A multi-type adaptive radial interference filtering method for multi-band weather radar

Abstract

The present invention relates to a multi-band weather radar multi-type adaptive radial interference filtering method, which belongs to the field of radio, and comprises: using a signal index to check whether there is radial interference in radar base data, and if so, classifying and distinguishing the detection distance of the radar; if the radar detection distance is within a first preset range, continuing to sample the signal index to check whether the radial interference of the radar base data is greater than a preset value, and if not, then sequentially executing six algorithms in different orders and outputting a quality control code, otherwise, sequentially executing six algorithms in different orders and outputting a quality control code; if the radar detection distance is within a second preset range and a third preset range, then sequentially executing six algorithms in different orders and outputting a quality control code. The present invention uses six algorithms to run in a certain order to achieve filtering and interpolation processing of different types of potential radial interference echoes, which can effectively avoid mis-filtering of data.

Description

Technical Field

[0001] The invention relates to the field of radio, and in particular to a multi-type adaptive radial interference filtering method for a multi-band weather radar. Background Art

[0002] In addition to being affected by factors such as ground objects, super refraction, and waves, the quality of Doppler weather radar reflectivity factor data is often interfered by other radars, electronic equipment, and other wireless power sources, as well as the influence of solar radiation, forming interference echoes on the radar PPI map, which seriously affects the detection effect of weather radar. In addition, the active interference echoes of weather radars in large cities are significantly more than those in remote areas, and the interference echoes are more obvious. Although forecasters can easily identify such echoes based on experience, the existing echo recognition algorithms do not have in-depth research on such interference echoes, and the recognition effect of some special echoes is poor. In areas where the active interference echoes are far away from the radar and mixed with precipitation echoes, it is difficult to separate them from precipitation echoes. Therefore, how to filter out the radial interference of weather radars in different bands is a problem that needs to be solved at present. Summary of the invention

[0003] The purpose of the present invention is to overcome the shortcomings of the prior art and provide a multi-band weather radar multi-type adaptive radial interference filtering method to solve the shortcomings of the prior art.

[0004] The object of the present invention is achieved by the following technical scheme: a multi-band weather radar multi-type adaptive radial interference filtering method, the interference filtering method comprising:

[0005] Step 1: Use signal indicators to check whether there is radial interference in the radar base data. If there is, classify and judge the detection distance of the radar;

[0006] Step 2: If the radar detection distance is within the first preset range, then continue to sample the signal index to check whether the radial interference of the radar base data is greater than the preset value. If it is not greater than the preset value, then execute the narrow peak algorithm, arc-shaped peak recognition algorithm, intermittent peak recognition algorithm, shorter longitudinal peak recognition algorithm and reverse weak wave peak recognition algorithm in sequence, and then output the quality-controlled data and the quality-control code of the data; if it is greater than the preset value, then execute the wide peak detection algorithm, narrow peak algorithm, arc-shaped peak recognition algorithm, intermittent peak recognition algorithm, shorter longitudinal peak recognition algorithm and reverse weak wave peak recognition algorithm in sequence, and then output the quality-controlled data and the quality-control code of the data;

[0007] If the radar detection distance is within the second preset range and the third preset range, the wide peak detection algorithm, narrow peak algorithm, arc-shaped peak recognition algorithm, intermittent peak recognition algorithm, shorter longitudinal peak recognition algorithm and reverse weak wave peak recognition algorithm are executed in sequence to output the quality-controlled data and the quality control code of the data.

[0008] After executing the intermittent peak recognition algorithm, it is also necessary to determine the size of the radar detection radius. If the radar detection radius is greater than or equal to the set value, the shorter longitudinal peak recognition algorithm with a radial threshold of 60 is executed; if the radar detection radius is less than the set value, the shorter longitudinal peak recognition algorithm with a radial threshold of 110 is executed.

[0009] If the radar detection distance is within the second preset range, the wide peak detection algorithm with a threshold of 55, the narrow peak algorithm with an azimuth angle not exceeding 7° or 11°, the arc-shaped peak recognition algorithm, the intermittent peak recognition algorithm, the shorter longitudinal peak recognition algorithm and the reverse weak wave peak recognition algorithm are sequentially executed to output the quality-controlled data and the quality control code of the data;

[0010] After executing the wide spike detection algorithm with a threshold of 55, the percentage of valid data in each radial direction is determined. If the quantiles of 95% and 80% are both greater than or equal to 80, the narrow spike algorithm with an azimuth angle not exceeding 11° is executed, otherwise the narrow spike algorithm with an azimuth angle not exceeding 7° is executed.

