A method and apparatus for removing electromagnetic interference from weather radar

Abstract

This application provides a method and apparatus for removing electromagnetic interference from weather radar. The method includes: acquiring detection data from the weather radar and calculating the echo intensity difference; setting a first threshold, a second threshold, and a third threshold; statistically analyzing the number of range databases and the number of effective echo range databases; determining whether the radial direction is the interference radial direction and numbering them in sequence; determining the edge orientation of the strip-shaped electromagnetic interference, and then performing interference removal and correction; and removing radial scattered noise. This application solves the technical problems of poor practical application effect and low recognition accuracy when using existing edge recognition methods to remove strip-shaped electromagnetic interference. The method provided by this application has clear logic, few parameters, and high execution efficiency. It can adapt to the ideal interference scenario of "covering the entire radial direction" and accurately deal with the actual detection scenario of "random and discontinuous distribution of interference", effectively improving the recognition accuracy and removal effect of strip-shaped electromagnetic interference.

Description

Technical Field

[0001] This application relates to the field of meteorological data technology, and in particular to a method and apparatus for removing electromagnetic interference from weather radar. Background Technology

[0002] Radar was initially used in the military. With the development of technology, its detection capabilities have become increasingly sophisticated, making it an important means of weather detection. It is widely used to observe weather phenomena such as clouds, rain, snow, thunderstorms, and turbulence, and is particularly valuable for monitoring and forecasting severe weather. Radar emits electromagnetic waves in the form of pulses. When these electromagnetic waves encounter various substances (raindrops, snowflakes, hail, and other non-meteorological targets), some of the electromagnetic wave energy is scattered by the target particles, while the rest is absorbed. The scattered energy is received by the radar, processed by radar hardware and calculated by software, and then converted into an image, which is displayed as a radar echo.

[0003] Radar echoes are categorized into meteorological and non-meteorological echoes. Non-meteorological echoes mainly include three types: ground object echoes, biological echoes, and electromagnetic interference echoes. Strip-shaped electromagnetic interference is a very common type of electromagnetic interference echo. It is generally caused by long-distance single-frequency electromagnetic wave interference and internal data processing anomalies. The interference echo is distributed in a narrow, elongated strip shape, often existing in one or several consecutive directions within the radar echo, and can generally extend to the radar's maximum detection range. In the presence of precipitation, especially when the precipitation area is large, the presence of electromagnetic interference can severely affect the quality of the actual precipitation echo assessment.

[0004] Currently, edge recognition is a commonly used method for removing strip-shaped electromagnetic interference. This method identifies the interference based on its edge characteristics, and it is particularly effective for interference that is continuously distributed along the entire radial direction. However, actual strip-shaped electromagnetic interference is often discontinuous in radar echoes, and the same interference is often distributed in multiple directions with large differences in the distribution of interference data in each direction. This results in poor practical application effect and low recognition accuracy of the edge recognition method. Summary of the Invention

[0005] The main purpose of this application is to provide a method and apparatus for removing electromagnetic interference from weather radar, which aims to solve the technical problems of poor practical application effect and low recognition accuracy when using edge recognition method to remove strip-shaped electromagnetic interference.

[0006] To achieve the above objectives, this application provides a method for removing electromagnetic interference from weather radar, comprising the following steps: Acquire weather radar detection data, and calculate the echo intensity difference between the same-layer elevation angle radial distance-by-distance database and the adjacent radial distance database based on the detection data; Set a first threshold for the echo intensity difference, a second threshold for the number of distance libraries that meet the first threshold, and a third threshold for the number of effective echo distance libraries in the radial direction; First, count the number of distance libraries in one radial direction that satisfy the condition that the absolute value of the echo intensity difference is greater than the first threshold. Then, count the number of effective echo distance libraries in the radial direction. Based on the number of distances in the first library, the number of effective echo distances in the second library, the second threshold, and the third threshold, determine whether the radial direction is an interfering radial direction, identify all interfering radial directions in sequence, and number them in order of priority. All numbers are grouped, the edge orientation of each group of banner-shaped electromagnetic interference is determined, and interference removal and correction are performed on the banner-shaped electromagnetic interference based on the edge orientation. Radial scatter noise removal is performed on the detection data after interference removal and correction to complete the removal of electromagnetic interference from weather radar detection data.

[0007] Optionally, the method for determining the radial direction of the interference is as follows: Compare the distance library quantity one with the second threshold. If the distance library quantity one is greater than the second threshold, then the radial direction is determined to be an interference radial direction. If the number of distances in the distance library is not greater than the second threshold, the previous radial direction is an interference radial direction, and the number of effective echo distances in the distance library is greater than the third threshold, then the radial direction is determined to be an interference radial direction.

[0008] Optionally, the interference removal method is to set all radial values ​​within the edge orientation to invalid values ​​to complete the interference removal.

[0009] Optionally, the interference correction method is as follows: For any distance library of any interference radial direction, correction is performed using the echoes from the window formed by the distance libraries of the previous radial direction and the next radial direction below the lower edge of the interference radial direction. When the number of valid echoes in the window is less than a preset fourth threshold, no correction is performed, and the invalid value is retained. When the number of valid echoes in the window is greater than or equal to the fourth threshold, correction is performed using the average value of the valid echoes in the window.

[0010] Optionally, the method for grouping all numbers and determining the edge orientation of each group of banner-shaped electromagnetic interference is as follows: different interferences are grouped according to the continuity of the numbers, and the next orientation of the first number and the orientation of the last number in each group are determined as the edge orientation of the interference in that group, thereby realizing the determination of the edge orientation of each group of banner-shaped electromagnetic interference.

[0011] Optionally, the absolute value of the echo intensity difference is calculated using the following expression:

[0012] in This represents the absolute value of the difference in echo intensity between adjacent locations at the same distance. Indicates the first Location, number The echo intensity value corresponding to the distance. Indicates the location number of the weather radar. This indicates the range index number of the weather radar. Indicates the current location The next adjacent location, the first j The echo intensity value corresponding to the distance from the library.

[0013] Optionally, the radial scattered noise removal method is as follows: taking the detection data of the same elevation angle layer as the processing object, by initializing the sliding window and the effective grid point threshold parameter of the neighborhood, the neighborhood information is extracted by traversing the grid point one by one, and the scattered noise is identified and marked as invalid value based on the comparison between the number of effective meteorological echo grid points in the neighborhood and the preset threshold. Then, the blank area is corrected by spatial interpolation method, and finally, all elevation angle layers are processed layer by layer to achieve radial scattered noise removal.

