An automatic analysis method for dual polarization weather radar data quality

Abstract

The present application discloses an automatic analysis method for the quality of dual-polarization weather radar base data, which relates to the field of dual-polarization weather radar. The method uses a fuzzy logic method to identify and remove ground object echoes in the volume scanning base data of the dual-polarization weather radar. After completing the data preprocessing of the volume scanning base data to obtain precipitation echo base data, data statistics and evaluation are performed according to the reflectivity factor, differential reflectivity, and differential phase. The base data quality is analyzed from multiple dimensions such as radar detection power, radial continuity detection results, initial differential phase stability, and layered cloud precipitation echo detection quantity. This method can help radar debugging and installation personnel accurately evaluate the quality of radar base data, and can also provide support for researchers engaged in radar data application to quickly understand the quality of radar base data.

Description

Technical Field

[0001] The present application relates to the field of dual-polarization weather radar, and in particular to an automatic analysis method for the quality of dual-polarization weather radar base data. Background Art

[0002] Compared with single polarization weather radar, dual polarization weather radar can estimate the shape and phase of precipitation particles more accurately, and can detect and identify non-spherical ice crystals, large raindrops or hail. In the evolution of precipitation systems, dual polarization weather radar can more directly reflect the echo structure of severe convective weather. Based on this, dual polarization weather radar has become an important means of monitoring disastrous weather.

[0003] With the increasing application of X-band dual-polarization weather radar, the requirements for radar data quality are also getting higher and higher. In order to ensure the accuracy of observation data, dual-polarization weather radar has put forward higher requirements for hardware performance, such as antenna accuracy, consistency and isolation of horizontal and vertical channels, which are key factors affecting radar data quality. In the actual observation process, X-band dual-polarization weather radar is easily affected by factors such as ground echo, environmental noise, signal attenuation, etc., and sometimes due to errors in radar calibration methods, signal processing algorithms and some losses of radar hardware, the detection performance of the radar cannot meet the indicators required by the radar factory test. These problems will lead to the decline in the quality of the base data of the dual-polarization weather radar and affect the detection accuracy. Therefore, it is necessary to analyze the data quality of the base data before applying it. At present, some mainstream practices are to perform quality analysis on a single detection quantity, which has great limitations, especially for the decline in data quality caused by signal processing algorithms and hardware losses. These problems are not obvious with conventional data correction algorithms, and the data quality of the base data cannot be accurately evaluated. Summary of the invention

[0004] In view of the above problems and technical requirements, this application proposes a method for automatically analyzing the quality of dual-polarization weather radar base data. The technical solution of this application is as follows:

[0005] A dual-polarization weather radar base data quality automatic analysis method, the dual-polarization weather radar base data quality automatic analysis method comprising:

[0006] The fuzzy logic method is used to identify and remove the ground object echo in the volume scanning base data of the dual polarization weather radar, and the data preprocessing of the volume scanning base data is completed to obtain the precipitation echo base data;

[0007] According to the reflectivity factor Z in the precipitation echo base data H , differential reflectivity Z DR , differential phase Φ DPObtain radar signal basic indicators and identify layered cloud precipitation echoes from precipitation echo base data. Radar signal basic indicators include radar detection power, radial continuity detection results and initial differential phase stability;

[0008] When all the basic indicators of radar signals are qualified and the detection quantity of layered cloud precipitation echo meets the corresponding requirements, it is determined that the data quality of the volume scanning base data of the dual polarization weather radar meets the standard; otherwise, it is determined that the data quality of the volume scanning base data of the dual polarization weather radar does not meet the standard.

[0009] A further technical solution is to obtain the radar detection power and detect whether it is qualified, including:

[0010] The minimum measurable reflectivity factor Z of all radial corresponding distance libraries at different elevation angles in the statistical precipitation echo base data H , and obtain the minimum measurable reflectivity factor Z H Distribution results with distance;

[0011] When the minimum measurable reflectivity factor Z H When the distribution result with distance meets the reflectivity factor distribution requirement, it is determined that the radar detection power is qualified; otherwise, it is determined that the radar detection power is unqualified.

[0012] A further technical solution is to obtain the radial continuity test result and test whether it is qualified, including:

[0013] Calculate the reflectivity factor Z in the precipitation echo base data H The first-order difference Z in the radial direction H_diff , differential reflectivity Z DR The first-order difference Z in the radial direction DR_diff and the differential phase Φ DP The first-order difference in the radial direction Φ DP_diff ;

[0014] When the radial direction Z H_diff are less than or equal to the corresponding threshold Z thd , and radially Z DR_diff Are less than or equal to the corresponding threshold ZDR thd , and radially Φ DP_diff are less than or equal to the corresponding threshold Φ thd When the radial continuity test result is qualified, it is determined that the radial continuity test result is qualified; otherwise, it is determined that the radial continuity test result is unqualified.