[0011] If the radar detection distance is within the third preset range, the wide peak detection algorithm with a threshold of 45, the narrow peak algorithm with an azimuth angle not exceeding 7° or 11°, the arc-shaped peak recognition algorithm, the intermittent peak recognition algorithm, the short longitudinal peak recognition algorithm and the reverse weak wave peak recognition algorithm are sequentially executed to output the quality-controlled data and the quality control code of the data;

[0012] After executing the wide spike detection algorithm with a threshold of 55, the percentage of valid data in each radial direction is determined. If the quantiles of 95% and 80% are both greater than or equal to 80, the narrow spike algorithm with an azimuth angle not exceeding 11° is executed, otherwise the narrow spike algorithm with an azimuth angle not exceeding 7° is executed.

[0013] The narrow spike algorithm includes:

[0014] Each reflectivity gate (α, l ) of the detected echo detects the reflectivity of the adjacent beam at the same distance from the radar radial range library l with an azimuth angle of α ± d, where α represents the azimuth of the radar and the azimuth angle offset is d = 5°;

[0015] If the conditions are met No echo and or No echo and , and meets the conditions , then the reflectivity gate (α, l ) is considered as potential interference, Z is the radar reflectivity, represents the reflectivity of the distance library l at azimuth α-d, It represents the variance of the distance library l with azimuth angle α in the radial direction;

[0016] The number of potential gates along a given radar beam is calculated. If this number is above 30%, and the number of potential gates accounts for 80% of the valid reflectivity gates, all gates with interference are replaced with reflectivities interpolated from adjacent beams, and the entire beam is considered to have interference. At this time, the radar beam is given an interference flag.

[0017] Among them, the gate specifically refers to a certain distance library, but the meaning of expression is different. The reflectivity gate is a certain radar reflectivity distance library, the potential gate value is that a certain distance library may be an interference callback, and the interference gate is the interference callback of a certain distance library.

[0018] The broad spike detection algorithm includes:

[0019] The coefficient of variation of the locally confirmed reflectivity was used to examine the variability of the returns across and along the radar beam;

[0020] If the coefficient of variation across the radar beam and the coefficient of variation along the radar beam meet the conditions at the same time, the reflectivity gate (α, l) is considered as a potential interference;

[0021] If more than 55% of the gates along a radar beam are classified as potentially interfering, all gates with interference are replaced with reflectivities interpolated from adjacent beams and the entire beam is considered to have interference, at which point the radar beam is given an interference designation.

[0022] The arc-shaped peak recognition algorithm includes:

[0023] For each radar radial starting from the end farthest from the radar, detect each reflectivity gate (α, l) of the echo and calculate the distance r between the most recent Z (α, l1)>−33dBZ and the next Z (α, l2)>−33dBZ;

[0024] When the number of times R(α) ≥ 3 for a radial r greater than 10 km, check the R(α) of the vector beam within the azimuth α ± d at the same distance from the radar, d = 10°;

[0025] If the conditions are met If the l1(α) of the adjacent beams is increasing or decreasing, the entire beam is marked as a potential interference, and all echo values ​​of the beam at the current α azimuth are assigned as missing values ​​without any interpolation or filling. Indicates the number of radials within the azimuth that satisfy the condition.

[0026] The intermittent spike identification algorithm includes:

[0027] For each azimuth angle α at a given elevation angle ε, the echo bins are counted along the entire radar beam to obtain the number of valid echoes. , the var(ε,α) value is determined as the larger of the variance values ​​between n(ε,α) and the values ​​in the adjacent azimuths α-1 and α+1, i.e. , where N represents the number of radar bins along the radar beam, Indicates whether there is an echo at an elevation angle of ε, an azimuth angle of α and a distance from the library l, and var(ε,α) indicates the variance of the elevation angle of ε and the azimuth angle of α;

[0028] For the analyzed azimuth angle α, two average values ​​are calculated in the range of ±4 and ±40 respectively. and , and determine the value ;

[0029] If the conditions are met or and , then the azimuth is considered to be a potential interference azimuth. At the same time, if the azimuths α-1 and α+1 are potential interferences, then the azimuth α is considered to be a potential interference azimuth.