[0014] Optionally, the method for determining the radial direction of the interference is as follows: Set a fifth threshold; Compare the distance library quantity one with the second threshold. If the distance library quantity one is greater than the second threshold, then the radial direction is determined to be an interference radial direction. If the number of distance libraries is not greater than the second threshold, the previous radial direction is an interference radial direction, and the difference between the number of effective echo distance libraries in the previous radial direction and the number of effective echo distance libraries is less than the fifth threshold, then the radial direction is determined to be an interference radial direction.

[0015] Furthermore, to achieve the above objectives, this application also provides a weather radar electromagnetic interference removal device for implementing the aforementioned weather radar electromagnetic interference removal method, comprising: The echo intensity difference calculation module is used to acquire the detection data of the weather radar and calculate the echo intensity difference between the same-layer elevation angle radial distance-by-distance library and the adjacent radial distance library based on the detection data. The threshold setting module is used to set a first threshold for the echo intensity difference, a second threshold for the number of distance libraries that meet the first threshold, and a third threshold for the number of effective echo distance libraries in the radial direction. The quantity statistics module is used to count the number of distance libraries in one radial direction that satisfy the condition that the absolute value of the echo intensity difference is greater than the first threshold, and the number of effective echo distance libraries in that radial direction. The interference radial identification module is used to determine whether a radial direction is an interference radial direction based on the number of distances in the distance library, the number of effective echo distances in the effective echo distance library, the second threshold, and the third threshold, and to identify the interference radial directions in all radial directions in sequence and number them according to their order. The edge orientation determination and interference processing module is used to group all interference radial numbers, determine the edge orientation of each group of strip electromagnetic interference, and perform interference removal and interference correction on the strip electromagnetic interference based on the edge orientation. The scattered noise removal module is used to remove radial scattered noise from the detection data after interference removal and correction, thereby removing electromagnetic interference from the weather radar detection data.

[0016] Optionally, the interference radial identification module determines whether a radial path is an interference radial path by comparing the first distance library quantity with the second threshold. If the first distance library quantity is greater than the second threshold, the radial path is determined to be an interference radial path. If the first distance library quantity is not greater than the second threshold, and the previous radial path is an interference radial path, and the second effective echo distance library quantity is greater than the third threshold, the radial path is determined to be an interference radial path.

[0017] Compared with the prior art, this application has the following beneficial effects: This application provides a method for removing electromagnetic interference from weather radar. By calculating the echo intensity difference between adjacent radial distances, setting three thresholds, performing distance database statistics, identifying the radial interference through dual-judgment logic, locating the edge orientation through interference grouping, and removing radial scattered noise, this method solves the technical problems of poor practical application effect and low recognition accuracy when using existing edge recognition methods to remove strip-shaped electromagnetic interference. This method has clear logic, few parameters, and high execution efficiency. It can adapt to the ideal interference scenario of "covering the entire radial area" and accurately deal with the actual detection scenario of "random and discontinuous distribution of interference", effectively improving the recognition accuracy and removal effect of strip-shaped electromagnetic interference. Attached Figure Description

[0018] Figure 1 This is a flowchart of the weather radar electromagnetic interference removal method in the embodiments of this application; Figure 2 This is a schematic diagram of the structure of the weather radar electromagnetic interference removal device in the embodiments of this application; Figure 3 This is the original echo intensity map in the embodiments of this application; Figure 4 This is the echo image after strip-shaped electromagnetic interference identification, removal, and correction in the embodiments of this application; Figure 5 This is an echo diagram of the strip-shaped electromagnetic interference identification, removal correction, and radial scatter noise removal in the embodiments of this application. Figure 6 This is a grouped echo diagram of the banner-shaped electromagnetic interference in the embodiments of this application; Figure 7 The illustration shows the correction diagram after removing the banner-shaped electromagnetic interference in the embodiments of this application.

[0019] The attached diagram is described below: 1-Echo intensity difference calculation module, 2-Threshold setting module, 3-Quantity statistics module, 4-Interference radial identification module, 5-Edge orientation determination and interference processing module, 6-Scattered noise removal module. Detailed Implementation

[0020] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0021] like Figure 1 As shown, this application provides a method for removing electromagnetic interference from weather radar, including the following steps: S1. Obtain the detection data of the weather radar, and calculate the echo intensity difference between the same-layer elevation angle radial distance-by-distance reservoir and the adjacent radial distance reservoir based on the detection data; S2. Set a first threshold for the echo intensity difference, a second threshold for the number of distance libraries that meet the first threshold, and a third threshold for the number of effective echo distance libraries in the radial direction; S3. Count the number of distance libraries in one radial direction that satisfy the absolute value of the echo intensity difference greater than the first threshold, and then count the number of effective echo distance libraries in the radial direction. S4. Based on the number of distances in the first library, the number of effective echo distances in the second library, the second threshold, and the third threshold, determine whether the radial direction is an interfering radial direction. Identify all interfering radial directions in sequence and number them according to their order. S5. Group all the numbers, determine the edge location of each group of banner-shaped electromagnetic interference, and perform interference removal and correction on the banner-shaped electromagnetic interference according to the edge location. S6. Perform radial scatter noise removal on the detection data after interference removal and correction to complete the removal of electromagnetic interference in the weather radar detection data.

[0022] Specifically, the purpose of step S1 in the weather radar electromagnetic interference removal method provided in this application is to acquire detection data for interference identification and establish a basis for distinguishing between interference and normal echoes. Weather radar detection data is the prerequisite for all subsequent interference identification and processing. One of the core differences between strip-shaped electromagnetic interference and normal meteorological echoes lies in the abrupt change in echo intensity between adjacent radial echoes at the same distance (the intensity of interference echoes often differs significantly from surrounding normal echoes). By calculating the echo intensity difference between adjacent radial echoes at the same distance and azimuth-by-azimuth, this intensity difference can be quantified, providing a quantifiable reference indicator for subsequently determining whether a certain radial direction is an interference radial direction, avoiding identification bias caused by subjective judgment alone.

[0023] The purpose of step S2 is to determine the criteria for interference identification, thereby standardizing and quantifying interference identification. Since the intensity characteristics and distribution patterns of radar echoes differ across regions and weather scenarios, a lack of unified criteria can lead to unstable interference identification results and low accuracy. A first threshold for echo intensity difference is set to distinguish between "normal intensity fluctuations" and "intensity abrupt changes caused by interference." A second threshold is set for the number of range databases that meet the first threshold to avoid misjudging "occasional intensity fluctuations in individual range databases" as interference. A third threshold is set for the number of effective radial echo range databases to filter out radial ranges that "have effective echo data and can be used for interference determination," eliminating interference from missing or invalid data and ensuring the rigor of interference identification.

