A weather radar echo type identification method, device, medium and equipment

Abstract

The application discloses a weather radar echo type identification method and device, a medium and equipment. The application obtains preset polarization radar observation data, accurately calculates the distribution characteristics of ocean clutter by using a statistical method, and further constructs a membership function, thereby providing a quantitative basis for a fuzzy logic algorithm. Through fuzzy logic algorithm processing, the membership degree of each pixel belonging to a preset echo category is obtained, and finally the echo category is determined according to the membership degree. The application solves the problem that the echo type of the weather radar cannot be accurately identified in the prior art.

Description

Technical Field

[0001] This invention relates to the field of weather radar echo type identification, and more particularly to a method, apparatus, medium, and device for weather radar echo type identification. Background Technology

[0002] Weather radar is an indispensable tool in meteorological observation. It detects various particles in the atmosphere, such as clouds, precipitation particles, and wildfire plume particles, by emitting and receiving electromagnetic waves. It can not only monitor weather conditions in real time but also provide meteorological early warning information, and is widely used in fields such as wildfire plume monitoring and early warning, and forest disaster prevention and mitigation. With the development of technology, dual-polarization radar has gradually become mainstream. In addition to reflectivity and radial wind speed, dual-polarization radar can also provide observations such as differential reflectivity, correlation coefficient, and differential phase ratio, providing a solid data foundation for distinguishing different types of particles (such as rain, snow, hail, and non-meteorological particles).

[0003] Numerous weather radars have been deployed along China's coastal areas. When observing at low elevation angles, sea clutter (i.e., ocean echoes) often affects the accuracy of radar observations. Sea clutter has the following characteristics: (1) its occurrence time is uncertain; (2) it usually occurs near low elevation angles (0.5° to 1.5°), and is more common within 100 kilometers of the radar station; (3) for large-area sea clutter, the average intensity is relatively weak, but the intensity of individual peaks can sometimes be extremely high; (4) sea clutter has certain Doppler radial velocity characteristics.

[0004] To address the problem of sea clutter identification, several studies have proposed different solutions. For example, Steiner et al. used a decision tree method to study a series of reflectivity parameters and found that the vertical extension of radar echoes, the horizontal variation of the reflectivity factor field, and the vertical gradient of reflectivity are the most effective ways to identify sea clutter. However, sea clutter identification methods based on polarization radar observations still warrant further investigation.

[0005] While existing technologies can identify sea clutter to some extent, these methods have some limitations:

[0006] Limited recognition accuracy: Existing methods have low recognition accuracy for sea clutter under complex weather conditions and are prone to misjudgment.

[0007] High computational complexity: Some methods involve large amounts of computation, making it difficult to process large amounts of radar data in real time.

[0008] Poor adaptability: Existing methods are poorly adaptable to radar data from different regions, making it difficult to maintain stable recognition results in different geographical environments.

[0009] These issues prevent existing technologies from accurately identifying the types of echoes from weather radar. Summary of the Invention

[0010] This invention provides a method, apparatus, medium, and device for identifying weather radar echo types, in order to solve the problem that existing technologies cannot accurately identify the echo types of weather radar.

[0011] Firstly, this application provides a method for identifying weather radar echo types, including:

[0012] Acquire preset polarization radar observation data;

[0013] Based on the polarization radar observation data and statistical methods, the distribution characteristics of the ocean echoes from the polarization radar observation data were calculated.

[0014] Based on the distribution characteristics of the ocean echoes, a membership function is constructed;

[0015] The polarization radar observation data is processed according to the fuzzy logic algorithm and the membership function to obtain the membership degree of each pixel of the polarization radar observation data belonging to the preset echo category.

[0016] Based on the membership degree of each pixel to a preset echo category, the various echo categories of the polarization radar observation data are determined.

