X-band phased array weather radar multi-beam scanning mode observation error calibration method and system based on statistical result

Abstract

The invention discloses a statistical result-based X-band phased array weather radar multi-beam scanning mode observation error calibration method and system. The method comprises the following steps of: S1, acquiring observation base data of an S-band dual-polarization Doppler weather radar and an X-band dual-polarization phased array weather radar as initial data; s2, performing data quality control on the radar-based data to obtain preprocessed data; and S3, by adopting a different radar time and space sampling volume coordinate matching method and taking an observation value of the S-band radar as a reference, counting a matching difference of the preprocessed data to obtain an observation error of the X-band radar and correcting the observation error.

Description

Technical Field

[0001] The present invention relates to the field of phased array technology, and particularly relates to a method and system for calibrating observation errors of a multi-beam scanning mode of an X-band phased array weather radar based on statistical results. Background Art

[0002] Disastrous weather such as heavy precipitation, thunderstorm gales, tornadoes and downbursts occurs frequently in China, but the operational forecasting ability is relatively low, which is an urgent problem to be solved in the field of weather forecasting. At present, conventional weather radar equipment at home and abroad is restricted by the spatio-temporal resolution and the observation blind area, and cannot meet the detection requirements for rapidly changing weather systems, resulting in insufficient forecasting and warning capabilities for disastrous weather. The application of phased array technology in weather radar detection can greatly improve the time resolution of detection data and thus improve the radar warning efficiency. The phased array dual-polarization weather radar has achieved preliminary application results in China. Wu Chong et al. (2014) carried out a quantitative comparative analysis of an S-band phased array weather radar (S-PAR) and a Doppler weather radar, and proposed a method for matching radar reflectivity factors with different resolutions at different geographical locations and a quantitative comparison method for observation data; at the same time, taking the observation data of a C-band dual-polarization radar (C-POL) as a reference, the influence of different width scanning beams of an X-band phased array radar (X-PAR) on the echo intensity was compared. In 2015, Liu Jun et al. further verified the reliability of the echo data of the phased array meteorological radar by analyzing the comparison of the observation data of the phased array meteorological radar and the X-band mechanical antenna meteorological radar. Liu Liping et al. (2015) carried out test calibration and field observation experiments of the X-PAR, aiming to improve the quality and detection ability of radar detection data. In addition, multiple X-PARs can also be used for joint network observation to generate radar mosaic results as an effective supplementary means for operational radars.

[0003] However, in practical applications, X-band phased array radars have problems such as unstable data quality, complex quality control algorithms, and room for improvement in quality control effects, which pose new challenges to the application of phased array weather radars. Therefore, when integrating phased array weather radar data into the operational radar observation network, strict calibration is required to ensure the effectiveness of operational applications. Currently, Guangdong Province has deployed a high-density radar observation network consisting of more than 50 X-band phased array dual-polarization weather radars to make up for the shortcomings of S-band operational radars, improve the temporal and spatial resolution of radars, and enhance the monitoring and early warning capabilities for severe convective weather. The X-PARs in the network adopted a single-beam scanning mode with narrow transmit and narrow receive in the initial stage of construction. In practical applications, to meet the refined observation requirements for mesoscale systems, some radars were upgraded to a multi-beam scanning mode with wide transmit and narrow receive starting from 2022. The technology of "wide transmit and narrow receive" can achieve the emission of a wide beam of 4.5 - 6.3° and the simultaneous reception of 5 - 7 elevation channels, increasing the observation elevation angle from a resolution of 1.8° and 16 layers to a resolution of 0.9° and 68 layers while keeping the volume scan time unchanged. The multi-beam scanning mode significantly improves the resolution of data and reduces the static cone blind area of the radar, but also brings problems such as inconsistent radar calibration methods, changes in meteorological radar equation parameters, and broadened beams.

[0004] Currently, there is little research on the sensitivity, changes in observation errors, etc. of the multi-beam scanning mode of X-PARs in Guangdong Province, and further analysis is needed. At the same time, X-PARs have problems such as unstable data quality, complex quality control algorithms, and room for improvement in quality control effects in practical applications, which pose new challenges to the application of phased array weather radars. Therefore, when integrating X-PAR data into the operational radar observation network, strict calibration is required to ensure the effectiveness of operational applications. Summary of the Invention

[0005] To solve the above technical problems, the present invention provides a calibration method based on the multi-beam scanning mode of X-band phased array weather radars, and the method includes:

[0006] Step S1: Collect the observation basic data of S-band dual-polarization Doppler weather radar S-POL and X-band dual-polarization phased array weather radar X-PAR as initial data;

[0007] Step S2: Perform data quality control on the radar basic data to obtain preprocessed data;

[0008] Step S3: Adopt different radar time and spatial sampling volume coordinate matching methods, and based on the observed values of S-POL, statistically match the differences of the preprocessed data to obtain the observation errors of X-PAR and correct them.

[0009] Optionally, in step S1, the initial data includes the observed basic data of S-POL and the observed basic data of X-PAR;

[0010] Among them, the operating frequency of S-POL is 2885 MHz, the peak power of the transmitter is greater than or equal to 650 kW, the maximum detection range is 460 km, the range resolution is 250 m, the horizontal beam width is 0.96°, and the vertical beam width is 0.65°;

[0011] The maximum detection range of the X-PAR is 60 km, the range resolution is 30 m, the horizontal beam width is 1.2°, the vertical beam width is 0.9°, the AXPT0364 type dual-polarization planar microstrip patch array antenna is adopted, the antenna array surface uses a rectangular aperture, the elevation angle of the antenna array surface is fixed at 15°, there are 64 rows of dual-polarization microstrip patch linear arrays in the pitch direction, and each row of linear arrays in the azimuth direction has 32 radiation units.

[0012] Optionally, in step S2, the process of performing data quality control on the radar basic data specifically includes:

[0013] Select the reflectivity factor Z from the initial data H greater than 15 dB, the differential reflectivity factor Z DR greater than or equal to 0 and less than 5 dB, the zero-lag cross-correlation coefficient CC greater than 0.9, and the signal-to-noise ratio SNR greater than 25 dB for consistency comparison analysis;

[0014] For the quality control of the two-way differential propagation phase difference Φ DP the initial phase is determined by the 3 km short-range echo method, and the 1 km median filtering is used to eliminate the noise information;

[0015] The observed parameters of the S-band and X-band radars are calculated respectively by using the raindrop size distribution meter observation data, and an exponential fit is performed on Z H and a polynomial fit is performed on Z DR The observed parameters of S-POL and X-PAR are interchanged so that the observed results of the S-band and X-band radars after interchange for the same target can be compared, and the data quality control of the radar basic data is completed.