[0024] The purpose of step S3 is to provide specific data support for the determination of the interference radial direction, realizing "data-quantified determination" instead of "experience-based judgment". Statistical analysis of "the number of distances in the distance database that satisfy the condition that the absolute value of the echo intensity difference is greater than the first threshold" can intuitively reflect the intensity difference between a certain radial direction and its adjacent radial directions. The more significant the difference, the larger the number of distances in the distance database, and the higher the probability that the radial direction is an interference radial direction. Statistical analysis of "the number of effective echo distances in the radial direction" can determine the completeness of the effective data for that radial direction, avoiding misjudgment of radial directions with severe data gaps. It also provides data basis for the subsequent "auxiliary determination when the previous radial direction is interference", solving the problem of "misjudging interference based solely on a single feature" in edge recognition methods.

[0025] Step S4 aims to accurately identify all interfering radial directions, laying the foundation for subsequent interfering grouping and edge localization. This is also a crucial step in addressing the low accuracy of edge recognition methods. Through a dual-judgment logic—comparing the distance database quantity with the second threshold and comparing the previous radial state with the effective echo quantity with the third threshold—it can identify both ideal interfering radial directions with "obvious interference characteristics covering the entire radial direction" and actual interfering radial directions in scenarios where "interference is discontinuous and the intermediate radial interference characteristics are not obvious." This avoids the problem of "misidentifying the intermediate radial direction of interference as an edge radial direction" in edge recognition methods. Furthermore, numbering all interfering radial directions sequentially clearly outlines the distribution pattern of the interference, facilitating subsequent grouping and ensuring a one-to-one correspondence between interference and edge locations.

[0026] Step S5 aims to achieve a precise correspondence between interference and edge azimuth, effectively removing interference and correcting data. This is a core step in solving key technical problems. Edge identification methods suffer from the limitation of "failing to achieve a one-to-one correspondence between interference and edge azimuth," leading to incomplete interference removal or accidental deletion of normal echoes. This step, by "grouping by radial numbering continuity," divides continuously distributed interference into independent groups. Then, it defines the "next azimuth of the first number" and the "azimuth of the last number" in each group as the edge azimuth. This ensures that each strip-shaped electromagnetic interference has a clear and unique edge range, completely resolving the problems of "mismatched edge azimuth and interference correspondence, missed edge detection, and multiple edge detection." Based on this, setting all radial values ​​within the edge azimuth to invalid values ​​quickly and thoroughly removes interference. Correcting the interference area using surrounding normal echoes fills in the gaps left after interference removal, ensuring the continuity and integrity of radar echo data and preventing data loss after interference removal from affecting subsequent weather observations and forecasts.

[0027] The purpose of step S6 is to completely remove residual radial scatter points, further improving the purity of radar echo data. After the processing in step S5, the large-area, continuous strip-shaped electromagnetic interference has been removed, but a small number of sparsely distributed, isolated radial scatter points may still remain. If these radial scatter points are not removed, they will still affect the accuracy of the echo data, leading to deviations in subsequent weather analysis. By using radial scatter point noise removal methods to eliminate the interference of residual radial scatter points, the interference is completely removed, ensuring that the output radar echo data is pure and complete, meeting the needs of weather radar operational observation and subsequent data processing.

[0028] This application provides a method for removing electromagnetic interference from weather radar, aiming to solve the technical problems of poor practical application effect and low recognition accuracy when using edge recognition methods to remove strip-shaped electromagnetic interference. This application forms a complete interference removal process through a progressive and coordinated six-step process: from basic data acquisition and judgment standard setting, to interference radial identification and precise edge location, and then to interference removal, correction, and residual radial noise removal. It solves the problem of "misjudging edges when interference is discontinuous" by identifying the interference radial direction through dual judgment logic; it solves the problem of "mismatch between interference and edges" by locating edge locations through interference grouping; and it ensures complete interference removal through radial noise removal. The entire method has clear logic, few parameters, and high execution efficiency. It can adapt to ideal interference scenarios "covering the entire radial direction" and accurately handle actual detection scenarios with "random and discontinuous interference distribution," effectively improving the recognition accuracy and removal effect of strip-shaped electromagnetic interference, overcoming the inherent defects of edge recognition methods, achieving radar echo data purification, providing reliable data support for subsequent weather monitoring and forecasting, and is easy to implement in engineering and practical applications.

[0029] In one embodiment, the method for determining the radial direction of interference in step S4 is as follows: S401. Compare the distance to the number of libraries with the second threshold. If the distance to the number of libraries is greater than the second threshold, then the radial direction is determined to be an interference radial direction. S402. If the distance from the number of libraries is not greater than the second threshold, the previous radial direction is the interference radial direction, and the effective echo distance from the number of libraries is greater than the third threshold, then the radial direction is determined to be the interference radial direction.

[0030] Specifically, in step S401 of this embodiment, the number of distance libraries represents the total number of distance libraries in a certain radial direction whose absolute value of the echo intensity difference with the adjacent radial direction at the same distance library is greater than the first threshold. This value directly reflects the degree of intensity difference between the current radial direction and the adjacent normal radial direction. When the number of distance libraries is greater than the second threshold, it indicates that there is a significant intensity abrupt change in most distance libraries in the current radial direction, and the difference from the surrounding normal meteorological echoes is significant, exhibiting typical strip-shaped electromagnetic interference characteristics. At this time, it is determined to be an interfering radial direction, which can quickly and accurately identify the ideal interference scenario of "obvious interference characteristics, covering the entire or most of the radial direction", avoiding misjudgment due to the accidental fluctuation of individual distance libraries, while taking into account the identification efficiency and ensuring that obvious interference is not missed.

[0031] Step S402 addresses the actual detection scenario of "discontinuous interference and unclear radial interference features in the middle." This scenario is the core pain point of the low accuracy of edge recognition methods. Edge recognition methods often mistakenly identify such radial interference as normal radial or edge radial interference because the radial interference features are not significant, resulting in incomplete interference identification. When the number of distances in the distance library is not greater than the second threshold, it indicates that the current radial interference features are not obvious and cannot be directly determined by intensity differences alone. At this time, the association judgment of "the previous radial is an interference radial" is introduced. Combined with the condition that "the number of effective echo distances in the distance library is greater than the third threshold," the situation of "continuously distributed interference but weak radial interference features in the middle" can be accurately captured: the previous radial is an interference radial, indicating that the current radial is within the range of continuous interference, and the number of effective echo distances in the distance library meets the standard, indicating that the current radial has sufficient effective data. This eliminates the problem of unclear intensity differences caused by missing data, and thus determines it as an interference radial, completely solving the problem of misjudgment and omission of discontinuous interference by edge recognition methods.