[0015] A further technical solution is that the automatic analysis method for dual-polarization weather radar base data quality also includes:

[0016] When there is Z in the radial direction H_diff Greater than the corresponding threshold Z thd , and is greater than the threshold Z thdZ H_diff When the corresponding reflectivity factor jump region is located at the pulse splicing position, the reflectivity factor Z of the reflectivity factor jump region is H Correction is made so that the radial Z H_diff are less than or equal to the corresponding threshold Z thd ; When the reflectivity factor jump region is not at the pulse splicing position, the reflectivity factor jump region is output;

[0017] When there is Z in the radial direction DR_diff Greater than the corresponding threshold ZDR thd , and is greater than the threshold ZDR thd Z DR_diff When the corresponding differential reflectivity jump region is located at the pulse splicing position, the differential reflectivity Z of the differential reflectivity jump region is DR Correction is made so that the radial Z DR_diff Are less than or equal to the corresponding threshold ZDR thd ; When the differential reflectivity jump region is not at the pulse splicing position, the differential reflectivity jump region is output;

[0018] When there is Φ in the radial direction DP_diff Greater than the corresponding threshold Φ thd , and is greater than the threshold Φ thd Φ DP_diff When the corresponding differential phase jump region is located at the pulse splicing position, the differential phase Φ of the differential phase jump region is DP Correction is made so that the radial Z H_diff are less than or equal to the corresponding threshold Z thd ; When the differential phase jump region is not at the pulse splicing position, the differential phase jump region is output;

[0019] When it is determined that the radial continuity test result is unqualified, at least one of a reflectivity factor jump area, a differential reflectivity jump area and a differential phase jump area is also output.

[0020] A further technical solution is to adjust the reflectivity factor Z in the reflectivity factor jump region. H , the differential reflectivity Z in the differential reflectivity jump region DR 、Differential phase Φ in the differential phase jump region DP The amendments include:

[0021] Determine the corrected reflectivity factor of the reflectivity factor jump region at the pulse splicing position as Z H (after)-Z H (front), Z H (after) is the average reflectivity factor of several distance bins after the pulse splicing position, Z H(front) is the average of the reflectivity factors of several distance bins before the pulse splicing position;

[0022] Determine the corrected differential reflectivity of the differential reflectivity jump region at the pulse splicing position as Z DR (after)-Z DR (front), Z DR (after) is the average of the differential reflectivity of several distance bins after the pulse splicing position, Z DR (front) is the average of the differential reflectivities of several range bins before the pulse splicing position;

[0023] Determine the corrected differential phase of the differential phase jump region at the pulse splicing position as Φ DP (after)-Φ DP (front), Φ DP (after) is the average of the differential phases of several distance bins after the pulse splicing position, Φ DP (front) is the average of the differential phases of several range bins before the pulse splicing position.

[0024] A further technical solution is to obtain the initial differential phase stability and detect whether it is qualified, including:

[0025] In each radial direction of each elevation angle, the average value of the differential phases of the last several distance bins of the effective precipitation echo starting point in the radial direction is taken as the initial differential phase of the radial direction, and the correlation coefficients of the continuous K distance bins searched outward from the effective precipitation echo starting point in each radial direction are greater than the correlation coefficient threshold and the standard deviation of the differential phases of the continuous K distance bins is less than the predetermined threshold;

[0026] Calculate the median value of all radial initial differential phases of each layer elevation angle to obtain the initial differential phase of each layer elevation angle;

[0027] Calculate the average value of the initial differential phases of all layer elevation angles in the precipitation echo base data to obtain the initial differential phase of the precipitation echo base data;

[0028] The standard deviation of the initial differential phase of several consecutive groups of volume scan basis data is calculated to obtain the initial differential phase stability. When the standard deviation is less than the standard deviation threshold, the initial differential phase stability is determined to be qualified, otherwise the initial differential phase stability is determined to be unqualified.

[0029] A further technical solution is to identify the stratiform cloud precipitation echo from the precipitation echo base data, including:

[0030] For the reflectivity factor Z H Any point of Thd1: When there is Z within the range of radius R1 with the point as the center H≥Thd1, it is determined that the point belongs to the convective cloud point; when the reflectivity factor Z of all points within the range of radius R1 with the point as the center H When Thd1 is less than Thd1, the average reflectivity factor of all points within the range of radius R2 with point as the center is calculated as the average background field intensity Aυ; when A υ ≤0 and the reflectivity factor Z of the point H -A υ ≥Thd2, the point is determined to be a convective cloud point. υ ≤0 and the reflectivity factor Z of the point H -A υ <Thd2, the point is determined to be a layered cloud point; when 0<A υ <Thd3 and the reflectivity factor of the point When 0<A υ <Thd3 and the reflectivity factor of the point When A υ ≥Thd3 and the reflectivity factor Z of the point H -A υ ≥0, the point is determined to belong to the convective cloud point. υ ≥Thd3 and the reflectivity factor Z of the point H -A υ When <0, the point is determined to be a layered cloud point;

[0031] The reflectivity factor Z H All other points within the influence radius of any convective cloud point <Hhd1 are corrected and attributed to convective cloud points. The larger the reflectivity factor of each convective cloud point, the larger the influence radius of the convective cloud point.

[0032] The stratiform cloud precipitation echo is obtained by comprehensively identifying all stratiform cloud points.

[0033] A further technical solution is to detect whether the detection quantity of the stratiform cloud precipitation echo meets the corresponding requirements, including:

[0034] The numerical distribution of each detection quantity of the stratiform cloud precipitation echo is statistically analyzed, including signal-to-noise ratio, pre-filter reflectivity, reflectivity factor, radial velocity, velocity spectrum width, differential reflectivity, correlation coefficient, differential phase and differential phase shift rate;

[0035] When the numerical distribution of each detection quantity is respectively within the corresponding preset distribution range, it is determined that the detection quantity of the layered cloud precipitation echo meets the corresponding requirements; otherwise, it is determined that the detection quantity of the layered cloud precipitation echo does not meet the corresponding requirements.

