrosion, the number of grayscale image A corrosion iterations is controlled by the threshold, and the threshold is Among them, SNR is the signal-to-noise ratio; a high signal-to-noise ratio does not require a high number of iterations, while a low signal-to-noise ratio requires more iterations; the purpose of performing image corrosion on the grayscale image A to obtain the grayscale image B is to remove the low-brightness area and retain the high-brightness area. The high-brightness area corresponds to the part with a strong signal in the original data, thereby reducing the difficulty of extracting the wind speed; since the frequency deviation position information at the location where the signal is stronger needs to be obtained in the present invention, the strength of the signal is reflected in the grayscale of the image. The higher the grayscale value, the stronger the signal. However, there is still certain interference and noise, so a certain amount of noise is removed by the image corrosion method.

[0044] The specific steps of obtaining the frequency deviation value by performing image processing on the grayscale image B are as follows:

[0045] a. Figure 6 As shown in the figure, the grayscale image B is cut evenly from top to bottom to obtain several cutting layers, and the number of cutting layers is the same as the number of wind profile radar scanning layers; the cutting layers from the top to the bottom are recorded as cutting layers L1, L2, L3, ... L i , Figure 6In the figure, the vertical dotted line where e is located represents the 0 frequency; it should be noted that Figure 6 It is a schematic diagram for explaining how to cut the grayscale image B and the position of the line f in each cutting layer and the position of the 0 frequency in the grayscale image B. In practice, the grayscale image B is cut as follows: Figure 6 cutting and traversal.

[0046] b. Start from the top layer of the cutting layer and traverse to the bottom layer. Each cutting layer is traversed from left to right in turn to find the midline f along its up and down direction (such as Figure 6 The pixel point on the (i,j) , traverse the gray value of each pixel, recorded as PQ (i,j) , where i is the number of cutting layers from top to bottom, j is the number of pixel arrangement positions from left to right, and both i and j are natural numbers; traverse the pixel with the largest grayscale value on each midline f, recorded as Pmax Q (i,j) ; After completing the traversal of all cutting layers, traverse to find out several continuous and uninterrupted Pmax Q on each midline f (i,j) , denoted as Pmax Q (i,j) Set; that is, judge PmaxQ (i,j) The continuity of the judgment step is: traverse and find each Pmax Q (i,j) The left and right sides, namely Pmax Q (i,j-1) 、PmaxQ (i,j+1) Is there a Pmax Q in all (i,j) , if it exists, it means that Pmax Q (i,j) Not at the edge; if not, it means that Pmax Q (i,j) Located at the edge. Then traverse each midline f where the Pmax Q is located at the edge (i,j) The total number of pixels is recorded as Cedge, and Cedge / 2 can be used to obtain Pmax Q on the midline f of the cutting layer. (i,j) The number of sets is denoted as n, where n is a natural number.

[0047] c. A Pmax Q for each pixel on the midline f of each cutting layer (i,j) The set is recorded as wind profile bright body D, that is, there are n wind profile bright bodies D in each cutting layer; when n = 1, it means that there is only one wind profile bright body D in the cutting layer, that is, there is only one Pmax Q (i,j) The set is recorded as a single wind profile bright body; when n ≥ 2, it means that there are multiple wind profile bright bodies D in the cutting layer, that is, there are multiple Pmax Q (i,j) The collection is recorded as the bright body of the windy profile.

[0048] d. Calculate the frequency deviation value fd in each cutting layer by the value of j at the midpoint of the single wind profile bright body or the midpoint of the multi-wind profile bright body in each cutting layer. The calculation formula is: p is the total number of pixels in each row, floor means rounding down, the midpoint pixel of the multi-wind profile bright body is the average value of the midpoint pixel of several wind profile bright bodies D in the multi-wind profile bright body; suppose there are m pixels Q in a single wind profile bright body (i,j) , each pixel from left to right is recorded as Then the pixel point at the midpoint of the bright body of the single wind profile is If m is an even number, The value is an integer. If m is an odd number, The value of is rounded up; that is, the value of j at the midpoint of the bright body of the single wind profile is The value of j at which The value of j at is known; for example, Figure 7 As shown, there are 5 pixels in the bright body of the single wind profile, and each pixel is recorded from left to right as Then the pixel point at the midpoint of the bright body of the single wind profile is Since 5 is an odd number, The value of is rounded up to 3, then for That is, the value of j at the midpoint of the bright body of the single wind profile is The j value at H is given by the j of each pixel (i.e., the number of pixel positions from left to right on the midline f of the cutting layer). If the j values ​​of these five pixels are 6, 7, 8, 9, and 10 from left to right, then The j is equal to 8.