[0017] This application ensures the representativeness and timeliness of pre-defined polarization radar observation data, providing a solid foundation for subsequent analysis. Then, statistical methods are used to calculate the ocean clutter distribution characteristics of the polarization radar observation data, providing a scientific basis for constructing membership functions. The construction of membership functions quantifies the characteristic ambiguity of ocean clutter, laying the foundation for the application of fuzzy logic algorithms. By combining fuzzy logic algorithms with membership functions, the polarization radar observation data is processed to obtain the membership degree of each pixel belonging to a pre-defined echo category. This process effectively handles the uncertainty and ambiguity of the data, improving the accuracy and robustness of identification. Finally, each echo category is determined based on the membership degree, ensuring the scientific nature and accuracy of the classification. This application solves the problem in existing technologies that cannot accurately identify the echo type of weather radar.

[0018] As a preferred embodiment of the first aspect, after determining the echo categories of the polarization radar observation data based on the membership degree of each pixel to a preset echo category, the method further includes:

[0019] Obtain the underlying surface type information;

[0020] Based on the underlying surface type information, echoes that do not conform to the underlying surface type in each echo category of the polarization radar observation data are removed.

[0021] In this preferred embodiment, after determining the various echo categories of the polarization radar observation data, this application further acquires underlying surface type information and eliminates echoes that do not conform to the underlying surface type based on this information. This process significantly improves the accuracy and reliability of echo identification. Specifically, by acquiring underlying surface type information, we can distinguish different surface features, such as ocean, land, and cities. Combining this information, we can more accurately identify and eliminate those unreasonable echo types on specific underlying surfaces, such as eliminating ocean clutter appearing on land. This step not only reduces misjudgments but also further purifies the radar data, improving the purity and quality of the data.

[0022] As a preferred embodiment of the first aspect, the acquisition of preset polarization radar observation data specifically includes:

[0023] Acquire initial polarization radar observation data at a preset time;

[0024] Based on the time, location, intensity, and Doppler characteristics of the ocean echoes, the preset polarization radar observation data is obtained by filtering from the initial polarization radar observation data.

[0025] In this preferred embodiment, this application obtains high-quality pre-defined polarization radar observation data by acquiring initial polarization radar observation data at a preset time and filtering it based on the time, location, intensity, and Doppler characteristics of ocean echoes. This process first ensures the timeliness and representativeness of the data. By limiting the time frame, the collected data reflects the meteorological characteristics within a specific time period, providing an accurate temporal background for subsequent analysis. Next, through a filtering step, observation data related to ocean echoes is accurately extracted from a large amount of initial data using the characteristic parameters of ocean echoes. This filtering mechanism effectively removes irrelevant data, reduces data noise, and improves the purity and relevance of the data. The filtered data not only reduces the computational load of subsequent processing but also improves the efficiency and accuracy of data processing. Therefore, this processing step of the present invention significantly improves the quality of polarization radar observation data, providing a solid data foundation for subsequent echo type identification and classification, thereby improving the application effect and reliability of weather radar in meteorological monitoring and early warning.

[0026] In a preferred embodiment of the first aspect, the distribution characteristics of the ocean echoes from the polarization radar observation data are calculated based on the polarization radar observation data and statistical methods, and a membership function is constructed based on the distribution characteristics of the ocean echoes, specifically as follows:

[0027] Based on statistical methods, the one-dimensional probability density distributions of the differential phase standard deviation, differential reflectivity, radial velocity, and signal-to-noise ratio standard deviation of the polarization radar observation data are calculated.

[0028] The formula for calculating the standard deviation of the differential phase is as follows:

[0029]

[0030] In the formula, σ X The standard deviation of the differential phase is σ, where σ is the standard deviation and X represents the polarization radar observation difference differential phase KDP or signal-to-noise ratio SNR. N represents the mean; N represents the standard deviation and the number of statistical samples.

[0031] Membership functions are constructed based on the various one-dimensional probability density distributions.

[0032] The step of constructing membership functions based on the respective one-dimensional probability density distributions is as follows:

[0033] The membership function is:

[0034]

[0035] In the formula, P(X) is the membership function value, X1 is 5% of the one-dimensional probability density distribution of the differential phase standard deviation, X2 is 25% of the one-dimensional probability density distribution of the differential reflectivity, X3 is 75% of the one-dimensional probability density distribution of the radial velocity, and X4 is 95% of the one-dimensional probability density distribution of the signal-to-noise ratio standard deviation.