[0016] Optionally, in step S3, the statistical process of the observation error specifically includes:

[0017] The single sampling volume of S-POL is equivalent to a frustum of a pyramid, and the elevation angle, azimuth angle, and slant range spherical coordinate information f of the 8 vertices of the frustum of the pyramid are calculated respectively i 0 of f i 0 (el i 0, az i 0 , r i 0 ), where \(i = 1\sim8\);

[0018] For each center point \(f\) of the sampling volume of X - PAR X The coordinates are converted from the spherical coordinates \(f\) based on X - PAR X (el X , az X , r X ) into longitude, latitude, and altitude information \(D\) in the Cartesian coordinate system X (lon X , lat X , h X );

[0019] Taking the S - POL radar as the origin of the spherical coordinate system, the Cartesian coordinates of X - PAR are inversely calculated into the S - POL spherical coordinate system to obtain the spherical coordinates \(f\) when the S - POL radar observes the X - PAR data points S (el S , az S , r S );

[0020] Establish a one - to - one correspondence relationship between the observation data of two radars with different positions. Based on the correspondence relationship, compare the position relationship between the center point of the X - PAR sampling volume and the 8 vertices of the frustum of a pyramid of the S - POL radar sampling volume. For the \(f\) S (el S , az S , r S ) that falls within the \(f\) i 0 (el i 0 , az i 0 , r i 0 ) range, the data is considered observation data with consistent spatial matching;

[0021] Extract the acquisition time of each radar radial data. Considering the antenna rotation speeds of S - POL and X - PAR, the matching points with a radial data acquisition time difference within ±30 s are the final observation data with consistent spatio - temporal matching;

[0022] In order to quantitatively analyze the data quality of X - PAR, taking the S - POL radar observation as the accurate value, the minimum measurable echo intensity \(Z\) of X - PAR in the multi - beam scanning mode after matching min 、\(Z\) H 、\(Z\) DR 、\(K\) DP are statistically analyzed to analyze the differences in X - PAR observation data and obtain the observation errors.

[0023] The present invention also discloses an observation error calibration system for the multi-beam scanning mode of an X-band phased array weather radar based on statistical results. The system includes:

[0024] A data acquisition module, configured to collect the observation basic data of an S-band dual-polarization Doppler weather radar S-POL and an X-band dual-polarization phased array weather radar X-PAR as initial data;

[0025] A data preprocessing module, configured to perform data quality control on the radar basic data to obtain preprocessed data;

[0026] A data calibration module, configured to adopt different radar time and spatial sampling volume coordinate matching methods, and based on the observation values of the S-POL, statistically match the differences of the preprocessed data to obtain the observation errors of the X-PAR and correct them.

[0027] Optionally, in the data acquisition module, the initial data includes the observation basic data of the S-POL and the observation basic data of the X-PAR;

[0028] Among them, the operating frequency of the S-POL is 2885 MHz, the peak power of the transmitter is greater than or equal to 650 kW, the maximum detection range is 460 km, the range resolution is 250 m, the horizontal beam width is 0.96°, and the vertical beam width is 0.65°;

[0029] The maximum detection range of the X-PAR is 60 km, the range resolution is 30 m, the horizontal beam width is 1.2°, the vertical beam width is 0.9°, it adopts an AXPT0364 type dual-polarization planar microstrip patch array antenna, the antenna array surface uses a rectangular aperture, the elevation angle of the antenna array surface is fixed at 15°, there are 64 rows of dual-polarization microstrip patch linear arrays in the pitch direction, and each row of linear arrays in the azimuth direction has 32 radiation units.

[0030] Optionally, in the data preprocessing module, the process of performing data quality control on the radar basic data specifically includes:

[0031] Select the reflectivity factor Z H greater than 15 dB, the differential reflectivity factor Z DR greater than or equal to 0 and less than 5 dB, the zero-lag cross-correlation coefficient CC greater than 0.9, and the signal-to-noise ratio SNR greater than 25 dB data to participate in the consistency comparison analysis;

[0032] For the quality control of the two-way differential propagation phase difference Φ DP adopt the 3-km short-range echo method to determine the initial phase, and use the 1-km median filtering to remove the noise information;

[0033] Using the raindrop spectrometer observation data, calculate the observation parameters of S-band and X-band radars respectively, and perform an exponential fit on Z H and perform a polynomial fit on Z DR Exchange the observation parameters of S-POL and X-PAR so that the observation results of the S-band and X-band radars after the exchange for the same target can be compared, and complete the data quality control of the radar base data.

[0034] Optionally, in the data calibration module, the statistical process of the observation error specifically includes:

[0035] Equivalent the single sampling volume of S-POL to a frustum of a pyramid, and calculate the elevation angle, azimuth angle, and slant range spherical coordinate information f i 0 of the 8 vertices of the frustum of the pyramid, where i = 1 to 8; i 0 (el i 0 ,az i 0 ,r i 0 )

[0036] Convert the center point f X coordinates of each sampling volume of X-PAR from the spherical coordinates based on X-PAR f X (el X ,az X ,r X ) to the longitude, latitude, and height information D X (lon X ,lat X ,h X ) in the Cartesian coordinate system;

[0037] Taking the S-POL radar as the origin of the spherical coordinates, back-calculate the Cartesian coordinates of X-PAR to the S-POL spherical coordinate system to obtain the spherical coordinates f S (el S ,az S ,r S ) when the S-POL radar observes the X-PAR data point;

[0038] Establish a one-to-one correspondence between the observation data of the two radars with different positions, compare the positional relationship between the center point of the X-PAR sampling volume and the 8 vertices of the frustum of the pyramid of the S-POL radar sampling volume based on the correspondence, and compare f S (el S ,az S ,r S ) falling within f i 0 (eli 0 , az i 0 , r i 0 Data within the range is considered observation data with consistent spatial matching.

[0039] Extract the acquisition time of each radar radial data. Considering the antenna rotation speeds of S-POL and X-PAR, matching points with a radial data acquisition time difference within ±30 s are the final observation data with consistent spatio-temporal matching.

[0040] To quantitatively analyze the data quality of X-PAR, using the S-POL radar observation as the accurate value, for the minimum detectable echo intensity Z of X-PAR in the multi-beam scanning mode after matching min , Z H , Z DR , K DP Perform statistics, analyze the differences in X-PAR observation data, and obtain the observation error.