[0030] Using the variance calculated across the radar beam, confirm that for each elevation angle, one of the above two conditions is met, the variance across the azimuth bin of the potential interfering echo , if the condition is met and , then it is believed that If it is a potential interference, all gates with interference are replaced by the reflectivity of adjacent beam interpolation, and the entire beam is considered to have interference. At this time, the radar beam is given an interference mark.

[0031] The shorter longitudinal spike identification algorithm includes:

[0032] If the shorter longitudinal interference echo in the radar library meets the conditions and and , then the interference echo is considered to be detected, all gates with interference are replaced with reflectivities interpolated from adjacent beams, and the entire beam is considered to have interference. At this time, the radar beam is given an interference mark. Indicates the number of reflectances along the radial direction greater than a certain threshold. Indicates the number of reflectances in the cross-radial direction greater than a certain threshold, It represents the variance of the distance library l with azimuth angle α along the radial direction.

[0033] The reverse weak wave peak recognition algorithm includes:

[0034] For each azimuth, if the condition is met and and , then it is considered that If the radar library is a potential interference echo, all gates with interference are replaced by reflectivities interpolated from adjacent beams, and the entire beam is considered to have interference. At this time, the radar beam is given an interference mark. Represents the slope of the reflectivity fitted by all distance libraries at elevation angle ε and azimuth angle α, represents the reflectivity at elevation angle ε, azimuth angle α-d and distance library l.

[0035] The present invention has the following advantages: a multi-band weather radar multi-type adaptive radial interference filtering method, which adopts six algorithms to run in a certain order to realize filtering and interpolation processing of different types of potential radial interference echoes, can effectively avoid erroneous filtering of data and increase the reliability of filtering. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 It is a schematic diagram of the process of the present invention. DETAILED DESCRIPTION

[0037] In order to make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the technical scheme in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. The components of the embodiments of the present application described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the detailed description of the embodiments of the present application provided below in conjunction with the drawings is not intended to limit the scope of protection of the application claimed for protection, but merely represents the selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work belong to the scope of protection of the present application. The present invention is further described below in conjunction with the drawings.

[0038] like Figure 1 As shown, in order to solve the problem of radial interference filtering of weather radars in different bands, the present invention specifically relates to a multi-band weather radar multi-type adaptive radial interference filtering method, which specifically includes the following contents:

[0039] Step 1: Use signal indicators to check whether there is radial interference in the radar base data. If so, classify and judge the detection distance of the radar. Otherwise, directly output the quality control code.

[0040] Furthermore, since the cloud area is relatively continuous within the radar detection range, in the ideal case without clutter, the reflectivity factor data of precipitation should also be relatively continuous, and its radial energy should also change continuously. If there is an interference echo, the radial energy may change suddenly near the clutter boundary. Therefore, it is currently considered to detect the accumulation of the difference between two adjacent beams of the radar echo in the radial direction, which can be used as a signal of radial interference in the later stage. The calculation method of the radial interference signal is as follows:

[0041] (1)

[0042] (2)

[0043] Among them, α represents the direction of the radar, represents the radial distance library, L represents the total number of distance libraries on a radial, count represents counting, where represents judgment, radial represents the total number of radials in the entire layer, energy represents the energy accumulation between two adjacent radials, and when the total number of radials with energy accumulation exceeding 1500dBZ accounts for the proportion (signal) of the total number of single-layer PPI radials > 0%, it is considered that an interference radial has occurred in the current PPI.

[0044] Step 2: If the radar detection distance is less than or equal to 200km, continue to sample the signal index to check whether the radial interference of the radar base data is greater than 5%. If it is not greater than 5%, then the narrow peak algorithm, arc-shaped peak recognition algorithm, intermittent peak recognition algorithm, shorter longitudinal peak recognition algorithm and reverse weak wave peak recognition algorithm are executed in sequence with an azimuth angle not exceeding 5°, and then the quality-controlled data and the quality-control code of the data are output; if it is greater than 5%, then the wide peak detection algorithm with a threshold of 55, the narrow peak algorithm, the arc-shaped peak recognition algorithm, the intermittent peak recognition algorithm, the shorter longitudinal peak recognition algorithm and the reverse weak wave peak recognition algorithm are executed in sequence, and then the quality-controlled data and the quality-control code of the data are output;

[0045] If the radar detection distance is greater than 300km and greater than 200km and less than or equal to 300km, the wide peak detection algorithm, narrow peak algorithm, arc-shaped peak recognition algorithm, intermittent peak recognition algorithm, shorter longitudinal peak recognition algorithm and reverse weak wave peak recognition algorithm are executed in sequence to output the quality-controlled data and the quality control code of the data.