[0036] A further technical solution is to use fuzzy logic to identify and remove ground object echoes in the volume scanning data of the dual polarization weather radar, including:

[0037] For each distance library in the volume scan base data, Calculate the reflectivity factor Z on the distance library H Horizontal texture T dBZ ,according to Calculate the reflectivity factor Z on the distance library H The radial variation between the libraries S PIN ,according to Calculate the differential reflectivity Z on the distance library DR Horizontal texture T ZDR ,according to The differential phase Φ on the distance library is calculated DP Horizontal texture Among them, Z i，j Represents the reflectivity factor of azimuthal index i and radial index j, Z i，j+1 represents the reflectivity factor of azimuthal index i and radial index j+1, Z thresh is the threshold for intensity change between distances, ZDR i，j The differential reflectivity of azimuthal index i and radial index j, ZDR i，j+1 represents the differential reflectivity of azimuthal index i and radial index j+1, Φ DPi，j represents the differential phase of azimuth index i and radial index j, Φ DPi，j+1 represents the differential phase of azimuth index i and radial index j+1, N A The total number of indices representing the orientation, N R Indicates the total number of radial indices;

[0038] The fuzzy logic method is used to calculate the horizontal texture T on each distance library. dBZ , the degree of variation between libraries S PIN , horizontal texture T ZDR and horizontal texture Check whether the distance library is a ground object echo;

[0039] All distance libraries of ground object echoes in the volume scanning base data are removed, and the data preprocessing of the volume scanning base data is completed to obtain precipitation echo base data.

[0040] A further technical solution is to use fuzzy logic method to calculate the horizontal texture T of each distance library. dBZ , the degree of variation between libraries S PIN , horizontal texture T ZDR and horizontal texture Detecting whether the distance library is a ground object echo includes:

[0041] Set the horizontal texture T on the distance library dBZ Enter membership functions in fuzzy logic Get T dBZ Membership degree f1(T dBZ );

[0042] The inter-library variation degree S on the distance library PIN Enter membership functions in fuzzy logic Get S PIN Membership degree f2(S PIN );

[0043] Set the horizontal texture T on the distance library ZDR Enter membership functions in fuzzy logic Get T ZDR Membership degree f3(T ZDR );

[0044] Set the horizontal texture on the distance library Enter membership functions in fuzzy logic get Membership

[0045] Set the distance library to T dBZ Membership degree f1(T dBZ ), S PIN Membership degree f2(S PIN ), T ZDR Membership degree f3(T ZDR )and Membership Perform weighted calculation to obtain the comprehensive membership of the distance library;

[0046] When the comprehensive membership of the distance library is greater than the membership threshold, the distance library is determined to be a ground object echo.

[0047] The beneficial technical effects of this application are:

[0048] The present application discloses an automatic analysis method for the quality of dual-polarization weather radar base data. The method removes ground object echoes from volume scanning base data and reduces the influence of clutter. The method analyzes the base data quality from multiple dimensions through data statistics and evaluation methods. The method can help radar debugging and installation personnel to comprehensively, accurately and reliably evaluate the quality of radar base data, and can also provide support for researchers engaged in radar data applications to quickly understand the quality of radar base data.

[0049] In addition, based on quality detection, this method can also detect and correct the phenomenon of discontinuous echo at the splicing of wide and narrow pulses caused by the use of solid-state transmitters. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1It is a method flow chart of a method for automatically analyzing the quality of dual-polarization weather radar-based data in one embodiment of the present application.

[0051] Figure 2 It is a flow chart of a method for preprocessing volume scan base data to obtain precipitation echo base data in one embodiment of the present application.

[0052] Figure 3 It is a membership function used when using fuzzy logic method to identify and remove ground object echoes in volume scanning data of dual polarization weather radar in one embodiment of the present application.

[0053] Figure 4 It is a flow chart of a method for obtaining radial continuity detection results in one embodiment of the present application.

[0054] Figure 5 It is a flow chart of a method for identifying layered cloud precipitation echoes from precipitation echo base data in one embodiment of the present application. DETAILED DESCRIPTION

[0055] The specific implementation of the present application is further described below in conjunction with the accompanying drawings.

[0056] The present application discloses a dual polarization weather radar base data quality automatic analysis method, the dual polarization weather radar base data quality automatic analysis method comprises the following steps, please refer to Figure 1 The flowchart shown:

[0057] Step 1: Use fuzzy logic method to identify and remove ground object echoes in the volume scanning base data of the dual-polarization weather radar, complete data preprocessing of the volume scanning base data to obtain precipitation echo base data. Data preprocessing can avoid the influence of ground object echoes on subsequent statistical analysis results, and improve the accuracy and reliability of the analysis results.

[0058] Step 2: According to the reflectivity factor Z in the precipitation echo base data H , differential reflectivity Z DR , differential phase Φ DP The basic indicators of radar signals are obtained and the stratiform cloud precipitation echoes are identified from the precipitation echo base data. The basic indicators of radar signals include radar detection power, radial continuity detection results and initial differential phase stability.

[0059] Step 3: When all radar signal basic indicators are qualified and the detection amount of stratiform cloud precipitation echo meets the corresponding requirements, it is determined that the data quality of the dual-polarization weather radar volume scanning base data meets the standard and can be used for subsequent use. Otherwise, it is determined that the data quality of the dual-polarization weather radar volume scanning base data does not meet the standard and can be considered to be discarded to avoid low data quality affecting subsequent detection effects.