[0032] In summary, this embodiment, through the dual judgment logic of S401 and S402, can efficiently identify radial interference with obvious characteristics and accurately capture complex scenarios with discontinuous interference. It effectively avoids the inherent defects of edge recognition methods, provides accurate radial interference data for subsequent interference grouping and edge localization, lays the foundation for the accuracy of the entire interference removal process, and ultimately improves the overall effect and reliability of strip-shaped electromagnetic interference removal.

[0033] In one embodiment, in step S5, the interference removal method is to set all radial values ​​within the edge orientation to invalid values ​​to complete the interference removal.

[0034] Specifically, this embodiment adopts the method of "setting all radial values ​​within the edge azimuth to invalid values." Essentially, this method directly marks invalid values ​​to completely separate interference signals from normal meteorological echo signals, preventing residual interference signals from affecting subsequent data processing and weather observation. The core characteristic of strip-shaped electromagnetic interference is its continuous radial distribution, concentrated within a specific edge azimuth range. Setting all radial values ​​within this range to invalid values ​​can cover the entire interference area at once, avoiding both "missed interference removal" and accidental deletion of normal echoes outside the edge azimuth. This achieves the dual goals of "complete interference removal and preservation of normal echoes," solving the problem of "incomplete interference removal" or "accidental deletion of normal echoes" caused by edge recognition methods due to edge misjudgment. This method is simple to operate, highly efficient, and adaptable to engineering application requirements.

[0035] In one embodiment, the interference correction method in step S5 is as follows: For any distance library of any interference radial direction, correction is performed using the echoes from the window formed by the distance libraries of the previous radial direction and the next radial direction below the lower edge of the interference radial direction. When the number of valid echoes in the window is less than a preset fourth threshold, no correction is performed, and the invalid value is retained. When the number of valid echoes in the window is greater than or equal to the fourth threshold, correction is performed using the average value of the valid echoes in the window.

[0036] Specifically, this embodiment employs a neighborhood windowing approach for echo correction based on the identified interference radial distance database. A calculation window is formed using data from the same distance database on both sides of the interference radial direction. Interference-affected data is repaired based on surrounding normal echoes, and a fourth threshold is introduced as a criterion for correction execution. The fourth threshold is a pre-set lower limit for the number of valid echoes. When the number of valid echoes within the window is less than this threshold, the sample is insufficient and has no computational significance; therefore, no correction is performed, and invalid values ​​are retained. When the number is greater than or equal to the threshold, the average of the valid echoes within the window is used to complete the correction. Setting a fourth threshold effectively avoids correction distortion caused by missing neighborhood data, prevents the generation of false echoes and secondary noise, and distinguishes different interference distribution scenarios. While repairing scattered interference data, it preserves the characteristics of contiguous interference areas, improving the reliability and adaptability of the overall processing results.

[0037] The fourth threshold is a pre-set lower limit for the number of valid echoes, which is the entry threshold for starting the mean correction algorithm. It is a fixed parameter manually configured based on the radar detection scenario and echo characteristics.

[0038] The core technical objectives of setting the fourth threshold include: (1) avoiding correction distortion caused by missing neighboring data. Strip-shaped electromagnetic interference is often distributed in multiple continuous radial areas. If the interference area is large, the selected correction window will also be covered by interference, resulting in too few effective meteorological echoes or even all invalid values. At this time, the number of samples in the window is insufficient, and the mean calculation is not statistically significant. Forcibly taking the mean will generate false echoes and confuse the real meteorological information. The fourth threshold directly intercepts such invalid corrections by setting a minimum number of samples. (2) ensuring the reliability and authenticity of the correction data. Effective radar echoes represent real meteorological targets such as raindrops and snowflakes. Only when the number of effective echo samples in the window reaches a certain amount can the mean objectively reflect the intensity level of the real meteorological echoes at that azimuth and distance. The fourth threshold ensures that the correction results are consistent with the actual weather conditions from the perspective of sample size. (3) Distinguish between areas with dense interference and areas with sparse interference, and adapt to different interference scenarios. When electromagnetic interference is distributed in a large area: there are few effective echoes in the window, which are below the fourth threshold. Invalid values ​​are retained, and the interference area is accurately marked without destroying the original data characteristics. When electromagnetic interference is scattered and only a few radial areas are affected: there are enough normal echoes in the window, which meet the fourth threshold. Mean correction is performed to repair the missing meteorological data.

[0039] Assuming the preset fourth threshold = 3: Scenario 1 (Correctable): 5 valid echoes are obtained in the correction window, and the number is ≥3 → Calculate the average of the 5 echoes to complete the distance library correction; Scenario 2 (Uncorrectable): If there is only 1 valid echo in the correction window, and the number is less than 3, correction is abandoned. The distance library remains invalid, and the interference marker is retained.

[0040] In one embodiment, in step S5, the method for grouping all numbers and determining the edge orientation of each group of banner-shaped electromagnetic interference is as follows: different interferences are grouped according to the continuity of the numbers, and the next orientation of the first number and the orientation of the last number in each group are determined as the edge orientation of the interference in that group, thereby realizing the determination of the edge orientation of each group of banner-shaped electromagnetic interference.

[0041] Specifically, the strip-shaped electromagnetic interference is distributed radially in a continuous manner, and its radial numbering is continuous. Grouping based on this numbering continuity allows for precise differentiation of individual strip-shaped electromagnetic interferences, avoiding confusion between different interference edge locations and resolving the pain point of miscorrelation between interference and edges in edge recognition methods. Defining the location of the next position after the first number and the location of the last number in each group as the edge location accurately captures the start and end boundaries of the interference, completely encompassing the interference area without including the normal radial direction, thus avoiding edge misjudgment. This method is logically simple and accurately positioned, providing a precise range basis for subsequent interference removal and correction, balancing accuracy and engineering efficiency, and effectively compensating for the inherent shortcomings of edge recognition methods.

[0042] In one embodiment, the absolute value of the echo intensity difference is calculated using the following equation (1): (1) In formula (1) This represents the absolute value of the difference in echo intensity between adjacent locations at the same distance. Indicates the first Location, number The echo intensity value corresponding to the distance. Indicates the location number of the weather radar. This indicates the range index number of the weather radar. Indicates the current location The next adjacent location, the first j The echo intensity value corresponding to the distance from the library.

[0043] Specifically, the core difference between strip-shaped electromagnetic interference and normal meteorological echoes is the abrupt change in echo intensity at the same distance from adjacent radial locations. This intensity difference can be quantified by calculating the echo intensity difference at the same distance from adjacent locations using equation (1), providing a core indicator for subsequent radial interference identification. The parameters in equation (1) are clearly defined. , By comparing the echo intensity of the same range database in the current and adjacent azimuth directions, the comparison target is accurately identified, avoiding invalid comparisons across range databases and azimuths, thus ensuring the accuracy of the calculation results. This calculation method is simple and efficient, requiring no complex calculations, and can be executed quickly azimuth-by-azimuth and range database-by-range method. It is suitable for real-time radar echo processing needs, laying a solid foundation for subsequent statistics on the number of range databases that meet the threshold and for determining the radial direction of interference. It effectively avoids the problem of misjudgment of interference caused by inaccurate intensity difference quantification in edge recognition methods.