[0046] Furthermore, after executing the intermittent peak recognition algorithm, it is also necessary to determine the size of the radar detection radius. If the radar detection radius is greater than or equal to 200 km, a shorter longitudinal peak recognition algorithm with a radial threshold of 60 is executed; if the radar detection radius is less than 200 km, a shorter longitudinal peak recognition algorithm with a radial threshold of 110 is executed.

[0047] Furthermore, if the radar detection distance is greater than 300km, the wide peak detection algorithm with a threshold of 55, the narrow peak algorithm with an azimuth angle not exceeding 7° or 11°, the arc-shaped peak recognition algorithm, the intermittent peak recognition algorithm, the shorter longitudinal peak recognition algorithm and the reverse weak wave peak recognition algorithm are sequentially executed to output the quality-controlled data and the quality control code of the data;

[0048] If the radar detection distance is greater than 200km and less than or equal to 300km, the wide peak detection algorithm with a threshold of 45, the narrow peak algorithm with an azimuth angle not exceeding 7° or 11°, the arc-shaped peak recognition algorithm, the intermittent peak recognition algorithm, the shorter longitudinal peak recognition algorithm and the reverse weak wave peak recognition algorithm are executed in sequence to output the quality-controlled data and the quality control code of the data.

[0049] Furthermore, after executing the wide spike detection algorithm with a threshold of 55, the percentage of valid data in each radial direction is determined. If the quantiles of 95% and 80% are both greater than or equal to 80, the narrow spike algorithm with an azimuth angle not exceeding 11° is executed, otherwise the narrow spike algorithm with an azimuth angle not exceeding 7° is executed.

[0050] Furthermore, the broad peak detection algorithm specifically includes the following contents:

[0051] For wide interference, an algorithm based on the analysis of the spatial structure of radar echoes is employed, using the locally determined coefficient of variation (ratio of standard deviation to mean) of the reflectivity to examine the variability of the echo across and along the radar beam, so that a given echo is classified as a potential interference if the first coefficient of variation is high and the latter is low, i.e.:

[0052] Across radar beam directions: (3),

[0053] Along the radar beam direction: (4),

[0054] If both conditions of equations (3) and (4) are satisfied, the reflectivity gate (α, l ) is considered as a potential interferer.

[0055] in is the coefficient of variation for a given gate calculated across the radar beam at distances not exceeding ± 5° in azimuth; is the coefficient of variation of a given radar gate calculated within ± 10 km along the radar beam; Z is the radar reflectivity expressed in dBZ.

[0056] If more than 55% of the gates along a radar beam are classified as potentially interfering, all gates with interference are replaced with reflectivities interpolated from adjacent beams and the entire beam is considered to have interference, at which point the radar beam is given an interference flag; along a radar beam is a ray scanned in the radial direction.

[0057] Furthermore, the narrow peak algorithm specifically includes the following contents:

[0058] To remove narrow interference, i.e. interference with an azimuth angle not exceeding 11°, the algorithm is applied to each reflectivity gate (α, l ) where an echo is detected, which means that Z (α, l ) > − 33 dBZ (this is the lowest reflectivity measured by the radar). For these reflectivity gates, the reflectivities of the neighboring beams at azimuth angles α ± d (where d = 5°) at the same distance from the radar range bin l are checked:

[0059] No echo and (5),

[0060] No echo and (6),

[0061] In order to avoid the longitudinal echo similar to radial interference caused by terrain occlusion, longitudinal variance detection is introduced after this step to retain normal echo data of this type:

[0062] (7),

[0063] Z is the radar reflectivity, express.

[0064] If one of the conditions in equation (5) or (6) is met and equation (7) is met, then the reflectivity gate (α, l ) is considered as a potential interferer. The process is repeated for d = 3° and 1°. Next, the number of potential gates along a given radar beam is calculated. If this number is higher than 30% and the number of potential gates accounts for 80% of the valid reflectivity gates, all gates with interference are replaced with reflectivities interpolated from adjacent beams, and the entire beam is considered to have interference. At this time, the radar beam is given an interference flag.