[0060] like Figure 1In order to clearly represent the logic, the diagram is illustrated by detecting in turn whether the basic indicators of each radar signal are qualified, and finally detecting whether the detection amount of the layered cloud precipitation echo meets the corresponding requirements. However, those skilled in the art can understand that there is no specific order for the detection of these aspects, and they are generally performed simultaneously.

[0061] In one embodiment, the method of performing data preprocessing on the volume scan base data in step 1 to identify and remove the ground object echo to obtain precipitation echo base data includes the following steps, please refer to Figure 2 Flowchart:

[0062] Step 210, for each distance library in the volume scan base data, calculating the characteristic quantity of the distance library includes:

[0063] (1) According to Calculate the reflectivity factor Z on the distance library H Horizontal texture T dBZ .

[0064] Among them, Z i，j Represents the reflectivity factor of azimuthal index i and radial index j, Z i，j+1 Represents the reflectivity factor of azimuthal index i and radial index j+1. N A The total number of indices representing the orientation, N R Indicates the total number of indices in the radial direction.

[0065] (2) According to Calculate the reflectivity factor Z on the distance library H The radial variation between the libraries S PIN .

[0066] in, Z thresh is the threshold of intensity change between distances, set to 3dBZ.

[0067] (3) According to Calculate the differential reflectivity Z on the distance library DR Horizontal texture T ZDR .

[0068] Among them, ZDR i，j The differential reflectivity of azimuthal index i and radial index j, ZDR i，j+1 Represents the differential reflectivity of azimuthal index i and radial index j+1.

[0069] (4) According to The differential phase Φ on the distance library is calculated DP Horizontal texture

[0070] Among them, Φ DPi，jrepresents the differential phase of azimuth index i and radial index j, Φ DPi，j+1 Represents the differential phase of azimuthal index i and radial index j+1.

[0071] Step 220, using fuzzy logic method according to the four feature quantities on each distance library, namely the horizontal texture T dBZ , the degree of variation between libraries S PIN , horizontal texture T ZDR and horizontal texture Check whether the distance library is a ground object echo. Including:

[0072] (1) The horizontal texture T on the distance library dBZ Enter membership functions in fuzzy logic Get T dBZ Membership degree f1(T dBZ ). That is, the horizontal texture T dBZ The closer it is to 20, the more likely it is a ground echo. dBZ If it is greater than 20, it is considered to be ground echo. Figure 3 As shown in (a) in .

[0073] (2) The inter-reservoir variation degree S on the distance reservoir PIN Enter membership functions in fuzzy logic Get S PIN Membership degree f2(S PIN ). That is, the degree of variation between libraries S PIN The closer it is to 30%, the greater the possibility that it is a ground object echo. The degree of variation between libraries S PIN If it is greater than 30%, it is considered to be ground echo. The membership function is as follows: Figure 3 As shown in (b) in .

[0074] (3) The horizontal texture T on the distance library ZDR Enter membership functions in fuzzy logic Get T ZDR Membership degree f3(T ZDR ). That is, the horizontal texture T ZDR If it is greater than 2, the possibility of ground object echo is greater. ZDR If it is greater than 2, it is considered to be ground echo. The membership function is as follows: Figure 3 As shown in (c) in .

[0075] (4) The horizontal texture on the distance library Enter membership functions in fuzzy logic get Membership That is, horizontal texture Between 0 and 10, it is not considered as ground echo, horizontal texture The closer it is to 30, the more likely it is a ground echo. If it is greater than 30, it is considered to be ground echo. The membership function is as follows: Figure 3 As shown in (d) in .

[0076] (5) T on the distance library dBZ Membership degree f1(T dBZ ), S PIN Membership degree f2(S PIN ), T ZDR Membership degree f3(T ZDR )and Membership Perform weighted calculation to obtain the comprehensive membership of the distance library α1, α2, α3 and α4 are the weights corresponding to each membership degree, and can be customized.

[0077] (6) When the comprehensive membership of the distance library F>F thd When , the distance library is determined to be a ground object echo, otherwise it is determined to be a precipitation echo. thd is the preset membership threshold.

[0078] Step 230, after each distance library is identified and detected according to step 210 and step 220, all distance libraries for ground object echoes in the volume scanning base data are removed, and the data preprocessing of the volume scanning base data is completed to obtain precipitation echo base data.

[0079] In the above steps 2 and 3, it is necessary to perform detection in four aspects: radar detection power, radial continuity detection result, initial differential phase stability, and layered cloud precipitation echo. The following embodiments introduce these four aspects respectively as follows:

[0080] 1. Radar detection power

[0081] The minimum measurable reflectivity factor Z of all radial corresponding distance libraries at different elevation angles in the statistical precipitation echo base data H , and obtain the minimum measurable reflectivity factor Z H Distribution results with distance. When the minimum measurable reflectivity factor Z H When the distribution result with distance meets the reflectivity factor distribution requirement, it is determined that the radar detection power is qualified; otherwise, it is determined that the radar detection power is unqualified.

[0082] In determining the minimum measurable reflectivity factor Z H When the distribution results with distance meet the reflectivity factor distribution requirements, it is possible to detect whether the overall distribution meets the corresponding requirements. Or detect the minimum measurable reflectivity factor Z at a predetermined distance.H Whether the corresponding requirements are met, such as the minimum measurable reflectivity factor Z at 100 km commonly used in practical applications H Whether the corresponding requirements are met.