[0041] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0042] Due to the multiple observation biases brought about by the dual complexities of phased array and multi-beam technologies, the maintenance and calibration of X-PAR are more cumbersome than those of S-POL. The existing X-PAR observation error calibration method is factory calibration. After the radar production and assembly are completed, the observation errors of each component of the radar are tested in a microwave anechoic chamber and a far-field calibration environment, and the error results are delivered to the user together with the radar. However, the actual observation environment is not exactly the same as the laboratory environment. Due to the influence of many factors such as the temperature, humidity, electromagnetic background, hardware loss, and random error pulsation of the observation environment, the comprehensive observation error of the radar will show unpredictable offsets, and some errors caused by the actual observation environment are difficult to accurately obtain under laboratory conditions. A method for calibrating the observation error of the multi-beam scanning mode of an X-band phased array weather radar based on statistical results according to the present invention uses S-POL, which has been strictly calibrated, operates stably in business, and has relatively high data reliability, as a benchmark, and performs real-time calibration on X-PAR, which has a more complex system composition and relatively low data reliability, based on the statistical results of a long time series of 1 month. Compared with the existing factory calibration method, the calibration link of this method is more comprehensive, there is no need to stop the radar for calibration operations, and the observation error calibration results can be updated in real time and rolled. Description of the Drawings

[0043] To more clearly illustrate the technical solutions of the present invention, the following briefly introduces the drawings required in the embodiments. Obviously, the following-described drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0044] Figure 1 It is a step diagram of a method for calibrating the observation error of the multi-beam scanning mode of an X-band phased array weather radar based on statistical results in an embodiment of the present invention;

[0045] Figure 2 It is a diagram of the relative position distribution and detection range of the radar in an embodiment of the present invention;

[0046] Figure 3 It is a schematic diagram of a method for matching the time and space sampling volume coordinates of different radars in an embodiment of the present invention;

[0047] Figure 4 It is a curve graph of the variation of the difference between X-PAR parameters and S-POL parameters with elevation angle and distance in an embodiment of the present invention; where Figure 4 (a) is the curve of ΔZ H varying with the elevation angle, Figure 4 (b) is the curve of ΔZ DR varying with the elevation angle, Figure 4 (c) is the curve of ΔK DP varying with the elevation angle, Figure 4 (d) is the curve of ΔZ H varying with the distance;

[0048] Figure 5 It is an example of calibrating ΔZ H of the ZG000 radar and an example of calibrating the factory antenna gain in an embodiment of the present invention; where Figure 5 (a) is a diagram of the example of calibrating the radar ΔZ H and the example of calibrating the factory antenna gain of the radar ZG000 in the multi-beam mode at 12:27 on August 25, 2022, Figure 5 (b) is a diagram of the example of calibrating the radar ΔZ H and the example of calibrating the factory antenna gain of the radar ZG000 in the multi-beam mode at 12:33 on August 25, 2022, Figure 5 (c) is a diagram of the example of calibrating the radar ΔZ H and the example of calibrating the factory antenna gain of the radar ZG000 in the multi-beam mode at 12:34 on August 25, 2022, Figure 5 (d) is a diagram of the example of calibrating the radar ΔZ H and the example of calibrating the factory antenna gain of the radar ZG000 in the multi-beam mode at 13:04 on August 25, 2022;

[0049] Figure 6 It is an example of calibrating ΔZ H of the ZG001 radar and an example of calibrating the factory antenna gain in an embodiment of the present invention; where Figure 6(a) Radar ΔZ of radar ZG000 in multi-beam mode at 13:18 on August 25, 2022 H Calibration example and factory antenna gain calibration example diagram; Figure 6 (b) Radar ΔZ of radar ZG000 in multi-beam mode at 13:19 on August 25, 2022 H Calibration example and factory antenna gain calibration example diagram; Figure 6 (c) Radar ΔZ of radar ZG000 in multi-beam mode at 13:20 on August 25, 2022 H Calibration example and factory antenna gain calibration example diagram; Figure 6 (d) Radar ΔZ of radar ZG000 in multi-beam mode at 13:21 on August 25, 2022 H Calibration example and factory antenna gain calibration example diagram. Detailed implementation manners

[0050] To make the above objects, features, and advantages of the present invention more obvious and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific implementation manners.

[0051] Example 1

[0052] A method for calibrating the observation error of the multi-beam scanning mode of an X-band phased array weather radar based on statistical results, as Figure 1 shown, the method includes:

[0053] Step S1: Collect the observation basic data of the S-band dual-polarization Doppler weather radar and the X-band dual-polarization phased array weather radar as initial data.

[0054] In this research and invention, the observation data of S-POL (station number Z9200) deployed in Guangzhou, Guangdong Province and three X-band dual-polarization phased array weather radars (X-PAR, station numbers are ZG000, ZG001, and ZG002) are selected. The relative position distribution and detection range of the 4 radars are as Figure 2 shown.

[0055] The main working parameters of S-POL and X-PAR are shown in Table 1. Z9200 is located at 113.36E, 23.00°N, was built and put into operation in 2001, and completed the dual-polarization upgrade transformation in 2016. Its working frequency is 2885 MHz, the peak power of the transmitter is ≥650 kW, the maximum detection distance is 460 km, the range resolution is 250 m, the horizontal beam width is 0.96°, and the vertical beam width is 0.95°.

[0056] Since the electromagnetic waves of S-band radar are less affected by precipitation attenuation, and Z9200 has strictly calibrated systems such as transmitters, receivers, waveguides, and antennas during its operation, the data reliability is relatively high. Therefore, in this study, the observation results of Z9200 are used as the accurate values to quantitatively compare and analyze the errors of X-PAR.

[0057] The three X-PARs are located in Panyu (ZG000), Huadu (ZG001), and Baiyun (ZG002) respectively. They are all developed by Guangdong Narui Radar Technology Co., Ltd. and were put into operation in July 2017, January 2018, and December 2018 respectively.

[0058] Its maximum detection range is 60 km, and the range resolution is accurate to 30 m. X-PAR can achieve digital multi-beam scanning with wide transmission and narrow reception. The horizontal beam width is 1.2°, the vertical beam width is 0.9°, the elevation coverage range is 0.9 - 61.2°, the elevation resolution is 1.8°, with a total of 68 elevation layers, and the volume scan time is 60 s. It is a fully solid-state coherent, active phased array, pulsed Doppler, dual-polarization weather radar.

[0059] The three X-PARs all adopt the AXPT0364 type dual-polarization planar microstrip patch array antenna. The antenna array surface uses a rectangular aperture. There are 64 rows of dual-polarization microstrip patch linear arrays in the elevation direction, and each row of linear arrays has 32 radiation units in the azimuth direction, with a total of 2048 radiation units. The 64 rows of dual-polarization microstrip patch linear arrays correspond to 64 radio frequency transceiver units, and 8 radio frequency transceiver control units are used to complete the amplitude / phase control of the pulses and environmental monitoring.