[0065] Furthermore, the arc-shaped peak recognition algorithm specifically includes the following contents:

[0066] For arc interference, that is, interference with an azimuth angle not exceeding 20°, for each radar radial starting from the end far away from the radar, detect each reflectivity gate (α, l ) of the echo, calculate the distance r between the most recent Z(α, l1)>−33dBZ and the next Z(α, l2)>−33dBZ (and so on), and the distance library l1(α) at ​​this time is recorded. When the number of times r on a radial is greater than 10km R(α)≥3; for these radials, check the R(α) of the adjacent beams within the azimuth angle α ± d (where d = 10°) at the same distance from the radar:

[0067] (8),

[0068] And the l1(α) of adjacent beams shows an increasing or decreasing trend; among them, is the number of radials within the azimuth that satisfy the condition.

[0069] If formula (8) is satisfied, the entire beam is given a potential interference mark. Due to the particularity of arc interference, all echo values ​​of the beam are assigned as missing values ​​here, and no interpolation filling processing is performed.

[0070] Furthermore, the intermittent peak recognition algorithm specifically includes the following contents:

[0071] This algorithm is suitable for interference echoes that are usually elongated in shape but discontinuous along and across the radar beam. Due to the discontinuity of its pattern, the above three sub-algorithms cannot handle such interference. The algorithm is divided into the following two steps:

[0072] Potential azimuths with intermittent interference are first detected and then confirmed by appropriate variability of the radar beam. For the lowest elevation angles, the first step is omitted, as this scan may be quite redundant, so the detection of potential interference may be inefficient, and the azimuth with potential interference is obtained from the adjacent higher elevation angles.

[0073] In the following procedure, for each azimuth angle α at a given elevation angle ε (except the lowest elevation angle), the echo bins are counted along the entire radar beam to obtain the value of the number of valid echoes n(ε,α):

[0074] (9),

[0075] in, , N is the number of radar bins along the radar beam, It indicates whether there is an echo at the elevation angle ε, the azimuth angle α and the distance from the library l, and var(ε,α) indicates the variance of the elevation angle ε and the azimuth angle α.

[0076] The var(ε,α) value is determined as the larger of the variance values ​​between n(ε,α) and the values ​​in the adjacent azimuths α-1 and α+1:

[0077] (10),

[0078] Then for the analyzed azimuth angle α, two average values ​​are calculated and In the range of ±4 and ±40 respectively, and determine the value Var(ε,α):

[0079] (11),

[0080] Potential intermittent interference is found in the radar beam at azimuths that meet the following conditions.

[0081] (12),

[0082] and (13),

[0083] If one of the above two conditions is met, the azimuth is considered to be a potential interference azimuth.

[0084] Due to the specific shape of this type of echo, an additional check is introduced in order to smooth the field of relevant potential interferers: An azimuth angle α is considered as a potential interferer if azimuth angles α-1 and α+1 are potential interferers.

[0085] In the final step, the variance calculated across the radar beam is used to identify the variance across azimuth bins that have such potential interfering echoes at each elevation angle. , it is also necessary to consider the case of no echo value. The azimuth box refers to a distance library at a certain azimuth.

[0086] and (14),

[0087] The variance Calculate the reflectivity (in dBZ), The unit of reflectivity is .

[0088] If the above conditions are met, it is considered If the gates with interference are replaced by reflectivities interpolated from adjacent beams, the entire beam is considered to have interference, and the radar beam is given an interference flag.

[0089] The concept behind the above formula is that intermittent interference echoes have a certain radar reflectivity variance among the echoes that pass through the radar beam. The variance is calculated in dBZ and is therefore logarithmic ( ), which is high for this type of echo. However, a high variance in dBZ is also characteristic of strong weather echoes, especially those originating from convective rainfall. On the other hand, in non-log values ​​( ), the variance of weather echoes is relatively high, since their values ​​are usually higher than those of interference echoes and their internal differentiality is greater. Therefore, the algorithm used to detect intermittent interference echoes assumes that they are composed of high variances calculated for reflectivity in dBZ and high variances calculated for reflectivity in dBZ. The reflectance calculated for the unit is characterized by a relatively low variance.

[0090] Furthermore, the shorter longitudinal peak recognition algorithm specifically includes the following contents:

[0091] This algorithm is only used for radars with a significant interference echo burden and only for the lowest two elevation angles. For each elevation angle, a reflectivity threshold of 10 dBZ is used for each radar bin. , calculate the number of consecutive neighbors along the dimension whose reflectivity is above the threshold and span size The number of times a given radar beam appears. Shorter longitudinal interference echoes in the radar library If the following two conditions are met, an interference echo is considered to be detected.