[0083] 2. Radial continuity test results

[0084] Please refer to Figure 4 The flowchart shown includes the following processes:

[0085] 1. Calculate the reflectivity factor Z in the precipitation echo base data H The first-order difference Z in the radial direction H_diff , differential reflectivity Z DR The first-order difference Z in the radial direction DR_diff and the differential phase Φ DP The first-order difference in the radial direction Φ DP_diff .

[0086] 2. When the radial direction Z H_diff are less than or equal to the corresponding threshold Z thd , and radially Z DR_diff Are less than or equal to the corresponding threshold ZDR thd , and radially Φ DP_diff are less than or equal to the corresponding threshold Φ thd When the radial continuity test result is qualified.

[0087] Similarly, Figure 4 In order to show clearly, the process is to detect Z in sequence. H_diff , Z DR_diff , Φ DP_diff For illustration, there is actually no specific order for these three detection steps and they can be executed simultaneously.

[0088] 3. When there is Z in the radial direction H_diff >Z thd , or there is Z in the radial direction DR_diff >ZDR thd , or there is Φ in the radial direction DP_diff >Φ thd , it is determined that the radial continuity test result is unqualified.

[0089] Since the all-solid-state dual-polarization weather radar uses a multi-pulse combination transmission mode to process echoes: medium pulses are used to fill the blind spots of long pulses, and short pulses are used to fill the blind spots of medium pulses. Discontinuities will appear at the splicing points of different pulses, so this radial discontinuity may be caused by pulse splicing. The pulse splicing position is at a fixed position preset by the system and can be predetermined. If it is determined that the radial continuity test result is unqualified, further:

[0090] (1) When there is a Z H_diff Greater than the corresponding threshold Z thd , and is greater than the threshold Z thd Z H_diff When the corresponding reflectivity factor jump area is located at the pulse splicing position, it indicates that it is a normal reflectivity factor jump caused by pulse splicing. H Correction is made so that the radial Z H_diff are less than or equal to the corresponding threshold Z thd .

[0091] In one embodiment, the correction method includes determining the corrected reflectivity factor of the reflectivity factor jump region at the pulse splicing position as Z H (after)-Z H (front). Among them. Z H (after) is the average reflectivity factor of several distance bins after the pulse splicing position, Z H (front) is the average of the reflectivity factors of several distance bins before the pulse splicing position.

[0092] However, when the reflectivity factor jump region is not at the pulse splicing position, it indicates that the reflectivity factor jump region is abnormal. Then, the reflectivity factor jump region is output to indicate the abnormal situation for further inspection.

[0093] (2) When there is Z in the radial direction DR_diff Greater than the corresponding threshold ZDR thd , and is greater than the threshold ZDR thd Z DR_diff When the corresponding differential reflectivity jump area is located at the pulse splicing position, it indicates that it is a normal differential reflectivity jump caused by pulse splicing. The differential reflectivity Z of the differential reflectivity jump area is DR Correction is made so that the radial Z DR_diff Are less than or equal to the corresponding threshold ZDR thd .

[0094] In one embodiment, the correction method includes determining the corrected differential reflectivity of the differential reflectivity jump region at the pulse splicing position as Z DR (after)-Z DR (front), Z DR (after) is the average of the differential reflectivity of several distance bins after the pulse splicing position, Z DR (front) is the average of the differential reflectivity of several distance bins before the pulse splicing position.

[0095] Similarly, when the differential reflectivity jump region is not at the pulse splicing position, the abnormal differential reflectivity jump region is output.

[0096] (3) When there is Φ in the radial direction DP_diff Greater than the corresponding threshold Φ thd , and is greater than the threshold Φ thd Φ DP_diff When the corresponding differential phase jump area is located at the pulse splicing position, it indicates that it is a normal differential phase jump caused by pulse splicing. Then, the differential phase Φ of the differential phase jump area is DP Correction is made so that the radial Z H_diff are less than or equal to the corresponding threshold Z thd .

[0097] In one embodiment, the correction method includes determining the corrected differential phase of the differential phase jump region at the pulse splicing position as Φ DP (after)-Φ DP (front), Φ DP (after) is the average of the differential phases of several distance bins after the pulse splicing position, Φ DP (front) is the average of the differential phases of several range bins before the pulse splicing position.

[0098] Similarly, when the differential phase jump region is not at the pulse splicing position, the differential phase jump region is output.

[0099] If the radial continuity test result is determined to be unqualified, at least one of the reflectivity factor jump area, the differential reflectivity jump area and the differential phase jump area is also output to indicate the radial discontinuity area. If the radial continuity test result is determined to be unqualified, when it is determined that the radial discontinuity problem is caused by pulse splicing, it can be directly corrected, and when it is determined that the radial discontinuity problem is caused by an abnormal phenomenon, the abnormal area can be indicated according to the output jump area.

[0100] 3. Initial Differential Phase Stability

[0101] In each radial direction of each layer of elevation angle, the average value of the differential phases of the last several distance bins of the radial effective precipitation echo starting point is used as the radial initial differential phase. Each radial effective precipitation echo starting point needs to meet two conditions: the correlation coefficients of the continuous K distance bins searched from the effective precipitation echo starting point are greater than the correlation coefficient threshold, and the standard deviation of the differential phases of the continuous K distance bins is less than the predetermined threshold. In one example, K = 9, the correlation coefficient threshold is 0.9, and the predetermined threshold is 3.