[0036] In this preferred embodiment, this application calculates the ocean echo distribution characteristics of polarization radar observation data using statistical methods, and constructs membership functions based on these characteristics, significantly improving the accuracy and reliability of ocean echo identification. Specifically, firstly, statistical methods are used to calculate the one-dimensional probability density distributions of the differential phase standard deviation, differential reflectivity, radial velocity, and signal-to-noise ratio standard deviation of the polarization radar observation data. These distribution characteristics provide a scientific basis for the quantitative analysis of ocean echoes, enabling the fuzziness of ocean echo characteristics to be quantified. By calculating the differential phase standard deviation, the characteristic description of ocean echoes is further refined, improving the distinguishability of features. Next, membership functions are constructed based on these one-dimensional probability density distributions. The construction of membership functions enables fuzzy logic algorithms to more effectively handle the uncertainty and fuzziness of data, improving the accuracy and robustness of identification. Therefore, through this series of scientifically sound steps, this invention not only improves the accuracy of ocean echo identification but also provides a solid foundation for subsequent echo type classification, significantly enhancing the application effect and reliability of weather radar in marine meteorological monitoring.

[0037] As a preferred embodiment of the first aspect, the step of processing the polarization radar observation data according to the fuzzy logic algorithm and the membership function to obtain the membership degree of each pixel of the polarization radar observation data belonging to a preset echo category specifically involves:

[0038] Based on the fuzzy logic algorithm, the preset equal weight principle, and the membership function, the polarization radar observation data is processed to obtain the membership degree of each pixel of the polarization radar observation data belonging to the preset echo category.

[0039] The preset echo categories include meteorological echoes, ground clutter, biological echoes, and ocean clutter.

[0040] In this preferred embodiment, this application processes polarization radar observation data by applying fuzzy logic algorithms and membership functions, combined with a preset equal-weight principle, to calculate the membership degree of each pixel belonging to a preset echo category. This process first utilizes fuzzy logic algorithms to handle the uncertainty and ambiguity of the data, enabling the algorithm to handle complex meteorological data more flexibly. The preset equal-weight principle ensures that each feature parameter contributes equally to the final result when calculating the membership degree, avoiding excessive influence from any single feature parameter and improving the fairness and reliability of the results. Through membership functions, the features of the polarization radar observation data are transformed into specific membership degree values, making the classification of each pixel into meteorological echoes, ground clutter, biological echoes, and oceanic clutter more explicit and scientific. Ultimately, this method not only improves the accuracy of echo type identification but also enhances the robustness of the algorithm, enabling weather radar to accurately identify and classify echo types under different meteorological conditions, providing higher-quality data support for meteorological monitoring and early warning.

[0041] Secondly, this application provides a weather radar echo type identification device. The weather radar echo type identification device includes an acquisition module, a calculation module, a construction module, and an identification module;

[0042] The acquisition module is used to acquire preset polarization radar observation data;

[0043] The calculation module is used to calculate the distribution characteristics of ocean echoes from the polarization radar observation data based on the polarization radar observation data and statistical methods.

[0044] The construction module is used to construct membership functions based on the distribution characteristics of the ocean echoes;

[0045] The identification module is used to process the polarization radar observation data according to the fuzzy logic algorithm and the membership function to obtain the membership degree of each pixel of the polarization radar observation data belonging to the preset echo category.

[0046] Based on the membership degree of each pixel to a preset echo category, the various echo categories of the polarization radar observation data are determined.

[0047] This device utilizes four modules that work in a coordinated manner to more accurately identify weather radar echo signals. By acquiring pre-defined polarization radar observation data, this application ensures the representativeness and timeliness of the data, providing a solid foundation for subsequent analysis. Then, statistical methods are used to calculate the ocean clutter distribution characteristics of the polarization radar observation data, providing a scientific basis for constructing membership functions. The construction of membership functions quantifies the characteristic ambiguity of ocean clutter, laying the foundation for the application of fuzzy logic algorithms. By combining fuzzy logic algorithms with membership functions, the polarization radar observation data is processed to obtain the membership degree of each pixel belonging to a pre-defined echo category. This process effectively handles the uncertainty and ambiguity of the data, improving the accuracy and robustness of identification. Finally, each echo category is determined based on the membership degree, ensuring the scientific nature and accuracy of the classification. This application solves the problem of inaccurate identification of weather radar echo types in existing technologies.