[0092] and (15),

[0093] (16),

[0094] All gates with interference are replaced with reflectivities interpolated from adjacent beams, and the entire beam is considered to have interference. The radar beam is then given an interference flag. Indicates the number of reflectances along the radial direction greater than a certain threshold. Indicates the number of reflectances in the cross-radial direction greater than a certain threshold, It represents the variance of the distance library l with azimuth angle α along the radial direction.

[0095] Furthermore, the reverse weak wave spike recognition algorithm specifically includes the following contents:

[0096] The algorithm targets interference shapes of radar sites visible on the radar picture. Such echoes are usually the result of beam blocking, but it is not possible to detect them with digital terrain maps (Dem). This sub-algorithm is used only for the two lowest altitudes. For each azimuth, find the radar bins with potential reverse interference echoes using the following criteria.

[0097] and and (17),

[0098] in Represents the slope of the reflectivity fitted by all distance libraries at elevation angle ε and azimuth angle α, represents the elevation angle ε, azimuth angle α-d and reflectivity on the distance library l. If the above conditions are met, it is considered that the The radar library is a potential interference echo and this process is performed for d values ​​from 1° to 2°.

[0099] All gates with interference are replaced with reflectivities interpolated from adjacent beams, and the entire beam is considered to have interference, at which point the radar beam is given an interference flag.

[0100] The radial interference filtering sub-algorithm verification of potential interference type echoes is used to check all echoes marked as potential interference echoes, that is, echoes suspected to be non-meteorological echoes. To this end, for a given azimuth angle α and elevation angle ε, the number of bins with detected potential interference echoes is calculated for each interference echo type: if it exceeds a preset threshold, the number of bins with detected potential interference echoes of all types at azimuth angle α and elevation angle ε is filtered out as real (confirmed) interference.

[0101] Furthermore, the interpolation algorithm replaced by the reflectivity interpolated from the adjacent beams includes the following two interpolations:

[0102] 1. For a given marked bin, check the adjacent bins and then check further azimuths (left and right) at the same distance from the radar site until the unmarked bin is found. The reflectivity value in the analysis bin is determined as the average of the two values ​​on the left and right, but if one of the interpolated values ​​is no-echo, the interpolated value is also considered no-echo.

[0103] 2. For a given radar marker box, a principle similar to the reverse speckle removal algorithm is adopted, using a 7×7 window for interpolation, and the corresponding weight t is 10; the algorithm can be executed twice and can effectively fill the unreasonable data holes.

[0104] Furthermore, in order to verify the reliability of the data and check the radar data in the future, if it is found that the data of a radar station often has more radial interference, the radar can be adjusted or the interference source can be found according to the actual situation; in this way, the data quality of the radar can be monitored. According to actual statistical analysis, a simple filtered quality control code is currently used to identify the quality control of the data; the specific quality control index is calculated as follows:

[0105] For a volume scan data, each elevation layer is processed, the number of interference radials that appear after processing is counted, and the percentage of interference radials in each elevation layer is calculated; then the percentages of all elevation layers are added; finally, the data quality control code is output according to the following rules:

[0106] The sum of the percentages is divided by 10 and rounded up. For data less than 10, the quality-controlled data and the data quality control code are output according to the actual value. If the data is greater than or equal to 10, 9 is output as the data quality control code. Therefore, the quality control code range of the overall data is 0~9. The quality control code of a certain data (quality control code is 5) can be understood as: the radial number of radial interference accounts for 40%~50% of the radial number of the entire layer. In this way, the quality control code of a certain data can intuitively reflect the quality of the body scan, and it is also convenient for regular inspection of data in the later stage.

[0107] The above is only a preferred embodiment of the present invention. It should be understood that the present invention is not limited to the form disclosed herein, and should not be regarded as excluding other embodiments, but can be used for various other combinations, modifications and improvements, and can be modified within the scope of the concept described herein through the above teachings or the technology or knowledge of the relevant field. The changes and modifications made by those skilled in the art do not deviate from the spirit and scope of the present invention, and should be within the scope of protection of the claims attached to the present invention.