[0102] Then the median of all radial initial differential phases of each layer elevation angle is calculated to obtain the initial differential phase of each layer elevation angle. Finally, the average value of the initial differential phases of all layers elevation angles in the precipitation echo base data is calculated to obtain the initial differential phase of the precipitation echo base data.

[0103] After obtaining the initial differential phase of each set of precipitation echo base data, the standard deviation of the initial differential phase of several consecutive sets of volume scan base data is calculated to obtain the initial differential phase stability. When the standard deviation is less than the standard deviation threshold, the initial differential phase stability is determined to be qualified, otherwise the initial differential phase stability is determined to be unqualified.

[0104] 4. Stratiform Cloud Precipitation Echo

[0105] The precipitation echo base data includes convective cloud precipitation echoes and stratiform cloud precipitation echoes. Therefore, it is first necessary to identify the stratiform cloud precipitation echoes from the precipitation echo base data. In one embodiment, the reflectivity factor of each point is used to determine whether the point belongs to a convective cloud point or a stratiform cloud point, including the following steps. Please refer to Figure 5 Flowchart:

[0106] All reflectivity factors Z H Points with a value ≥ Thd1 are classified as convective cloud points, where Thd1 is a preset threshold value and is set to 40 dB in this example.

[0107] For the reflectivity factor Z H Any point <Thd1 is identified according to (1) to (5) as follows:

[0108] (1) When there is Z within the range of radius R1 with this point as the center H ≥Thd1, that is, when there is a convective cloud point that has been determined within the range, it is determined that the point also belongs to the convective cloud point. R1 is the predetermined threshold and is also 5km in the example.

[0109] (2) When the reflectivity factor Z of all points within the range of radius R1 with this point as the center H When Thd1 is less than 1, the average reflectivity factor of all points within the range of radius R2 with this point as the center is calculated as the background field average intensity A υ R2 is also a predetermined threshold and is generally larger than R1, and in this example is taken as 12 km.

[0110] (3) When A υ ≤0 and the reflectivity factor Z of the point H -A υ ≥Thd2, it is determined that the point belongs to the convective cloud point. υ ≤0 and the reflectivity factor Z of the point H -A υWhen <Thd2, the point is determined to be a layered cloud point. Thd2 is a preset threshold and is also taken as 10dBZ in the example.

[0111] (4) When 0<A υ <Thd3 and the reflectivity factor of the point When 0<A υ <Thd3 and the reflectivity factor of the point , it is determined that the point belongs to a layered cloud point.

[0112] (5)When A υ ≥Thd3 and the reflectivity factor Z of the point H -A υ ≥0, the point is determined to belong to the convective cloud point. υ ≥Thd3 and the reflectivity factor Z of the point H -A υ When Thd3 is less than Thd1, it is determined that the point belongs to a layered cloud point. Thd3 is a preset threshold value and is less than Thd1, for example, it can be taken as 38dBZ in the example.

[0113] Reflectivity factor Z H After any point of Thd1 is identified according to (1) to (5) above, the reflectivity factor Z is further considered. H <Thd1 convection cloud point influence radius, reflectivity factor Z H All other points within the influence radius of any convective cloud point <Thd1 are also corrected and attributed to convective cloud points. After the correction is completed, the stratiform cloud precipitation echo is obtained by comprehensively identifying all the stratiform cloud points finally determined, and all other convective cloud points constitute the convective cloud precipitation echo.

[0114] Among them, the larger the reflectivity factor of each convective cloud point, the larger the influence radius of the convective cloud point. The influence radius corresponding to the values ​​of different reflectivity factors can be customized in advance. For example, in one example, when the reflectivity factor of the convective cloud point is less than 10dBZ, the corresponding influence radius is 0, that is, the influence radius is not considered. When the reflectivity factor of the convective cloud point is in the range of 10dBZ to 25dBZ, the corresponding influence radius is 1km. When the reflectivity factor of the convective cloud point is in the range of 25dBZ to 30dBZ, the corresponding influence radius is 2km. When the reflectivity factor of the convective cloud point is in the range of 30dBZ to 35dBZ, the corresponding influence radius is 3km. When the reflectivity factor of the convective cloud point is in the range of 35dBZ to 40dBZ, the corresponding influence radius is 4km. When the reflectivity factor of the convective cloud point is greater than 40dBZ, the corresponding influence radius is 5km.

[0115] After the layered cloud precipitation echo is identified, the numerical distribution of each detection quantity of the layered cloud precipitation echo is statistically analyzed, and the detection quantity includes signal-to-noise ratio, pre-filter reflectivity, reflectivity factor, radial velocity, velocity spectrum width, differential reflectivity, correlation coefficient, differential phase and differential phase shift rate. When the numerical distribution of each detection quantity is within the corresponding preset distribution range, it is determined that the detection quantity of the layered cloud precipitation echo meets the corresponding requirements, otherwise it is determined that the detection quantity of the layered cloud precipitation echo does not meet the corresponding requirements.

[0116] The above is only a preferred embodiment of the present application, and the present application is not limited to the above embodiments. It is understood that other improvements and changes directly derived or associated by those skilled in the art without departing from the spirit and concept of the present application should be considered to be included in the protection scope of the present application.