[0048] As a preferred embodiment of the second aspect, after determining the echo categories of the polarization radar observation data based on the membership degree of each pixel to a preset echo category, the method further includes:

[0049] Obtain the underlying surface type information;

[0050] Based on the underlying surface type information, echoes that do not conform to the underlying surface type in each echo category of the polarization radar observation data are removed.

[0051] In this preferred embodiment, after determining the various echo categories of the polarization radar observation data, this application further acquires underlying surface type information and eliminates echoes that do not conform to the underlying surface type based on this information. This process significantly improves the accuracy and reliability of echo identification. Specifically, by acquiring underlying surface type information, we can distinguish different surface features, such as ocean, land, and cities. Combining this information, we can more accurately identify and eliminate those unreasonable echo types on specific underlying surfaces, such as eliminating ocean clutter appearing on land. This step not only reduces misjudgments but also further purifies the radar data, improving the purity and quality of the data.

[0052] Thirdly, this application provides a computer-readable storage medium comprising a stored computer program, wherein, when the computer program is executed, it controls the device containing the computer-readable storage medium to perform the weather radar echo type identification method as described above. Its beneficial effects are the same as those of the weather radar echo type identification method provided in the first aspect of this application.

[0053] Fourthly, this application provides a terminal device including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor executes the computer program to implement any of the weather radar echo type identification methods described in the first aspect. Attached Figure Description

[0054] Figure 1 : A flowchart illustrating an embodiment of the weather radar echo type identification method provided in this application;

[0055] Figure 2 : A schematic diagram of the structure of one embodiment of elevation angle observation provided in this application;

[0056] Figure 3 : A schematic diagram of the structure of one embodiment of the membership function of ocean clutter provided in this application;

[0057] Figure 4 : A schematic diagram of the structure of one embodiment of the polarization radar ocean clutter identification results provided in this application;

[0058] Figure 5 : A schematic diagram of an embodiment of the weather radar echo type identification device provided in this application. Detailed Implementation

[0059] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0060] Example 1

[0061] Please refer to Figure 1 This invention provides a method for identifying weather radar echo types.

[0062] In this embodiment, the process of the weather radar echo type identification method in this application is described in detail through steps S01-S05.

[0063] S01: Acquire preset polarization radar observation data.

[0064] As a preferred embodiment of Embodiment 1, the acquisition of preset polarization radar observation data specifically includes:

[0065] Acquire initial polarization radar observation data for a preset time period, wherein the preset time period is 30 days;

[0066] Based on the time, location, intensity, and Doppler characteristics of the ocean echoes, the preset polarization radar observation data is obtained by filtering from the initial polarization radar observation data.

[0067] More specifically, the steps for collecting and organizing ocean clutter examples are as follows: Ocean clutter examples are selected based on their temporal, spatial distribution, intensity, and other characteristics. Figure 2 This paper presents a case study of ocean clutter observed by the Zhanjiang polarization radar at 01:30 on December 5, 2019. The ocean clutter was mainly distributed to the east of the radar, with an average reflectivity intensity between 0 and 30 dBZ. Figure 2 The data was obtained from the Zhanjiang polarization radar at 01:30 (UTC) on December 5, 2019, at an elevation angle of 0.5°. (a) is reflectivity; (b) is differential reflectivity; (c) is radial velocity; and (d) is correlation coefficient.

[0068] The radar station observation data described in this embodiment comes from the Guangdong Provincial Operational Observation System, which has 11 S-band polarization radars throughout the province.