Claims

1. A multi-band weather radar multi-type adaptive radial interference filtering method, characterized by: The interference filtering method comprises: Step 1: Use the signal index to check whether there is radial interference in the radar base data. If there is, classify and judge the detection distance of the radar. The signal index refers to the ratio of the total number of radial energy accumulation exceeding 1500dBZ to the total number of single-layer PPI radial energy being greater than 0%; Step 2: If the radar detection distance is within the first preset range, then continue to sample the signal index to check whether the radial interference of the radar base data is greater than the preset value. If it is not greater than the preset value, then execute the narrow peak algorithm, arc-shaped peak recognition algorithm, intermittent peak recognition algorithm, shorter longitudinal peak recognition algorithm and reverse weak wave peak recognition algorithm in sequence, and then output the quality-controlled data and the quality-control code of the data; if it is greater than the preset value, then execute the wide peak detection algorithm, narrow peak algorithm, arc-shaped peak recognition algorithm, intermittent peak recognition algorithm, shorter longitudinal peak recognition algorithm and reverse weak wave peak recognition algorithm in sequence, and then output the quality-controlled data and the quality-control code of the data; If the radar detection distance is within the second preset range and the third preset range, the wide peak detection algorithm, narrow peak algorithm, arc-shaped peak recognition algorithm, intermittent peak recognition algorithm, shorter longitudinal peak recognition algorithm and reverse weak wave peak recognition algorithm are executed in sequence to output the quality-controlled data and the quality control code of the data.

2. The multi-band weather radar multi-type adaptive radial interference filtering method according to claim 1, characterized in that: After executing the intermittent peak recognition algorithm, it is also necessary to determine the size of the radar detection radius. If the radar detection radius is greater than or equal to the set value, the shorter longitudinal peak recognition algorithm with a radial threshold of 60 is executed; if the radar detection radius is less than the set value, the shorter longitudinal peak recognition algorithm with a radial threshold of 110 is executed.

3. The multi-band weather radar multi-type adaptive radial interference filtering method according to claim 1, characterized in that: If the radar detection distance is within the second preset range, the wide peak detection algorithm with a threshold of 55, the narrow peak algorithm with an azimuth angle not exceeding 7° or 11°, the arc-shaped peak recognition algorithm, the intermittent peak recognition algorithm, the shorter longitudinal peak recognition algorithm and the reverse weak wave peak recognition algorithm are sequentially executed to output the quality-controlled data and the quality control code of the data; After executing the wide spike detection algorithm with a threshold of 55, the percentage of valid data in each radial direction is determined. If the quantiles of 95% and 80% are both greater than or equal to 80, the narrow spike algorithm with an azimuth angle not exceeding 11° is executed, otherwise the narrow spike algorithm with an azimuth angle not exceeding 7° is executed.

4. The multi-band weather radar multi-type adaptive radial interference filtering method according to claim 1, characterized in that: If the radar detection distance is within the third preset range, the wide peak detection algorithm with a threshold of 45, the narrow peak algorithm with an azimuth angle not exceeding 7° or 11°, the arc-shaped peak recognition algorithm, the intermittent peak recognition algorithm, the short longitudinal peak recognition algorithm and the reverse weak wave peak recognition algorithm are sequentially executed to output the quality-controlled data and the quality control code of the data; After executing the wide spike detection algorithm with a threshold of 55, the percentage of valid data in each radial direction is determined. If the quantiles of 95% and 80% are both greater than or equal to 80, the narrow spike algorithm with an azimuth angle not exceeding 11° is executed, otherwise the narrow spike algorithm with an azimuth angle not exceeding 7° is executed.

5. A multi-band weather radar multi-type adaptive radial interference filtering method according to any one of claims 1 to 4, characterized in that: The narrow spike algorithm includes: Each reflectivity gate (α, l) of the detected echo detects the reflectivity of the adjacent beam at the same distance from the radar radial range library l with an azimuth angle of α ± d, where α represents the azimuth of the radar and the azimuth angle offset is d = 5°; If the condition Z(α-d,l) has no echo and Z(α,l)>-23dBZ or Z(α+d,l) has no echo and Z(α,l)>-23dBZ, and the condition , then the reflectivity gate (α, l ) is considered as potential interference, Z is the radar reflectivity, Z(α-d, l) represents the reflectivity of the distance bin l on the azimuth α-d, It represents the variance of the distance library l with azimuth angle α in the radial direction; The number of potential gates along a given radar beam is calculated. If this number is above 30%, and the number of potential gates accounts for 80% of the valid reflectivity gates, all gates with interference are replaced with reflectivities interpolated from adjacent beams, and the entire beam is considered to have interference. At this time, the radar beam is given an interference flag.