[0044] In one embodiment, in step S6, the method for removing radial scattered noise is as follows: taking the detection data of the same elevation angle layer as the processing object, by initializing the sliding window and the effective grid point threshold parameter of the neighborhood, the neighborhood information is extracted by traversing the grid point one by one. Based on the comparison between the number of effective meteorological echo grid points in the neighborhood and the preset threshold, scattered noise is identified and marked as invalid values. Then, the blank area is corrected by spatial interpolation. Finally, all elevation angle layers are processed layer by layer to achieve the removal of radial scattered noise.

[0045] Specifically, in this embodiment, the radial scattered noise removal method addresses the core problem of "poor practical application effect and low recognition accuracy when using edge recognition to remove strip-shaped electromagnetic interference," serving as a supplement and improvement to interference removal. After interference removal and correction in step S5, sparse and isolated radial scattered points remain, affecting the purity of echo data, which edge recognition cannot effectively handle. This method uses data from the same elevation angle layer as the processing object, aligning with the inter-layer independence characteristics of radar detection data. By initializing the sliding window and neighborhood threshold, a standard for radial scattered point discrimination is provided. Neighborhood information is extracted by traversing each grid point, accurately distinguishing isolated radial scattered points from continuous normal echoes. Based on the comparison between the number of effective echoes in the neighborhood and the threshold, radial scattered points are marked and set as invalid values. Spatial interpolation is used to correct blank areas, and then all elevation angle layers are processed layer by layer. This thoroughly removes radial scattered points while ensuring echo continuity. The method is simple and efficient, further compensating for the shortcomings of edge recognition, ensuring complete removal of electromagnetic interference, and guaranteeing accurate and reliable radar echo data.

[0046] Spatial interpolation is used to correct blank areas. On the one hand, it can fill the data gaps caused by the removal of scattered noise and ensure the integrity and continuity of radar echo data. On the other hand, it can restore the true echo intensity by combining the spatial distribution pattern of meteorological echoes, avoid blank grid points from distorting meteorological characteristics, and eliminate the interference of missing data on subsequent operational operations such as precipitation estimation and weather identification, thereby reducing secondary errors.

[0047] Processing is performed layer by layer according to elevation angle, which conforms to the radar's layered scanning and the independent detection principle of data at each altitude level. This can achieve comprehensive removal of scattered noise across the entire airspace and eliminate residual noise at local elevation angles. At the same time, the processing rules are uniformly implemented for each layer, ensuring that the quality of echo data at different altitudes remains consistent, and guaranteeing the accuracy and effectiveness of subsequent applications such as three-dimensional echo analysis and stereoscopic meteorological mosaics.

[0048] In one embodiment, the method for determining the radial direction of interference in step S4 is as follows: S401, Set the fifth threshold; S402. Compare the distance to the number of libraries with the second threshold. If the distance to the number of libraries is greater than the second threshold, then the radial direction is determined to be an interference radial direction. S403. If the number of distances to the library is not greater than the second threshold, the previous radial direction is an interference radial direction, and the difference between the number of effective echo distances to the library and the number of effective echo distances to the library in the previous radial direction is less than the fifth threshold, then the radial direction is determined to be an interference radial direction.

[0049] Specifically, in this embodiment, step S401 pre-sets a fifth threshold, which serves as a numerical judgment benchmark and is used to assist in subsequent discrimination based on the difference in the number of effective echoes in adjacent radial directions. Since actual strip-shaped electromagnetic interference is often continuously distributed, a single indicator cannot accurately identify edges and weak interference radial directions. Therefore, this threshold is added to provide a quantitative basis for determining the continuity of continuous interference radial directions and improve the completeness of identification.

[0050] Step S402 uses the number of distance database entries and the second threshold as the core judgment criteria. The more distance database entries with excessive inner echo intensity differences in a single radial direction, the more obvious the abnormal characteristics of the radial data. When the number of entries exceeds the second threshold, it indicates that the proportion of abnormal points in the radial direction is high, which is consistent with the data characteristics of strip-shaped electromagnetic interference. Therefore, the radial direction is directly determined to be an interfering radial direction, realizing the rapid identification of typical strong interference radial directions.

[0051] Step S403 makes supplementary judgments for weak interference and radial interference edges that cannot be determined by a single indicator. When the current radial abnormal distance library quantity does not meet the standard, but the previous radial direction has been confirmed as an interference radial direction, and the difference between the effective echo distance library quantities of the two is less than the fifth threshold, it indicates that the two radial data characteristics are highly similar and belong to the category of continuous interference. Therefore, the current radial direction is also determined to be an interference radial direction, effectively identifying discontinuous and gradual strip interference and making up for the shortcomings of a single judgment rule.

[0052] The fifth threshold is used to distinguish between continuous radial interference and normal meteorological echoes. Striped electromagnetic interference is distributed continuously in strips, and the difference in the effective echo distance and number of adjacent interference radial directions is minimal; while real meteorological echoes such as clouds and rain will naturally fluctuate in adjacent directions, and the difference in number is significantly greater. This threshold is the critical value for distinguishing between the two.

[0053] The fifth threshold is a non-negative integer, and its magnitude directly determines the severity of the judgment. The smaller the threshold, the stricter the judgment, only identifying continuous interference with highly consistent data, resulting in a low false positive rate and easy to miss weak interference; The larger the threshold, the wider the judgment range, the more complete the entire interference band can be captured, the lower the missed judgment rate, and the easier it is to misjudge normal echoes.

[0054] Based on domestic conventional weather radar engineering applications, a general range of values ​​is given: Standard operational radar (range 50m-100m, unidirectional radial range 300-800 units): Recommended value 15-25; For short-range detection radar and close-range observation scenarios (limited single-radial range library): recommended value is 10-20; For long-range detection radars and long-range echoes that exhibit natural jitter, a value of 20-30 is recommended.