[0060] The elevation angle of the antenna array surface is fixed at 15° and does not change with the radar operating mode. X-PAR uses the phased array electronic scanning method in the vertical direction. First, a continuous and non-stop phased array electronic RHI scan is completed at one azimuth angle, then switched to the next azimuth angle, and then continue to complete the non-stop phased array electronic RHI scan. For the VPPI mode of traditional mechanical scanning weather radar, taking the VCP21 mode as an example, the delay of its vertical profile RHI structure data is as high as 6 minutes, which is likely to cause discontinuity of radar echoes in the vertical direction.

[0061] However, the phased array electronic scanning method of X-PAR shortens the observation period to only 1 minute. When the VRHI volume scan mode is used to complete a 360° volume scan, the radar antenna only needs to rotate mechanically for one circle, which not only avoids the delay when traditional mechanical radars switch different elevation angles but also can obtain real-time and accurate RHI data.

[0062] The vertical structure of meteorological processes is an important basis for analyzing and judging weather processes. The high-elevation, high-resolution, and non-delayed RHI data provided by X-PAR provide effective support for the monitoring, early warning, and research of weather processes. The main working parameters of POL and X-PAR are shown in Table 1.

[0063] Table 1

[0064]

[0065] In this embodiment, several precipitation cases in August 2022 are used for quantitative comparative analysis between multiple radars. The general situation of the precipitation cases observed in the radar experiment is shown in Table 2 (Beijing time, the same below).

[0066] Table 2

[0067]

[0068] Step S2: Perform data quality control on the radar base data to obtain preprocessed data.

[0069] Before quantitative analysis of radar base data, it is necessary to eliminate ground clutter interference and clear-sky echoes. Select data sets where Z H is greater than 15 dB, Z DR is greater than or equal to 0 and less than 5 dB, CC is greater than 0.9, and SNR is greater than 25 dB for consistency comparative analysis. Among them, the base data of X-PAR integrates the fixed coefficient "ZPHI" precipitation profile attenuation correction algorithm based on South China raindrop spectrum observations provided by Guangdong Narui Radar Technology Co., Ltd.

[0070] Based on the continuous volume scan results from 19 to 20 on August 25, 2022, the azimuths with severe low-elevation beam blockage are manually extracted and marked as fixed interference azimuths, and the blocked azimuths are removed during data preprocessing. For the quality control of Φ DP , the 3 km close-range echo method is used to determine the initial phase, and 1 km median filtering is used to remove noise information. Due to the scattering property differences of precipitation particles for electromagnetic waves of different wavelengths, the observation results of S-POL and X-PAR for the same target cannot be directly compared. Geng Fei (2023) calculated the S-band and X-band radar observation parameters using the raindrop spectrum observation data of Guangdong Longmen Station respectively, performed exponential fitting on Z H , performed polynomial fitting on Z DR , and interchanged the observation parameters of S-POL and X-PAR so that the observation results of the interchanged S-band and X-band radars for the same target can be compared.

[0071] Step S3: Adopt different radar time and spatial sampling volume coordinate matching methods. Based on the observed values of the S-band radar, statistically match the differences in the preprocessed data to obtain the observation error of the X-band radar and make corrections. Since there are obvious differences in the volume scan modes, geographical locations, sampling volumes, etc. between S-POL and X-PAR, a direct corresponding relationship cannot be established for the observed data. Different radar time and spatial sampling volume coordinate matching methods need to be adopted to achieve coordinate unification. As Figure 3 shown, a single sampling volume of S-POL can be equivalent to a frustum of a pyramid, and the elevation angle, azimuth angle, and slant range spherical coordinate information f i 0 (i = 1 - 8) of its 8 vertices can be calculated respectively. f i 0 (el i 0 , az i 0 , r i 0 ) of the center point of each sampling volume of X-PAR is converted from the spherical coordinates f X based on X-PAR f X (el X , az X , r X ) to the longitude, latitude, and height information D X (lon X , lat X , h X ) in the Cartesian coordinate system. Subsequently, with the S-POL radar as the origin of the spherical coordinates, the Cartesian coordinates of X-PAR are then inversely calculated into the S-POL spherical coordinate system, and the spherical coordinates f S (el S , az S , r S ) when the S-POL radar observes the X-PAR data points are obtained, and a one-to-one corresponding relationship is established for the observed data of the two radars with different positions. Compare the positional relationship between the center point of the X-PAR sampling volume and the 8 vertices of the frustum of the pyramid of the S-POL radar sampling volume, and f S (el S , az S , r S ) falling within f i 0 (el i 0 , az i 0 , r i 0) The data within the range is considered as the observation data with consistent spatial matching. Finally, the acquisition time of each radar radial data is extracted. Considering the antenna rotation speeds of S-POL and X-PAR, the matching points with a radial data acquisition time difference within ±30 s are considered as the final observation data with consistent spatio-temporal matching.

[0072] To quantitatively analyze the data quality of X-PAR, taking the S-POL radar observation as the accurate value, a one-to-one correspondence between the data of X-PAR and S-POL is established library by library to obtain the observation deviation (ΔParameter = Parameter _X - Parameter _S ) and the distribution relationship with the elevation angle, as Figure 4 shown, Figure 4 In it, the circular markers are the parameter differences between radar ZG000 and radar Z9200, and the triangular markers are the parameter differences between radar ZG001 and radar Z9200. In the multi-beam mode, a total of 68 elevation angles are scanned. Due to the deterioration of the antenna parameters at high elevation angles above 40 layers and the further increase of the transmit beam width, and less meteorological echo information at high elevation angles, the data at high elevation angles above 40 layers are discarded. ΔZ H and ΔZ DR both show a periodic change trend with 5 per group of elevation angles, which is consistent with the number of narrow receiving sub-beams within the wide transmit beam in Table 1. The periodic change trend of the observation deviation is consistent with the analysis results of Wu Chong et al. (2014). The reason for the periodic change of the observation deviation is that the actual antenna radiation intensity is unevenly distributed within the wide transmit beam. The normal angle of the X-PAR antenna array is fixed at 15°. When the receiving elevation angle steps 0.9°, the corresponding elevation angle numbers are between 16 and 17, and ΔZ H , ΔZ DR and ΔK DP all have extreme values with relatively small absolute errors at this elevation angle. During the preprocessing process, there may be phenomena such as misidentification or incomplete elimination of the occluded azimuth, resulting in relatively large ΔZ H deviations at the bottom 2 elevation angles. Among them, Figure 4 (a) is the variation curve of ΔZ H with the elevation angle, Figure 4 (b) is the variation curve of ΔZ DR with the elevation angle, Figure 4 (c) is the variation curve of ΔK DP with the elevation angle, Figure 4 (d) is the variation curve of ΔZ H with the distance.