[0049] Assume that there are y wind profile bright bodies D in the multi-wind profile bright body. The pixel point judgment method at the midpoint of each wind profile bright body D is the same as that of the single wind profile bright body. The j of the pixel point at the midpoint of each wind profile bright body D is recorded as j1, j2, j3...j from left to right. y , the pixel at the midpoint of the bright body of the windy profile For example, Figure 8 As shown, there are three wind profile bright bodies D in the multi-wind profile bright body, and the midpoint pixel of each wind profile bright body D is recorded as H1, H2, and H3 respectively. The j of the three H1, H2, and H3 is recorded as j1, j2, and j3 from left to right respectively. If the values ​​of j1, j2, and j3 are 3, 7, and 11 respectively (the specific algorithm for obtaining the values ​​of j1, j2, and j3 is the same as the method for obtaining the value of j of the pixel at the midpoint of the single wind profile bright body mentioned above), then the midpoint pixel of the multi-wind profile bright body is That is, H2 is the pixel at the midpoint of the bright body of the windy profile;

[0050] It should be noted that if there is a single wind profile bright body in the cutting layer, the pixel point at the midpoint of the single wind profile bright body is the frequency deviation of the cutting layer; if there are multiple wind profile bright bodies in the cutting layer, the average value of the pixel points at the midpoints of several wind profile bright bodies D in the multiple wind profile bright body (i.e., the pixel point at the midpoint of the multiple wind profile bright body) is the frequency deviation of the cutting layer; the position of the pixel point at the midpoint of the multiple wind profile bright body may be outside the multiple wind profile bright body (for example, the black pixel point between each wind profile bright body D in the multiple wind profile bright body) in addition to being on the multiple wind profile bright body; Figure 7 , Figure 8 The examples are just for the sake of ease of understanding. In practical applications, the actual data shall prevail.

[0051] e. The radial velocity of the wind speed in each cutting layer is further calculated by the frequency deviation value fd in each cutting layer. Specifically, the calculation formula of the radial velocity is: Where V is the radial velocity and λ is the wavelength, which is a known value in wind profiler radar.

[0052] Finally, the wind speed and direction can be calculated from the radial velocity measured by the beam at each cutting layer.

[0053] The above embodiments are only for illustrating the technical concept and features of the present invention, and their purpose is to enable people familiar with this technology to understand the content of the present invention and implement it accordingly, and they cannot be used to limit the scope of protection of the present invention. For those skilled in the art, it is obvious that the present invention is not limited to the details of the above exemplary embodiments, and the present invention can be implemented in other specific forms without departing from the spirit or basic features of the present invention. Therefore, no matter from which point of view, the embodiments should be regarded as exemplary and non-restrictive. The scope of the present invention is limited by the attached claims rather than the above description, and it is intended to include all changes within the meaning and scope of the equivalent elements of the claims in the present invention.