[0055] like Figure 2 As shown, this application provides a weather radar electromagnetic interference removal device, which can implement the above-described weather radar electromagnetic interference removal method. The device includes: The echo intensity difference calculation module 1 is used to acquire the detection data of the weather radar and calculate the echo intensity difference between the same-layer elevation angle radial distance-by-distance library and the adjacent radial distance library based on the detection data. The threshold setting module 2 is used to set a first threshold for the echo intensity difference, a second threshold for the number of distance libraries that meet the first threshold, and a third threshold for the number of effective echo distance libraries in the radial direction. The quantity statistics module 3 is used to count the number of distance libraries in one radial direction that satisfy the condition that the absolute value of the echo intensity difference is greater than the first threshold, and the number of effective echo distance libraries in that radial direction. Interference radial identification module 4 is used to determine whether the radial direction is an interference radial direction based on the number of distances in the distance library, the number of effective echo distances in the distance library, the second threshold, and the third threshold, and to identify the interference radial directions in all radial directions in sequence and number them in order. The edge orientation determination and interference processing module 5 is used to group all the radial numbers of the interference, determine the edge orientation of each group of strip electromagnetic interference, and perform interference removal and interference correction on the strip electromagnetic interference according to the edge orientation. The scattered noise removal module 6 is used to remove radial scattered noise from the detection data after interference removal and correction, thereby completing the removal of electromagnetic interference from the weather radar detection data.

[0056] Specifically, in this embodiment, the echo intensity difference calculation module 1 is the core of basic data processing. Its function corresponds to step S1 in the method. By accurately acquiring radar detection data and calculating the echo intensity difference of adjacent radial distance libraries in a radial and distance-by-distance library, the intensity difference between interference and normal echo is quantified, providing core data support for subsequent interference identification. This solves the identification deviation problem caused by the lack of quantitative basis for intensity difference in the edge identification method and ensures the accuracy of subsequent module processing.

[0057] The threshold setting module 2 is the core of the standard setting for interference identification. It corresponds to step S2 in the method. By setting the first threshold, the second threshold, and the third threshold, it provides a unified and quantifiable judgment standard for judging the radial direction of interference and screening effective echoes. It adapts to the radar echo characteristics of different regions and different weather scenarios, avoids misjudgment and missed judgment of interference due to ambiguity of the judgment standard, and ensures the universality of the device and the rigor of the identification.

[0058] The quantity statistics module 3 is a key data support module for interference determination, corresponding to step S3 in the method. By statistically analyzing the number of distances in the first library that meet the intensity difference threshold and the number of effective echo distances in the second library, it provides specific determination data for the interference radial identification module 4, realizing "data quantification determination" instead of "experience judgment", further improving the accuracy of interference identification and avoiding the errors caused by subjective judgment in the edge identification method.

[0059] Interference radial identification module 4 is the core identification module of the device, corresponding to step S4 in the method. By combining the data provided by the quantity statistics module 3 and the threshold set by the threshold setting module 2, it uses dual judgment logic to identify and number the interference radials. It can identify radials with obvious interference characteristics and capture complex scenes with discontinuous interference. It solves the pain points of low accuracy of interference identification and easy misjudgment of the middle radials in the edge identification method, and provides accurate interference radial information for subsequent edge positioning and interference processing.

[0060] The edge orientation determination and interference processing module 5 is the core execution module for interference removal, corresponding to step S5 in the method. It determines the edge orientation by grouping the interference radially, realizing a one-to-one correspondence between the interference and the edge orientation, and completely solving the problem of disordered correspondence between interference and edge in the edge recognition method. At the same time, it performs interference removal and correction operations, which not only completely removes the interference signal, but also fills the gap with the surrounding normal echoes, ensuring the continuity of echo data and laying the foundation for subsequent scattered noise removal.

[0061] The scattered noise removal module 6 is a supplementary module of the device, corresponding to step S6 in the method. For radial scattered points that are not residual after interference processing in S5, residual interference is completely eliminated through neighborhood discrimination, invalidation marking and interpolation correction. This further improves the purity of radar echo data, makes up for the deficiency of edge recognition method in handling sparse scattered interference, and ensures that electromagnetic interference is completely removed.

[0062] In summary, the six modules of this device have clear division of labor and are progressively integrated, precisely corresponding to each step of the aforementioned weather radar electromagnetic interference removal method. They work together to achieve the entire process of interference identification, edge localization, interference removal, correction, and radial scatter point clearing, effectively solving the inherent defects of the edge identification method. This approach balances identification accuracy with engineering execution efficiency, ensuring that the device can stably and efficiently complete weather radar electromagnetic interference removal, providing reliable support for weather radar operational observations.

[0063] In one embodiment, the interference radial identification module 4 determines whether a radial path is an interference radial path by comparing the distance library quantity one with the second threshold. If the distance library quantity one is greater than the second threshold, the radial path is determined to be an interference radial path. If the distance library quantity one is not greater than the second threshold, and the previous radial path is an interference radial path and the effective echo distance library quantity two is greater than the third threshold, the radial path is determined to be an interference radial path.

[0064] Specifically, in this embodiment, the interference radial identification module 4 accurately distinguishes between interference radials and normal radials. Its dual-judgment logic aligns with the actual distribution characteristics of strip-shaped electromagnetic interference. The distance library quantity one reflects the intensity difference between the current radial and adjacent radials. When the distance library quantity one is greater than the second threshold, it indicates that most distance libraries in the current radial have significant intensity abrupt changes, exhibiting typical interference characteristics. In this case, it is determined to be an interference radial, allowing for rapid identification of radials with significant interference characteristics, avoiding missed detections while maintaining efficiency. When the distance library quantity one is not greater than the second threshold, it indicates that the current radial interference characteristics are not obvious. In this case, the association judgment of "the previous radial is an interference radial" and the condition of "the effective echo distance library quantity two is greater than the third threshold" are introduced. This accurately captures scenarios where interference is continuously distributed but the intermediate radial interference characteristics are weak, eliminating the possibility of insignificant intensity differences due to missing data measurements, and avoiding the problem of edge identification methods mistakenly classifying such radials as normal radials or edge radials. This judgment method is precisely linked with the threshold setting module 2 and the quantity statistics module 3. It relies on quantitative data to make judgments instead of experience-based judgments, which greatly improves the accuracy of interference identification and provides accurate radial information of interference for subsequent edge location determination and interference processing, effectively making up for the inherent defects of the edge recognition method.

[0065] Example 1 This embodiment provides a method for removing electromagnetic interference from weather radar, including the following steps: Step 1: Obtain the detection data of the same elevation angle of the weather radar, covering 360 radial (azimuth number 1-360) and the corresponding distance database data of each radial. Calculate the echo intensity difference between the current radial and the adjacent radial at the same distance database according to radial and distance database. Use formula (1) to calculate the absolute value of the echo intensity difference.

[0066] Step 2: Based on the radar echo climate characteristics of the area where the weather radar is located, set three thresholds: the first threshold is set to 20 dBZ, the second threshold is set to 150, the third threshold is set to 180, and the fifth threshold is set to 15 for continuous interference radial auxiliary judgment; at the same time, set the window size for interference correction to 3 (i.e., each window contains the target range database and one range database before and after it), and set the fourth threshold for judging the number of valid echoes for interference correction to 4, to ensure that the setting of all the above thresholds fits the actual detection scenario and provides a unified standard for subsequent interference identification and correction.