[0069] In this preferred embodiment, this application obtains high-quality pre-defined polarization radar observation data by acquiring initial polarization radar observation data at a preset time and filtering it based on the time, location, intensity, and Doppler characteristics of ocean echoes. This process first ensures the timeliness and representativeness of the data. By limiting the time frame, the collected data reflects the meteorological characteristics within a specific time period, providing an accurate temporal background for subsequent analysis. Next, through a filtering step, observation data related to ocean echoes is accurately extracted from a large amount of initial data using the characteristic parameters of ocean echoes. This filtering mechanism effectively removes irrelevant data, reduces data noise, and improves the purity and relevance of the data. The filtered data not only reduces the computational load of subsequent processing but also improves the efficiency and accuracy of data processing. Therefore, this processing step of the present invention significantly improves the quality of polarization radar observation data, providing a solid data foundation for subsequent echo type identification and classification, thereby improving the application effect and reliability of weather radar in meteorological monitoring and early warning.

[0070] S02: Based on the polarization radar observation data and statistical methods, the distribution characteristics of the ocean echoes from the polarization radar observation data are calculated.

[0071] As a preferred embodiment of Example 1, the calculation of the distribution characteristics of the ocean echoes from the polarization radar observation data based on the polarization radar observation data and statistical methods specifically includes:

[0072] Based on statistical methods, the one-dimensional probability density distributions of the differential phase standard deviation, differential reflectivity, radial velocity, and signal-to-noise ratio standard deviation of the polarization radar observation data are calculated.

[0073] The formula for calculating the standard deviation of the differential phase is as follows:

[0074]

[0075] In the formula, σ X The standard deviation of the differential phase is σ, where σ is the standard deviation and X represents the polarization radar observation difference differential phase KDP or signal-to-noise ratio SNR. The value represents the mean; N represents the standard deviation and the number of statistical samples; in this case, it is taken as 8 consecutive ranges from the radar radial observation.

[0076] In this preferred embodiment, this application calculates the ocean echo distribution characteristics of polarization radar observation data using statistical methods, and constructs membership functions based on these characteristics, significantly improving the accuracy and reliability of ocean echo identification. Specifically, firstly, statistical methods are used to calculate the one-dimensional probability density distributions of the differential phase standard deviation, differential reflectivity, radial velocity, and signal-to-noise ratio standard deviation of the polarization radar observation data. These distribution characteristics provide a scientific basis for the quantitative analysis of ocean echoes, enabling the fuzziness of ocean echo characteristics to be quantified. By calculating the differential phase standard deviation, the characteristic description of ocean echoes is further refined, improving the distinguishability of features. Next, membership functions are constructed based on these one-dimensional probability density distributions. The construction of membership functions enables fuzzy logic algorithms to more effectively handle the uncertainty and fuzziness of data, improving the accuracy and robustness of identification. Therefore, through this series of scientifically sound steps, this invention not only improves the accuracy of ocean echo identification but also provides a solid foundation for subsequent echo type classification, significantly enhancing the application effect and reliability of weather radar in marine meteorological monitoring.

[0077] S03: Construct a membership function based on the distribution characteristics of the ocean echoes.

[0078] As a preferred embodiment of Example 1, the step of constructing a membership function based on the distribution characteristics of the ocean echo specifically involves:

[0079] Membership functions are the core of fuzzy logic methods, used to quantify the fuzziness of oceanic clutter characteristics. In this case, a trapezoidal function is used to represent membership function values. The membership function is as follows:

[0080]

[0081] In the formula, P(X) is the membership function value, X1 is 5% of the one-dimensional probability density distribution of the differential phase standard deviation, X2 is 25% of the one-dimensional probability density distribution of the differential reflectivity, X3 is 75% of the one-dimensional probability density distribution of the radial velocity, and X4 is 95% of the one-dimensional probability density distribution of the signal-to-noise ratio standard deviation.

[0082] The steps for statistically analyzing membership functions are as follows: The probability density distribution of several ocean clutter cases is statistically analyzed using polarization radar observations; and the threshold values ​​for the membership functions are given using the 5%, 25%, 75%, and 95% quantiles. The results are as follows: Figure 3 As shown, Figure 3 In the table, (a) represents the standard deviation of the differential phase; (b) represents the differential reflectivity; (c) represents the radial velocity; and (d) represents the standard deviation of the signal-to-noise ratio.

[0083] S04: Based on the fuzzy logic algorithm and the membership function, the polarization radar observation data is processed to obtain the membership degree of each pixel of the polarization radar observation data belonging to the preset echo category.