Claims

1. A method for extracting radial velocity of wind speed using wind profiler radar data, characterized in that: The following steps are involved: Obtain the original spectrum data of the wind profiler radar and perform zero-frequency processing on the original spectrum data; The spectrum data after zero-frequency removal is standardized to obtain a standardized data set, and a contour coloring map is drawn for the standardized data set; The contour coloring map is converted into a grayscale map A, a threshold is set, and the grayscale map A is eroded according to the threshold to obtain a grayscale map B; The frequency deviation value is obtained by performing image processing on the grayscale image B. The specific steps are as follows: a. Cut the grayscale image B from top to bottom to obtain several cutting layers, the number of the cutting layers is the same as the number of wind profile radar scanning layers; b. Starting from the top layer of the cutting layer, traverse to the bottom layer. Each cutting layer traverses the pixel points on the midline f along its up and down direction from left to right, recorded as Q (i,j) , traverse the gray value of each pixel, recorded as PQ (i,j) , where i is the number of cutting layers from top to bottom, j is the number of pixel arrangement positions from left to right, and both i and j are natural numbers; traverse the pixel with the largest grayscale value on each midline f, recorded as PmaxQ (i,j) ; After completing the traversal of all cutting layers, traverse to find out several continuous and uninterrupted PmaxQ on each midline f (i,j) , denoted as PmaxQ (i,j) Set; then traverse each midline f where the PmaxQ is located on the edge (i,j) The total number of pixels is recorded as Cedge, and Cedge / 2 can be used to obtain PmaxQ on the midline f of the cutting layer. (i,j) The number of sets is denoted as n, where n is a natural number; c. A PmaxQ for each pixel on the midline f of each cutting layer (i,j) The set is recorded as wind profile bright body D, that is, there are n wind profile bright bodies D in each cutting layer; when n = 1, it means that there is only one wind profile bright body D in the cutting layer, that is, there is only one PmaxQ (i,j) Set, recorded as a single wind profile bright body; when n ≥ 2, it means that there are multiple wind profile bright bodies D in the cutting layer, that is, there are multiple PmaxQ (i,j) The collection is recorded as the bright body of the windy profile; d. Calculate the frequency deviation value fd in each cutting layer by the value of j at the midpoint of the single wind profile bright body or the midpoint of the multi-wind profile bright body in each cutting layer. The calculation formula is: p is the total number of pixels in each row, floor means rounding down, the midpoint pixel of the multi-wind profile bright body is the average value of the midpoint pixel of several wind profile bright bodies D in the multi-wind profile bright body; suppose there are m pixels Q in a single wind profile bright body (i,j) , each pixel from left to right is recorded as Then the pixel point at the midpoint of the bright body of the single wind profile is If m is an even number, The value is an integer. If m is an odd number, The value of is rounded up; that is, the value of j at the midpoint of the bright body of the single wind profile is The value of j at which The j value at is known; suppose there are y wind profile bright bodies D in the multi-wind profile bright body, the pixel point judgment method at the midpoint of each wind profile bright body D is the same as that of the single wind profile bright body, and the j of the pixel point at the midpoint of each wind profile bright body D is recorded as j1, j2, j3...j from left to right. y , the pixel at the midpoint of the bright body of the windy profile e. The radial velocity of the wind speed in each cutting layer is further calculated through the frequency deviation value fd in each cutting layer.

2. The method for extracting radial velocity of wind speed using wind profiler radar data according to claim 1, characterized in that: The process of obtaining the original spectrum data of the wind profiler radar, removing the zero frequency from the original spectrum data, and standardizing the spectrum data after the zero frequency removal to obtain a standardized data set is as follows: Assume that the total number of original spectrum points is k. If k is an odd number, remove 2 points for rounding up and down; if k is an even number, remove and 2 points at ; Traverse the data of each layer of the original spectrum from top to bottom, and find the maximum and minimum values ​​of each layer, which are recorded as Sp_max and Sp_min respectively; Each data Sp_data[t] of each layer is normalized: Sp_normal[t]=(Sp_data[t]-Sp_min) / (Sp_max-Sp_min), where Sp_normal is the data set after the normalization process.

3. The method for extracting radial velocity of wind speed using wind profiler radar data according to claim 2, characterized in that: Assume that the size of the standardized data set Sp_normal is R×s, where R is the number of layers and s is the number of spectrum points in each layer; the classic R-layer one-dimensional line graph is constructed as a two-dimensional contour coloring graph of size R×s; The setting in drawing the contour coloring map is: the higher the signal value, the closer the grayscale is to 225, and the lower the signal value, the closer the grayscale is to 0.

4. The method for extracting radial velocity of wind speed using wind profiler radar data according to claim 3, characterized in that: In the grayscale image A, the grayscale value of white is 255, and the grayscale value of black is 0; During the erosion of the grayscale image A, the number of erosion iterations of the grayscale image A is controlled by a threshold value, and the threshold value is I. Here, SNR is the signal-to-noise ratio.

5. The method for extracting radial velocity of wind speed using wind profiler radar data according to claim 4, characterized in that: The cutting layers are sequentially recorded as cutting layers L1, L2, L3, ... L from the top layer to the bottom layer. i ; Traverse and find out several continuous Pmax Q on the line f in each cutting layer (i,j) The steps are: traverse and find each Pmax Q (i,j) The left and right sides, namely Pmax Q (i,j-1) 、Pmax Q (i,j+1) Is there a Pmax Q in all (i,j) , if it exists, it means that Pmax Q (i,j) Not at the edge; if not, it means that Pmax Q (i,j) Located on the edge.

6. The method for extracting radial velocity of wind speed using wind profiler radar data according to claim 5, characterized in that: The radial velocity is calculated as follows: Wherein λ represents the wavelength, which is a known value in wind profiler radar.

Images