[0067] Step 3: For the absolute value of the echo intensity difference calculated in Step 1, count two key quantities radially: First, the number of distance libraries in each radial direction that satisfy the absolute value of the echo intensity difference greater than 20BZ; Second, the number of effective echo distance libraries in each radial direction (the total number of distance libraries after removing missing or invalid data).

[0068] Step 4: Combining the statistical data from Step 3 and the three thresholds set in Step 2, determine whether each radial direction is an interfering radial direction, and number all interfering radial directions in order of their azimuth sequence.

[0069] Based on the analysis, the interference radial lines identified in this detection data are numbered as follows: [1, 2, 4, 5, 12, 13, 21, 22, 34, 35, 41, 42, 43, 44, 45, 46, 48, 49, 78, 79, 80, 81, 82, 83, 84, 162, 163, 186, 187, 188, 351, 352, 353], totaling 33 interference radial lines.

[0070] Step 5: Interference radial grouping, edge orientation determination, and interference removal and correction 1. Interference Grouping: The 33 radial interferences identified in step S4 are grouped into 11 groups, as follows: (1) [1, 2], (2) [4, 5], (3) [12, 13], (4) [21, 22], (5) [34, 35], (6) [41, 42, 43, 44, 45, 46], (7) [48, 49], (8) [78, 79, 80, 81, 82, 83, 84], (9) [162, 163], (10) [186, 187, 188], (11) [351, 352, 353]. Each group corresponds to an independent strip-shaped electromagnetic interference, realizing a one-to-one correspondence between interference and grouping.

[0071] 2. Edge location determination: The edge location of the 11 groups of strip-shaped electromagnetic interference is determined by the method of "determining the next location of the first number and the location of the last number in each group as the edge location of the interference". (The front edge is the next location of the first number in each group, and the back edge is the location of the last number in each group). For example: Group (1) [1, 2], the front edge is the next location of 1 (location 2), and the back edge is the location of 2 (location 2); Group (2) [4, 5], the front edge is the next location of 4 (location 5), and the back edge is the location of 5 (location 5); Group (8) [78, 79, 80, 81, 82, 83, 84], the front edge is the next location of 78 (location 79), and the back edge is the location of 84 (location 84). The edge location of each group accurately covers the corresponding interference area without omission or redundancy.

[0072] 3. Interference Removal: The method of "setting all radial values ​​within the edge azimuth to invalid values" was adopted to invalidate all radial values ​​within the edge azimuth of the 11 groups of strip-shaped electromagnetic interference. Taking the second group [4, 5] (single radial interference, edge azimuth is azimuth 5) as an example, the echo values ​​of all distances in azimuth 5 were set to invalid values ​​to complete the removal of the interference in this group; the other groups were treated in the same way, setting the radial values ​​within the corresponding edge azimuth to invalid values ​​to completely remove the strip-shaped electromagnetic interference signal.

[0073] 4. Interference Correction: Correction is performed on each distance library for all interference radial directions. The window size is set to 3, that is, the window of each distance library to be corrected consists of the distance libraries before and after the distance library of the same distance library of the previous radial direction of the interference radial direction, and the distance libraries before and after the distance library of the same distance library of the next radial direction of the rear edge. Taking the group (2) [4, 5] (single radial interference, edge azimuth is azimuth 5) as an example, the distance library to be corrected is all invalid distance libraries of azimuth 5. Its window composition is as follows: the distance libraries before and after the distance library of the same distance library of the previous radial direction (azimuth 4) of the interference radial direction (azimuth 5) form window 1, and the distance libraries before and after the distance library of the same distance library of the next radial direction (azimuth 6) of the rear edge (azimuth 5) form window 2. The number of valid echoes in the two windows is counted. If the number of valid echoes is not less than the fourth threshold 4, the average value of all valid echoes in the two windows is taken to correct the distance library; if the number of valid echoes is less than 4, the invalid value is kept and no correction is performed. The other interference groups are corrected one by one according to this logic.

[0074] like Figure 7 As shown, assuming The azimuth is the interference azimuth and the interference is a uniradial interference (e.g., the interference in group (2) in this example). For the distance library that needs correction, after interference removal, , , All are set to invalid values. Assuming the window size is 3, the window formed by the distances before and after the previous radial distance library of the interference is [ The window formed by the front and rear distance libraries of the next radial distance library with the trailing edge orientation (in this example, the front and rear edges of the interference are the same) is [ The above two sets of windows are used to count all valid values. If the number of valid values ​​is greater than or equal to the fourth threshold of 4, the mean of all valid values ​​is used to correct the distance library; otherwise, it remains invalid, and no further correction is performed. The corrected echo map is shown below. Figure 4 As shown.

[0075] Step 6: Using the detection data after interference removal and correction at the same elevation angle layer as the processing object, initialize the 3×3 sliding window and the effective grid point discrimination threshold in the neighborhood (6), traverse the grid point to extract neighborhood information, identify and mark the radial scattered points and set them as invalid values, then correct the blank area by spatial interpolation, and finally complete the processing of all elevation angle layers to achieve the complete removal of radial scattered point noise.

[0076] This embodiment completes the joint determination of interference radial lines through steps one to four, combined with the fifth threshold 15, and accurately identifies 33 interference radial lines, solving the problem of low interference identification accuracy of the edge identification method; through grouping and edge positioning in step five, a one-to-one correspondence between 11 groups of interference and edge orientation is achieved, avoiding edge identification errors; relying on the fourth threshold 4 to control the correction logic, interference signals are completely stripped and gaps are filled to ensure the continuity of echo data; through scattered noise removal in step six, residual interference is removed to ensure the purity of echo data.

[0077] Figure 3 This is the original echo intensity map in this embodiment. Figure 4 This is the echo image after strip-shaped electromagnetic interference identification, removal, and correction in this embodiment. Figure 5 This is the echo map after strip-shaped electromagnetic interference identification, removal correction, and removal of radial scatter noise in this embodiment. Figure 6 This is the echo diagram of the strip-shaped electromagnetic interference group in this embodiment. Figure 7 This is an illustration of the correction after removing the banner-shaped electromagnetic interference in this embodiment.