[0073] In this embodiment, the observation deviation of Z H in the multi-beam mode is corrected. It should be noted that in the multi-beam mode, the variation trend of ΔZ H with the detection distance is not obvious (Figure 4 (d)), so in the correction of the X-PAR elevation error in the multi-beam mode, all available azimuth and range bin data at a certain elevation are uniformly processed and calculated, and the distance factor is not a sensitive item for the error in the multi-beam mode.

[0074] This embodiment uses a correction method based on statistical results for the Z of X-PAR H for calibration. All valid precipitation observation cases in August 2022 in Table 2 are used, the differences in X-PAR and S-POL reflectivity factors at each elevation of all observation data are statistically analyzed, and based on the statistical results, the Z of X-PAR H is corrected.

[0075] The solid line in the figure is the curve of the difference in echo intensity between X-PAR and S-POL obtained by the spatio-temporal coordinate matching algorithm versus the X-PAR observation elevation, ΔZ H has 5 obvious periodic fluctuations in each group and the overall deviation is about -3.5 dB. After statistical calibration, the ΔZ H -Elevation curve is as shown by the dash line in Figure 5 , the standard deviation is significantly reduced (about 0.46 dB), and the average deviation of ΔZ H is significantly reduced to about -0.8 dB (Table 3). In addition, this embodiment also compares the error calibration results of the existing factory antenna gain calibration method. The calibration values of the horizontal transmit gain G TH , horizontal receive gain G RH , vertical transmit gain G TV and vertical receive gain G RV of the AXPT0364 type dual-polarized planar microstrip patch array antenna for elevations 1-40 (0.9-36°) calibrated in the microwave anechoic chamber when the radar leaves the factory are used for error correction. The corrected ΔZ H -Elevation curve is as shown by the dotted line in Figure 5 . After the factory antenna gain correction, the data standard deviation is about 0.53 dB, and the average deviation of ΔZ H is about -2.5 dB. Comparing the existing factory antenna gain calibration method and the statistical calibration method of this embodiment, it can be seen that the overall calibration effect of the statistical calibration method of this embodiment is significantly better.

[0076] Figure 5 (a) is the radar ΔZ H calibration example and factory antenna gain calibration example diagram of radar ZG000 in the multi-beam mode at 12:27 on August 25, 2022, Figure 5 (b) is the radar ΔZ HCalibration example and factory antenna gain calibration example diagram Figure 5 (c) Radar ΔZ of radar ZG000 in multi-beam mode at 12:34 on August 25, 2022 H Calibration example and factory antenna gain calibration example diagram Figure 5 (d) Radar ΔZ of radar ZG000 in multi-beam mode at 13:04 on August 25, 2022 H Calibration example and factory antenna gain calibration example diagram.

[0077] Figure 6 And Figure 5 Similarly, 4 cases of radar ZG001 in multi-beam mode are selected. Among them, the solid line is the original uncalibrated curve, the short dash line is the curve after calibration in the embodiment of the present invention, and the dotted line is the curve after factory calibration. The case times are from 13:18 to 13:21 on August 25, 2022. Considering all cases, the factory antenna gain calibration method can appropriately eliminate the periodic fluctuation of ΔZ H and increase the average observation deviation from about -3 dB to about -2 dB, but the calibration effect is not restricted by S-POL data, and the calibration effect will also deteriorate when the attenuation of X-PAR cases is obvious; the statistical calibration method of this embodiment has the best comprehensive performance, and the average deviation of ΔZ H after calibration is about -0.6 dB, and the maximum error is within -1 to -2 dB, and the standard deviation of each elevation angle data is about 0.4 dB. Table 3 shows the evaluation of the existing calibration and the calibration effect of this embodiment for radar ΔZ H of ZG000 and ZG001.

[0078] Figure 6 (a) Radar ΔZ of radar ZG000 in multi-beam mode at 13:18 on August 25, 2022 H Calibration example and factory antenna gain calibration example diagram; Figure 6 (b) Radar ΔZ of radar ZG000 in multi-beam mode at 13:19 on August 25, 2022 H Calibration example and factory antenna gain calibration example diagram; Figure 6 (c) Radar ΔZ of radar ZG000 in multi-beam mode at 13:20 on August 25, 2022 H Calibration example and factory antenna gain calibration example diagram; Figure 6 (d) Radar ΔZ of radar ZG000 in multi-beam mode at 13:21 on August 25, 2022 H Calibration example and factory antenna gain calibration example diagram.

[0079] Table 3

[0080]

[0081] Data unit: dB; *: Case 1 - 4 respectively correspond to Figure 5 the 4 cases in Figure 6 ; **: Case 5 - 8 respectively correspond to

[0082] Embodiment 3

[0083] A calibration system based on the multi - beam scanning mode of an X - band phased array weather radar, specifically: a data acquisition module, which is used to collect the observation basic data of an S - band dual - polarization Doppler weather radar S - POL and an X - band dual - polarization phased array weather radar X - PAR as initial data.

[0084] The present invention selects the observation data of S - POL (station number Z9200) deployed in Guangzhou, Guangdong Province and three X - band dual - polarization phased array weather radars (X - PAR, station numbers are ZG000, ZG001, ZG002) respectively. The relative position distribution and detection range of the 4 radars are as Figure 2 shown.

[0085] The main working parameters of S - POL and X - PAR are shown in Table 1. Z9200 is located at 113.36E, 23.00°N, was built and put into operation in 2001, and completed the dual - polarization upgrade transformation in 2016. Its working frequency is 2885 MHz, the peak power of the transmitter is ≥650 kW, the maximum detection distance is 460 km, the range resolution is 250 m, the horizontal beam width is 0.96°, and the vertical beam width is 0.95°. Since the electromagnetic wave of the S - band radar is less affected by precipitation attenuation, and Z9200 has strictly calibrated systems such as the transmitter, receiver, waveguide, and antenna during the operation process, the data reliability is relatively high. Therefore, the observation results of Z9200 are used as the accurate values in this study to quantitatively compare and analyze the errors of X - PAR.