Claims

1. A method for automatically analyzing the quality of dual-polarization weather radar base data, characterized in that: The dual polarization weather radar base data quality automatic analysis method comprises: The fuzzy logic method is used to identify and remove the ground object echo in the volume scanning base data of the dual polarization weather radar, and the data preprocessing of the volume scanning base data is completed to obtain the precipitation echo base data; According to the reflectivity factor Z in the precipitation echo base data H , differential reflectivity Z DR , differential phase Φ DP Obtaining radar signal basic indicators and identifying layered cloud precipitation echoes from precipitation echo base data, wherein the radar signal basic indicators include radar detection power, radial continuity detection results, and initial differential phase stability; When all the basic indicators of radar signals are qualified and the detection amount of the layered cloud precipitation echo meets the corresponding requirements, it is determined that the data quality of the volume scanning base data of the dual polarization weather radar meets the standard; otherwise, it is determined that the data quality of the volume scanning base data of the dual polarization weather radar does not meet the standard; Detecting whether the detection quantity of the layered cloud precipitation echo meets the corresponding requirements includes: statistically analyzing the numerical distribution of each detection quantity of the layered cloud precipitation echo, the detection quantity includes signal-to-noise ratio, reflectivity before filtering, reflectivity factor, radial velocity, velocity spectrum width, differential reflectivity, correlation coefficient, differential phase and differential phase shift rate; when the numerical distribution of each detection quantity is within the corresponding preset distribution range, it is determined that the detection quantity of the layered cloud precipitation echo meets the corresponding requirements, otherwise it is determined that the detection quantity of the layered cloud precipitation echo does not meet the corresponding requirements.

2. The automatic analysis method of dual polarization weather radar base data quality according to claim 1 is characterized in that: Obtaining radar detection power and testing whether it is qualified includes: The minimum measurable reflectivity factor Z of all radial corresponding distance libraries at different elevation angles in the statistical precipitation echo base data H , and obtain the minimum measurable reflectivity factor Z H Distribution results with distance; When the minimum measurable reflectivity factor Z H When the distribution result with distance meets the reflectivity factor distribution requirement, it is determined that the radar detection power is qualified; otherwise, it is determined that the radar detection power is unqualified.

3. The automatic analysis method of dual polarization weather radar base data quality according to claim 1 is characterized in that: Obtaining radial continuity test results and testing whether they are qualified include: Calculate the reflectivity factor Z in the precipitation echo base data H The first-order difference Z in the radial direction H_diff , differential reflectivity Z DR The first-order difference Z in the radial direction DR_diff and the differential phase Φ DP The first-order difference in the radial direction Φ DP_diff ; When the radial direction Z H_diff are less than or equal to the corresponding threshold Z thd , and radially Z DR_diff Are less than or equal to the corresponding threshold ZDR thd , and radially Φ DP_diff are less than or equal to the corresponding threshold Φ thd When the radial continuity test result is qualified, it is determined that the radial continuity test result is qualified; otherwise, it is determined that the radial continuity test result is unqualified.

4. The automatic analysis method of dual polarization weather radar base data quality according to claim 3 is characterized in that: The dual polarization weather radar base data quality automatic analysis method also includes: When there is Z in the radial direction H_diff Greater than the corresponding threshold Z thd , and is greater than the threshold Z thd Z H_diff When the corresponding reflectivity factor jump region is located at the pulse splicing position, the reflectivity factor Z of the reflectivity factor jump region is H Correction is made so that the radial Z H_diff are less than or equal to the corresponding threshold Z thd ; When the reflectivity factor jump region is not at the pulse splicing position, the reflectivity factor jump region is output; When there is Z in the radial direction DR_diff Greater than the corresponding threshold ZDR thd , and is greater than the threshold ZDR thd Z DR_diff When the corresponding differential reflectivity jump region is located at the pulse splicing position, the differential reflectivity Z of the differential reflectivity jump region is DR Correction is made so that the radial Z DR_diff Are less than or equal to the corresponding threshold ZDR thd ; When the differential reflectivity jump region is not at the pulse splicing position, the differential reflectivity jump region is output; When there is Φ in the radial direction DP_diff Greater than the corresponding threshold Φ thd , and is greater than the threshold Φ thd Φ DP_diff When the corresponding differential phase jump region is located at the pulse splicing position, the differential phase Φ of the differential phase jump region is DP Correction is made so that the radial Z H_diff are less than or equal to the corresponding threshold Z thd ; When the differential phase jump region is not at the pulse splicing position, the differential phase jump region is output; When it is determined that the radial continuity test result is unqualified, at least one of a reflectivity factor jump area, a differential reflectivity jump area and a differential phase jump area is also output.

5. The automatic analysis method of dual polarization weather radar base data quality according to claim 4 is characterized in that: Reflectivity factor Z in the reflectivity factor jump region H , the differential reflectivity Z in the differential reflectivity jump region DR 、Differential phase Φ in the differential phase jump region DP The amendments include: Determine the corrected reflectivity factor of the reflectivity factor jump region at the pulse splicing position as Z H (after)-Z H (front), Z H (after) is the average reflectivity factor of several distance bins after the pulse splicing position, Z H (front) is the average of the reflectivity factors of several distance bins before the pulse splicing position; Determine the corrected differential reflectivity of the differential reflectivity jump region at the pulse splicing position as Z DR (after)-Z DR (front), Z DR (after) is the average of the differential reflectivity of several distance bins after the pulse splicing position, Z DR (front) is the average of the differential reflectivities of several range bins before the pulse splicing position; Determine the corrected differential phase of the differential phase jump region at the pulse splicing position as Φ DP (after)-Φ DP (front), Φ DP (after) is the average of the differential phases of several distance bins after the pulse splicing position, Φ DP (front) is the average of the differential phases of several range bins before the pulse splicing position.