[0078] From the perspective of implementation results, after processing with the above method in this embodiment, the original echo intensity map (as shown in the figure) is significantly improved. Figure 3 The strip-shaped electromagnetic interference shown in the figure (corresponding to) Figure 6 The radial lines marked in the box have all been effectively removed, and the corrected echo map (such as...) Figure 4As shown, there is no obvious residual interference, and the echo data is continuous and accurate, verifying the effectiveness of the above method. Among them, the correction process of the single radial interference in group (2) fully demonstrates the practicality of the interference correction method. By correcting the effective echo mean through the window, the blank defect after interference removal is effectively made up for, further proving that the above method provided in this application can effectively solve the core technical problems of poor practical application effect and low recognition accuracy of edge recognition method, and the logic is clear and easy to implement in engineering.

[0079] The above are merely preferred embodiments of this application and do not limit the patent scope of this application. Any equivalent structural or procedural transformations made using the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.

Claims

1. A method for removing electromagnetic interference from weather radar, characterized in that, Includes the following steps: Acquire weather radar detection data, and calculate the echo intensity difference between the same-layer elevation angle radial distance-by-distance database and the adjacent radial distance database based on the detection data; Set a first threshold for the echo intensity difference, a second threshold for the number of distance libraries that meet the first threshold, and a third threshold for the number of effective echo distance libraries in the radial direction; First, count the number of distance libraries in one radial direction that satisfy the condition that the absolute value of the echo intensity difference is greater than the first threshold. Then, count the number of effective echo distance libraries in the radial direction. Based on the number of distances in the first library, the number of effective echo distances in the second library, the second threshold, and the third threshold, determine whether the radial direction is an interfering radial direction, identify all interfering radial directions in sequence, and number them in order of priority. All numbers are grouped, the edge orientation of each group of banner-shaped electromagnetic interference is determined, and interference removal and correction are performed on the banner-shaped electromagnetic interference based on the edge orientation. Radial scatter noise removal is performed on the interference-removed and interference-corrected detection data to complete the removal of electromagnetic interference from weather radar detection data.

2. The weather radar electromagnetic interference removal method as described in claim 1, characterized in that, The method for determining the radial direction of the interference is as follows: Compare the distance library quantity one with the second threshold. If the distance library quantity one is greater than the second threshold, then the radial direction is determined to be an interference radial direction. If the number of distances in the distance library is not greater than the second threshold, the previous radial direction is an interference radial direction, and the number of effective echo distances in the distance library is greater than the third threshold, then the radial direction is determined to be an interference radial direction.

3. The method for removing electromagnetic interference from weather radar as described in claim 1, characterized in that, The interference removal method is as follows: set all radial values ​​within the edge orientation to invalid values ​​to complete the interference removal.

4. The weather radar electromagnetic interference removal method as described in claim 1, characterized in that, The interference correction method is as follows: For any distance library of any interference radial direction, correction is performed using the echoes from the window formed by the distance libraries of the previous radial direction and the next radial direction below the lower edge of the interference radial direction. When the number of valid echoes in the window is less than a preset fourth threshold, no correction is performed, and the invalid value is retained. When the number of valid echoes in the window is greater than or equal to the fourth threshold, correction is performed using the average value of the valid echoes in the window.

5. The method for removing electromagnetic interference from weather radar as described in claim 1, characterized in that, The method for grouping all numbers and determining the edge orientation of each group of banner-shaped electromagnetic interference is as follows: different interferences are grouped according to the continuity of the numbers, and the next orientation of the first number and the orientation of the last number in each group are determined as the edge orientation of the interference in that group, thereby realizing the determination of the edge orientation of each group of banner-shaped electromagnetic interference.

6. The method for removing electromagnetic interference from weather radar as described in claim 1, characterized in that, The absolute value of the echo intensity difference is calculated using the following expression: in This represents the absolute value of the difference in echo intensity between adjacent locations at the same distance. Indicates the first Location, number The echo intensity value corresponding to the distance. Indicates the location number of the weather radar. This indicates the range index number of the weather radar. Indicates the current location The next adjacent location, the first j The echo intensity value corresponding to the distance from the library.

7. The weather radar electromagnetic interference removal method as described in any one of claims 1-6, characterized in that, The method for removing radial scattered noise is as follows: taking the detection data of the same elevation angle layer as the processing object, by initializing the sliding window and the threshold parameter of the effective grid point in the neighborhood, the neighborhood information is extracted by traversing the grid point one by one. Based on the comparison between the number of effective meteorological echo grid points in the neighborhood and the preset threshold, scattered noise is identified and marked as invalid values. Then, the blank area is corrected by spatial interpolation. Finally, all elevation angle layers are processed layer by layer to achieve the removal of radial scattered noise.

8. The method for removing electromagnetic interference from weather radar as described in claim 2, characterized in that, The method for determining the radial direction of the interference is as follows: Set a fifth threshold; Compare the distance library quantity one with the second threshold. If the distance library quantity one is greater than the second threshold, then the radial direction is determined to be an interference radial direction. If the number of distance libraries is not greater than the second threshold, the previous radial direction is an interference radial direction, and the difference between the number of effective echo distance libraries in the previous radial direction and the number of effective echo distance libraries is less than the fifth threshold, then the radial direction is determined to be an interference radial direction.

9. A weather radar electromagnetic interference removal device, used to implement the weather radar electromagnetic interference removal method as described in any one of claims 1-8, characterized in that, include: The echo intensity difference calculation module is used to acquire the detection data of the weather radar and calculate the echo intensity difference between the same-layer elevation angle radial distance-by-distance library and the adjacent radial distance library based on the detection data. The threshold setting module is used to set a first threshold for the echo intensity difference, a second threshold for the number of distance libraries that meet the first threshold, and a third threshold for the number of effective echo distance libraries in the radial direction. The quantity statistics module is used to count the number of distance libraries in one radial direction that satisfy the condition that the absolute value of the echo intensity difference is greater than the first threshold, and the number of effective echo distance libraries in that radial direction. The interference radial identification module is used to determine whether a radial direction is an interference radial direction based on the number of distances in the distance library, the number of effective echo distances in the effective echo distance library, the second threshold, and the third threshold, and to identify the interference radial directions in all radial directions in sequence and number them according to their order. The edge orientation determination and interference processing module is used to group all interference radial numbers, determine the edge orientation of each group of strip electromagnetic interference, and perform interference removal and interference correction on the strip electromagnetic interference based on the edge orientation. The scattered noise removal module is used to remove radial scattered noise from the detection data after interference removal and correction, thereby removing electromagnetic interference from the weather radar detection data.

10. The weather radar electromagnetic interference removal device as described in claim 9, characterized in that, The specific method by which the interference radial identification module determines whether a radial path is an interference radial path is as follows: compare the size of the distance library quantity one with the second threshold. If the distance library quantity one is greater than the second threshold, then the radial path is determined to be an interference radial path. If the distance library quantity one is not greater than the second threshold, and the previous radial path is an interference radial path, and the effective echo distance library quantity two is greater than the third threshold, then the radial path is determined to be an interference radial path.

Images