6. A multi-band weather radar multi-type adaptive radial interference filtering method according to any one of claims 1 to 4, characterized in that: The broad spike detection algorithm includes: The coefficient of variation of the locally confirmed reflectivity was used to examine the variability of the returns across and along the radar beam; If the coefficient of variation across the radar beam and the coefficient of variation along the radar beam meet the conditions at the same time, the reflectivity gate (α, l) is considered as a potential interference; where the condition is met in the direction across the radar beam: , along the radar beam direction, the conditions are met: , then the reflectivity gate (α, l) is considered as a potential interference, where is the coefficient of variation for a given gate calculated across the radar beam at distances not exceeding ±5° in azimuth; is the coefficient of variation of a given radar gate calculated within ±10 km along the radar beam; Z is the radar reflectivity; If more than 55% of the gates along a radar beam are classified as potentially interfering, all gates with interference are replaced with reflectivities interpolated from adjacent beams and the entire beam is considered to have interference, at which point the radar beam is given an interference designation.

7. A multi-band weather radar multi-type adaptive radial interference filtering method according to any one of claims 1 to 4, characterized in that: The arc-shaped peak recognition algorithm includes: For each radar radial starting from the end farthest from the radar, detect each reflectivity gate (α, l) of the echo and calculate the distance r between the most recent Z(α, l1)>−33dBZ and the next Z(α, l2)>−33dBZ, where Z is the radar reflectivity, α represents the azimuth of the radar, the azimuth offset d=5°, and l is the range bin; When r in the radial direction is greater than 10 km, the number of When , check the R(α) of the vector beam within the azimuth α ± d at the same distance from the radar, d = 10°; If the conditions are met If the l1(α) of the adjacent beams is increasing or decreasing, the entire beam is marked as a potential interference, and all echo values ​​of the beam at the current α azimuth are assigned as missing values ​​without any interpolation or filling processing, where num(R(α)) represents the number of radials that meet the conditions within the azimuth.

8. A multi-band weather radar multi-type adaptive radial interference filtering method according to any one of claims 1 to 4, characterized in that: The intermittent spike identification algorithm includes: For each azimuth angle α at a given elevation angle ε, the echo bins are counted along the entire radar beam to obtain the number of valid echoes. , the var (ε, α) value is determined as the larger of the variance values ​​between n (ε, α) and the adjacent azimuths α-1 and α+1, that is, , where N represents the number of radar bins along the radar beam, i(ε,α,l) represents the presence or absence of echo at elevation ε, azimuth α and distance bin l, and var(ε,α) represents the variance at elevation ε and azimuth α; For the analyzed azimuth angle α, two average values ​​are calculated in the range of ±4 and ±40 respectively. and , and determine the value ; If the conditions are met or and , then the azimuth is considered to be a potential interference azimuth. At the same time, if the azimuths α-1 and α+1 are potential interferences, then the azimuth α is considered to be a potential interference azimuth. Using the variance calculated across the radar beam, confirm that for each elevation angle, one of the above two conditions is met, the variance across the azimuth bin of the potential interfering echo , if the condition is met and , then (ε, α, l) is considered as potential interference, all gates with interference are replaced by the reflectivity of adjacent beam interpolation, and the entire beam is considered to have interference. At this time, the radar beam is given an interference mark.

9. A multi-band weather radar multi-type adaptive radial interference filtering method according to any one of claims 1 to 4, characterized in that: The shorter longitudinal spike identification algorithm includes: If the short longitudinal interference echo in the radar library meets the conditions and and , then the interference echo is considered to be detected, all gates with interference are replaced with reflectivity interpolated from adjacent beams, and the entire beam is considered to have interference. At this time, the radar beam is given an interference mark. Indicates the number of reflectances along the radial direction greater than a certain threshold. Indicates the number of reflectances in the cross-radial direction greater than a certain threshold, It represents the variance of the distance library l with azimuth angle α along the radial direction.

10. A multi-band weather radar multi-type adaptive radial interference filtering method according to any one of claims 1 to 4, characterized in that: The reverse weak wave peak recognition algorithm includes: For each azimuth, if the condition is met and And Slope(Z(ε,α))<0, the (ε,α,l) radar library is considered to be a potential interference echo, then all gates with interference are replaced by reflectivities interpolated from adjacent beams, and the entire beam is considered to have interference. At this time, the radar beam is given an interference mark, Slope(Z(ε,α)) represents the slope of the reflectivity fitted by all distance libraries at elevation ε and azimuth α, and Z(ε,α-d,l) represents the reflectivity at elevation ε, azimuth α-d and distance library l.

Images