6. The automatic analysis method of dual polarization weather radar base data quality according to claim 1 is characterized in that: Obtaining the initial differential phase stability and testing whether it is qualified includes: In each radial direction of each layer of elevation angle, the average value of the differential phases of the last several distance bins of the effective precipitation echo starting point in the radial direction is used as the initial differential phase of the radial direction, and the correlation coefficients of the continuous K distance bins searched outward from the effective precipitation echo starting point in each radial direction are greater than the correlation coefficient threshold and the standard deviation of the differential phases of the continuous K distance bins is less than the predetermined threshold; Calculate the median value of all radial initial differential phases of each layer of elevation angle to obtain the initial differential phase of each layer of elevation angle; Calculate the average value of the initial differential phases of all layer elevation angles in the precipitation echo base data to obtain the initial differential phase of the precipitation echo base data; The standard deviation of the initial differential phase of several consecutive groups of volume scan basis data is calculated to obtain the initial differential phase stability. When the standard deviation is less than the standard deviation threshold, the initial differential phase stability is determined to be qualified, otherwise the initial differential phase stability is determined to be unqualified.

7. The automatic analysis method of dual polarization weather radar base data quality according to claim 1 is characterized in that: Identifying stratiform cloud precipitation echoes from precipitation echo base data includes: For the reflectivity factor Z H Any point of Thd1: When there is a Z within the range of radius R1 with the point as the center H ≥Thd1, the point is determined to belong to the convective cloud point; when the reflectivity factor Z of all points within the range of radius R1 with the point as the center H When Thd1 is less than Thd1, the average reflectivity factor of all points within the range of radius R2 with the point as the center is calculated as the background field average intensity A v ; when A v ≤0 and the reflectivity factor Z of the point H -A v ≥Thd2, it is determined that the point belongs to a convective cloud point. v ≤0 and the reflectivity factor Z of the point H -A v <Thd2, it is determined that the point belongs to a layered cloud point; when 0 v <Thd3 and the reflectivity factor of the point When , it is determined that the point belongs to the convective cloud point, when 0 v <Thd3 and the reflectivity factor of the point When A v ≥Thd3 and the reflectivity factor Z of the point H -A v ≥0, it is determined that the point belongs to a convective cloud point. v ≥Thd3 and the reflectivity factor Z of the point H -A v When <0, it is determined that the point belongs to a layered cloud point;​​ The reflectivity factor Z H All other points within the influence radius of any convective cloud point <Thd1 are corrected and attributed to convective cloud points. The larger the reflectivity factor of each convective cloud point, the larger the influence radius of the convective cloud point. The stratiform cloud precipitation echo is obtained by comprehensively identifying all stratiform cloud points.

8. The automatic analysis method of dual polarization weather radar base data quality according to claim 1 is characterized in that: The fuzzy logic method is used to identify and remove ground object echoes from the volume scanning data of dual polarization weather radar, including: For each distance library in the volume scan base data, Calculate the reflectivity factor Z on the distance library H Horizontal texture T dBZ ,according to Calculate the reflectivity factor Z on the distance library H The radial variation between the libraries S PIN ,according to The differential reflectivity Z on the distance library is calculated DR Horizontal texture T ZDR ,according to The differential phase Φ on the distance library is calculated DP Horizontal texture Among them, Z i，j Represents the reflectivity factor of azimuthal index i and radial index j, Z i，j+1 represents the reflectivity factor of azimuthal index i and radial index j+1, Z thresh is the threshold for intensity change between distances, ZDR i，j The differential reflectivity of azimuthal index i and radial index j, ZDR i，j+1 represents the differential reflectivity of azimuthal index i and radial index j+1, Φ DPi，j represents the differential phase of azimuth index i and radial index j, Φ DPi，j+1 represents the differential phase of azimuth index i and radial index j+1, N A The total number of indices representing the orientation, N R Indicates the total number of radial indices; The fuzzy logic method is used to calculate the horizontal texture T on each distance library. dBZ , the degree of variation between libraries S PIN , horizontal texture T ZDR and horizontal texture Detecting whether the distance library is a ground object echo; All distance libraries of ground object echoes in the volume scanning base data are removed, and the data preprocessing of the volume scanning base data is completed to obtain precipitation echo base data.

9. The automatic analysis method of dual polarization weather radar base data quality according to claim 8, characterized in that: The fuzzy logic method is used to calculate the horizontal texture T on each distance library. dBZ , the degree of variation between libraries S PIN , horizontal texture T ZDR and horizontal texture Detecting whether the distance library is a ground object echo includes: The horizontal texture T on the distance library dBZ Enter membership functions in fuzzy logic Get T dBZ Membership degree f1(T dBZ ); The inter-library variation degree S of the distance library PIN Enter membership functions in fuzzy logic Get S PIN Membership degree f2(S PIN ); The horizontal texture T on the distance library ZDR Enter membership functions in fuzzy logic Get T ZDR Membership degree f3(T ZDR ); Set the distance library on the level texture Enter membership functions in fuzzy logic get Membership The distance library T dBZ Membership degree f1(T dBZ ), S PIN Membership degree f2(S PIN ), T ZDR Membership degree f3(T ZDR )and Membership Performing weighted calculation to obtain the comprehensive membership of the distance library; When the comprehensive membership of the distance library is greater than the membership threshold, the distance library is determined to be a ground object echo.

Images