[0086] The three X-PARs are located in Panyu (ZG000), Huadu (ZG001), and Baiyun (ZG002) respectively, and were put into operation in July 2017, January 2018, and December 2018 respectively. Their maximum detection range is 60 km, and the range resolution is accurate to 30 m. The X-PAR can achieve digital multi-beam scanning with wide transmission and narrow reception. The horizontal beam width is 1.2°, the vertical beam width is 0.9°, the elevation coverage range is 0.9 - 61.2°, the elevation resolution is 1.8°, there are 68 elevation layers in total, and the volume scan time is 60 s. It is a fully solid-state coherent, active phased array, pulsed Doppler, dual-polarization weather radar. All three X-PARs use the AXPT0364 type dual-polarization planar microstrip patch array antenna. The antenna array surface uses a rectangular aperture. There are 64 rows of dual-polarization microstrip patch linear arrays in the elevation direction, and each row of linear arrays has 32 radiation units in the azimuth direction, with a total of 2048 radiation units. The 64 rows of dual-polarization microstrip patch linear arrays correspond to 64 radio frequency transceiver units, and 8 radio frequency transceiver control units are used to complete the amplitude / phase control of the pulses and environmental monitoring. The elevation angle of the antenna array surface is fixed at 15° and does not change with the radar operating mode. The X-PAR uses the phased array electronic scanning method in the vertical direction. First, a continuous and non-stop phased array electronic RHI scan is completed at one azimuth angle, then switched to the next azimuth angle, and then the non-stop phased array electronic RHI scan is continued. Taking the VPPI mode of the traditional mechanical scanning weather radar as an example, in the VCP21 mode, the delay of the vertical profile RHI structure data is as high as 6 min, which is likely to cause discontinuity of the radar echo in the vertical direction. However, the phased array electronic scanning method of the X-PAR shortens the observation cycle to only 1 min. When the VRHI volume scan mode is used to complete a 360° volume scan, the radar antenna only needs to rotate mechanically for one circle, which not only avoids the delay when the traditional mechanical radar switches different elevation angles but also can obtain real-time and accurate RHI data. The vertical structure of the meteorological process is an important basis for analyzing and judging weather processes. The high-elevation, high-resolution, and non-delay RHI data provided by the X-PAR provide effective support for the monitoring, early warning, and research of weather processes. The main working parameters of POL and X-PAR are shown in Table 4.

[0087] Table 4

[0088]

[0089] In this embodiment, several precipitation cases in August 2022 are used for quantitative comparative analysis among multiple radars. The general situation of the precipitation cases observed in the radar experiment is shown in Table 5 (Beijing time, the same below).

[0090] Table 5

[0091]

[0092] The data preprocessing module is used to perform data quality control on the radar-based data to obtain preprocessed data.

[0093] Before quantitative analysis of radar-based data, it is necessary to eliminate ground clutter interference and clear sky echoes. Select datasets where Z H is greater than 15 dB, Z DR is greater than or equal to 0 and less than 5 dB, CC is greater than 0.9, and SNR is greater than 25 dB for consistency comparison analysis. Among them, the base data of X-PAR integrates the fixed coefficient "ZPHI" precipitation profile attenuation correction algorithm based on South China raindrop spectrum observations provided by Guangdong Narui Radar Technology Co., Ltd.

[0094] Based on the continuous volume scan results from 19 to 20 on August 25, 2022, the azimuths with severe low elevation beam blockage are manually extracted and marked as fixed interference azimuths, and the blocked azimuths are excluded during data preprocessing. For the quality control of Φ DP , the 3 km short-range echo method is used to determine the initial phase, and 1 km median filtering is used to eliminate noise information. Due to the scattering property differences of precipitation particles for electromagnetic waves of different wavelengths, the observation results of S-POL and X-PAR for the same target cannot be directly compared. Geng Fei (2023) calculated the observation parameters of S-band and X-band radars using the raindrop spectrum observation data of Guangdong Longmen Station, and performed exponential fitting on Z H , and polynomial fitting on Z DR . The observation parameters of S-POL and X-PAR are interchanged so that the observation results of the interchanged S-band and X-band radars for the same target can be compared.

[0095] The data calibration module is used to adopt different radar time and spatial sampling volume coordinate matching methods, and based on the observed values of S-POL, statistically match the differences of the preprocessed data to obtain the observation errors of X-PAR and correct them.

[0096] Since there are obvious differences in the volume scan mode, geographical location, sampling volume, etc. between S-POL and X-PAR, the observed data cannot directly establish a corresponding relationship, and different radar time and spatial sampling volume coordinate matching methods need to be adopted to achieve coordinate unification. As Figure 3 shown, a single sampling volume of S-POL can be equivalent to a frustum of a pyramid, and the elevation angle, azimuth angle, and slant range spherical coordinate information f i 0 (i = 1 - 8) of its 8 vertices can be calculated respectively, f i 0 (el i 0 , az i 0 , ri 0 ), the center point f of each sampling volume of X - PAR X The coordinates are converted from the spherical coordinates f X (el X , az X , r X ) of X - PAR into longitude, latitude, and altitude information D in the Cartesian coordinate system X (lon X , lat X , h X ). Subsequently, with the S - POL radar as the origin of the spherical coordinate system, the Cartesian coordinates of X - PAR are then inversely calculated into the S - POL spherical coordinate system, and the spherical coordinates f S (el S , az S , r S ) when the S - POL radar observes the X - PAR data points are obtained, and a one - to - one correspondence relationship is established for the observation data of the two radars with different positions. The positional relationship between the center point of the X - PAR sampling volume and the 8 vertices of the frustum of a pyramid of the S - POL radar sampling volume is compared, and the f S (el S , az S , r S ) falling within f i 0 (el i 0 , az i 0 , r i 0 ) are considered as observation data with consistent spatial matching. Finally, the acquisition time of each radar radial data is extracted. Considering the antenna rotation speeds of S - POL and X - PAR, the matching points with a radial data acquisition time difference within ±30 s are the final observation data with consistent spatio - temporal matching.

[0097] To quantitatively analyze the data quality of X - PAR, with the S - POL radar observation as the accurate value, a one - to - one correspondence relationship is established between the data of X - PAR and S - POL library by library, and the distribution relationship between the observation deviation of X - PAR (ΔParameter = Parameter _X - Parameter _S ) and the elevation angle is obtained, as shown in Figure 4 . In the multi - beam mode, a total of 68 elevation angle layers are scanned. Since the antenna parameters deteriorate and the transmitted beam width further increases above 40 elevation angles, and there is less meteorological echo information at high elevation angles, the data above 40 elevation angles are discarded. ΔZ H and ΔZ DRAll show a periodic change trend of five per elevation angle group, which coincides with the number of narrow receiving sub-beams within the wide emission beam in Table 4. The periodic change trend of the observation deviation is consistent with the analysis results of Wu Chong et al. (2014). The reason for the periodic change in the observation deviation is that the actual antenna radiation intensity is unevenly distributed within the wide emission beam. The normal angle of the X-PAR antenna array is fixed at 15°, and the elevation angle number corresponding to the receiving elevation angle stepping of 0.9° is between the 16th and 17th numbers, while ΔZ H 、ΔZ DR and ΔK DP all have minima with relatively small absolute errors at this elevation angle. During the preprocessing process, there may be phenomena such as misidentification or incomplete elimination of the occluded azimuth, resulting in relatively large ΔZ H deviations at the bottom two elevation angles.

[0098] In this embodiment, the observation deviation of Z H in the multi-beam mode is corrected. It should be noted that the change trend of ΔZ H with the detection distance is not obvious in the multi-beam mode ( Figure 4 (d)), so in the X-PAR elevation angle error correction in the multi-beam mode, all available azimuth and distance library data at a certain elevation angle are uniformly processed and calculated, and the distance factor is not a sensitive item for errors in the multi-beam mode.

[0099] This embodiment uses a correction method based on statistical results to calibrate the Z H of X-PAR. All valid precipitation observation cases in August 2022 in Table 5 are used. The differences in the X-PAR and S-POL reflectivity factors at each elevation angle of all observation data are statistically calculated, and based on the S-POL observation results as the accurate values, the Z H of X-PAR is corrected based on the statistical results. Figure 5 Four cases of the radar ZG000 in the multi-beam mode are selected, and the case times are 12:27, 12:33, 12:34, and 13:04 on August 25, 2022. The data with obvious interference at the bottom two elevation angles are discarded. The gray solid line in the figure is the curve of the difference in the echo intensity between X-PAR and S-POL obtained by the spatio-temporal coordinate matching algorithm with respect to the X-PAR observation elevation angle. ΔZ H has obvious periodic fluctuations of five per group and the overall deviation is about -3.5 dB. The ΔZ H -Elevation curve after statistical calibration is as shown by the red solid line in Figure 5 , and the standard deviation is significantly reduced (about 0.46 dB), and ΔZ HThe average deviation is significantly reduced to about -0.8 dB (Table 6). In addition, in this embodiment, the error calibration results of the existing antenna gain calibration method at the factory are compared, and the calibration values of the horizontal transmission gain G TH , horizontal reception gain G RH , vertical transmission gain G TV and vertical reception gain G RV of the AXPT0364 type dual-polarization planar microstrip patch array antenna calibrated in the anechoic chamber when the radar leaves the factory are used for error correction. The corrected ΔZ H - Elevation curve is shown as the blue solid line in Figure 5 . After the antenna gain at the factory is corrected, the standard deviation of the data is about 0.53 dB, and the average deviation of ΔZ H is about -2.5 dB. By comparing the existing antenna gain calibration method at the factory and the statistical calibration method of this embodiment, it can be seen that the overall calibration effect of the statistical calibration method of this embodiment is significantly better.

[0100] Figure 6 Similar to Figure 5 , 4 cases of radar ZG001 in the multi-beam mode are selected, and the case times are from 13:18 to 13:21 on August 25, 2022. Combining all cases, the antenna gain calibration method at the factory can appropriately eliminate the periodic fluctuation of ΔZ H , and increase the average observation deviation from about -3 dB to about -2 dB. However, the calibration effect is not restricted by S-POL data, and the calibration effect will also deteriorate when the attenuation of the X-PAR case is obvious; the statistical calibration method of this embodiment has the best comprehensive performance. After calibration, the average deviation of ΔZ H is about -0.6 dB, the maximum error is within -1 to -2 dB, and the standard deviation of the data at each elevation angle is about 0.4 dB. Table 6 shows the evaluation of the existing calibration and the calibration effect of this embodiment for the radar ZG000 and ZG001 ΔZ H .

[0101] Table 6

[0102]

[0103] Data unit: dB; *: Cases 1-4 correspond to the 4 cases in Figure 5 respectively; **: Cases 5-8 correspond to the 4 cases in Figure 6 respectively.

[0104] The embodiments described above are only descriptions of the preferred embodiments of the present invention, and do not limit the scope of the present invention. Without departing from the design spirit of the present invention, various deformations and improvements made by those of ordinary skill in the art to the technical solutions of the present invention should fall within the protection scope determined by the claims of the present invention.

Claims

1. A method for calibrating observation errors of multi-beam scanning mode of X-band phased array weather radar based on statistical results, characterized in that: The method comprises: Step S1, collecting observation base data of S-band dual polarization Doppler weather radar S-POL and X-band dual polarization phased array weather radar X-PAR as initial data; Step S2, performing data quality control on the radar-based data to obtain preprocessed data; Step S3, using different radar time and space sampling volume coordinate matching methods, taking the observation value of the S-POL as a reference, statistically analyzing the matching differences of the pre-processed data, obtaining the observation error of X-PAR and correcting it.

2. The method for calibrating observation errors of multi-beam scanning mode of X-band phased array weather radar based on statistical results according to claim 1 is characterized in that: In the step S1, the initial data includes observation base data of S-POL and observation base data of X-PAR; The operating frequency of the S-POL is 2885 MHz, the peak power of the transmitter is greater than or equal to 650 kW, the maximum detection distance is 460 km, the distance resolution is 250 m, the horizontal beam width is 0.96°, and the vertical beam width is 0.65°; The maximum detection distance of the X-PAR is 60km, the distance resolution is 30m, the horizontal beam width is 1.2°, the vertical beam width is 0.9°, and it adopts the AXPT0364 dual-polarization planar microstrip patch array antenna. The antenna array surface adopts a rectangular aperture, the elevation angle of the antenna array surface is fixed at 15°, there are 64 rows of dual-polarization microstrip patch linear arrays in the pitch direction, and each row of the linear array in the azimuth direction has 32 columns of radiation units.

3. The method for calibrating observation errors of multi-beam scanning mode of X-band phased array weather radar based on statistical results according to claim 2 is characterized in that: In step S2, the process of performing data quality control on the radar-based data specifically includes: Select the reflectivity factor Z from the initial data H Greater than 15dB, differential reflectivity factor Z DR Data with a value greater than or equal to 0 and less than 5dB, a zero-lag cross-correlation coefficient CC greater than 0.9, and a signal-to-noise ratio SNR greater than 25dB participate in the consistency comparison analysis; For the two-way differential propagation phase difference Φ DP The quality control uses the 3km close-range echo method to determine the initial phase, and uses the 1km median filter to remove noise information; The S-band and X-band radar observation parameters are calculated using the raindrop spectrometer observation data, and the Z H Perform exponential fitting to Z DR Polynomial fitting is performed to interchange the observation parameters of S-POL and X-PAR, so that the observation results of the interchanged S-band and X-band radars for the same target can be compared, thereby completing the data quality control of the radar base data.

4. The method for calibrating observation errors of multi-beam scanning mode of X-band phased array weather radar based on statistical results according to claim 3 is characterized in that: In step S3, the statistical process of the observation error specifically includes: The single sampling volume of S-POL is equivalent to a tetrahedron, and the eight vertices f of the tetrahedron are calculated respectively. i 0 Elevation, azimuth, and slant range spherical coordinate information f i 0 (el i 0 ,az i 0 ,r i 0 ), wherein i=1 to 8; Each sampling volume center point f of X-PAR X The coordinates are determined by the spherical coordinates based on X-PAR X (el X ,az X ,r X ) is converted into longitude, latitude, and altitude information in the Cartesian coordinate system D X (lon X ,lat X ,h X ); Taking the S-POL radar as the origin of the spherical coordinates, the Cartesian coordinates of the X-PAR are back-calculated into the S-POL spherical coordinate system to obtain the spherical coordinates f when the S-POL radar observes the X-PAR data point S (el S ,az S ,r S ); A one-to-one correspondence is established between the two radar observation data at different locations. Based on the correspondence, the positional relationship between the center point of the X-PAR sampling volume and the eight vertices of the S-POL radar sampling volume quadrangle is compared. S (el S ,az S ,r S ) falls on f i 0 (el i 0 ,az i 0 ,r i 0 ) is considered to be spatially consistent with observed data; The acquisition time of each radar radial data is extracted, and the antenna rotation speed of S-POL and X-PAR is considered. The matching points with radial data acquisition time difference within ±30s are considered to be the observation data with consistent final time-space matching. In order to quantitatively analyze the data quality of X-PAR, the minimum measurable echo intensity Z of X-PAR in the matched multi-beam scanning mode is calculated, taking the S-POL radar observation as the accurate value. min , Z H , Z DR , K DP Perform statistics and analyze the differences in X-PAR observation data to obtain the observation error.

5. An X-band phased array weather radar multi-beam scanning mode observation error calibration system based on statistical results, the system is used to use the observation error calibration method according to any one of claims 1 to 4, characterized in that the system include: The data acquisition module is used to collect the observation base data of the S-band dual-polarization Doppler weather radar S-POL and the X-band dual-polarization phased array weather radar X-PAR as initial data; A data preprocessing module, used for performing data quality control on the radar-based data to obtain preprocessed data; The data calibration module is used to adopt different radar time and space sampling volume coordinate matching methods, take the observation value of the S-POL as a reference, perform statistical matching differences on the pre-processed data, obtain the observation error of the X-PAR and correct it.

6. The X-band phased array weather radar multi-beam scanning mode observation error calibration system based on statistical results according to claim 5, characterized in that: In the data acquisition module, the initial data includes observation base data of S-POL and observation base data of X-PAR; The operating frequency of the S-POL is 2885 MHz, the peak power of the transmitter is greater than or equal to 650 kW, the maximum detection distance is 460 km, the distance resolution is 250 m, the horizontal beam width is 0.96°, and the vertical beam width is 0.65°; The maximum detection distance of the X-PAR is 60km, the distance resolution is 30m, the horizontal beam width is 1.2°, the vertical beam width is 0.9°, and it adopts the AXPT0364 dual-polarization planar microstrip patch array antenna. The antenna array surface adopts a rectangular aperture, the elevation angle of the antenna array surface is fixed at 15°, there are 64 rows of dual-polarization microstrip patch linear arrays in the pitch direction, and each row of the linear array in the azimuth direction has 32 columns of radiation units.

7. The X-band phased array weather radar multi-beam scanning mode observation error calibration system based on statistical results according to claim 5, characterized in that: In the data preprocessing module, the process of performing data quality control on the radar-based data specifically includes: Select the reflectivity factor Z from the initial data H Greater than 15dB, differential reflectivity factor Z DR Data with a value greater than or equal to 0 and less than 5dB, a zero-lag cross-correlation coefficient CC greater than 0.9, and a signal-to-noise ratio SNR greater than 25dB participate in the consistency comparison analysis; For the two-way differential propagation phase difference Φ DP The quality control uses the 3km close-range echo method to determine the initial phase, and uses the 1km median filter to remove noise information; The S-band and X-band radar observation parameters are calculated using the raindrop spectrometer observation data, and the Z H Perform exponential fitting to Z DR Polynomial fitting is performed to interchange the observation parameters of S-POL and X-PAR, so that the observation results of the interchanged S-band and X-band radars for the same target can be compared, thereby completing the data quality control of the radar base data.

8. The X-band phased array weather radar multi-beam scanning mode observation error calibration system based on statistical results according to claim 7, characterized in that: In the data calibration module, the statistical process of the observation error specifically includes: The single sampling volume of S-POL is equivalent to a tetrahedron, and the eight vertices f of the tetrahedron are calculated respectively. i 0 Elevation, azimuth, and slant range spherical coordinate information f i 0 (el i 0 ,az i 0 ,r i 0 ), wherein i=1 to 8; Each sampling volume center point f of X-PAR X The coordinates are determined by the spherical coordinates based on X-PAR X (el X ,az X ,r X ) is converted into longitude, latitude, and altitude information in the Cartesian coordinate system D X (lon X ,lat X ,h X ); Taking the S-POL radar as the origin of the spherical coordinates, the Cartesian coordinates of the X-PAR are back-calculated into the S-POL spherical coordinate system to obtain the spherical coordinates f when the S-POL radar observes the X-PAR data point S (el S ,az S ,r S ); A one-to-one correspondence is established between the two radar observation data at different locations. Based on the correspondence, the positional relationship between the center point of the X-PAR sampling volume and the eight vertices of the S-POL radar sampling volume quadrangle is compared. S (el S ,az S ,r S ) falls on f i 0 (el i 0 ,az i 0 ,r i 0 ) is considered to be spatially consistent with observed data; The acquisition time of each radar radial data is extracted, and the antenna rotation speed of S-POL and X-PAR is considered. The matching points with radial data acquisition time difference within ±30s are considered to be the observation data with consistent final time-space matching. In order to quantitatively analyze the data quality of X-PAR, the minimum measurable echo intensity Z of X-PAR in the matched multi-beam scanning mode is calculated, taking the S-POL radar observation as the accurate value. min , Z H , Z DR , K DP Perform statistics and analyze the differences in X-PAR observation data to obtain the observation error.

Images