[0084] As a preferred embodiment of Example 1, the step of processing the polarization radar observation data according to the fuzzy logic algorithm and the membership function to obtain the membership degree of each pixel of the polarization radar observation data belonging to a preset echo category is as follows:

[0085] The system uses an equal-weighted principle to infer the membership degree of each pixel to a specified echo category. Finally, the echo category (meteorological echo, ground clutter, biological echo, and ocean clutter) is determined based on the inference results and output.

[0086] In this preferred embodiment, this application processes polarization radar observation data by applying fuzzy logic algorithms and membership functions, combined with a preset equal-weight principle, to calculate the membership degree of each pixel belonging to a preset echo category. This process first utilizes fuzzy logic algorithms to handle the uncertainty and ambiguity of the data, enabling the algorithm to handle complex meteorological data more flexibly. The preset equal-weight principle ensures that each feature parameter contributes equally to the final result when calculating the membership degree, avoiding excessive influence from any single feature parameter and improving the fairness and reliability of the results. Through membership functions, the features of the polarization radar observation data are transformed into specific membership degree values, making the classification of each pixel into meteorological echoes, ground clutter, biological echoes, and oceanic clutter more explicit and scientific. Ultimately, this method not only improves the accuracy of echo type identification but also enhances the robustness of the algorithm, enabling weather radar to accurately identify and classify echo types under different meteorological conditions, providing higher-quality data support for meteorological monitoring and early warning.

[0087] S05: Determine the echo categories of the polarization radar observation data based on the membership degree of each pixel to a preset echo category.

[0088] As a preferred embodiment of Embodiment 1, the step of determining the echo categories of the polarization radar observation data based on the membership degree of each pixel to a preset echo category specifically involves:

[0089] The ocean clutter identification algorithm based on fuzzy logic effectively distinguishes between ocean clutter, meteorological echoes, ground object echoes, and biological echoes, as shown in the results. Figure 4 As shown, at an elevation angle of 0.5°, a total of 14,937 pixels were identified as ocean clutter. Figure 4 The results are as follows: Marine clutter identification results from Zhanjiang polarization radar at 01:30 (UTC) on December 5, 2019: SC: Marine clutter; GC: Ground clutter; BS: Biological echo; ME: Meteorological echo.

[0090] As a preferred embodiment of Embodiment 1, after determining the echo categories of the polarization radar observation data based on the membership degree of each pixel to a preset echo category, the method further includes:

[0091] Obtain the underlying surface type information;

[0092] Based on the underlying surface type information, echoes that do not conform to the underlying surface type in each echo category of the polarization radar observation data are removed.

[0093] In this preferred embodiment, after determining the various echo categories of the polarization radar observation data, this application further acquires underlying surface type information and eliminates echoes that do not conform to the underlying surface type based on this information. This process significantly improves the accuracy and reliability of echo identification. Specifically, by acquiring underlying surface type information, we can distinguish different surface features, such as ocean, land, and cities. Combining this information, we can more accurately identify and eliminate those unreasonable echo types on specific underlying surfaces, such as eliminating ocean clutter appearing on land. This step not only reduces misjudgments but also further purifies the radar data, improving the purity and quality of the data.

[0094] This application ensures the representativeness and timeliness of pre-defined polarization radar observation data, providing a solid foundation for subsequent analysis. Then, statistical methods are used to calculate the ocean clutter distribution characteristics of the polarization radar observation data, providing a scientific basis for constructing membership functions. The construction of membership functions quantifies the characteristic ambiguity of ocean clutter, laying the foundation for the application of fuzzy logic algorithms. By combining fuzzy logic algorithms with membership functions, the polarization radar observation data is processed to obtain the membership degree of each pixel belonging to a pre-defined echo category. This process effectively handles the uncertainty and ambiguity of the data, improving the accuracy and robustness of identification. Finally, each echo category is determined based on the membership degree, ensuring the scientific nature and accuracy of the classification. This application solves the problem in existing technologies that cannot accurately identify the echo type of weather radar.

[0095] Example 2

[0096] Please refer to Figure 5This application provides a weather radar echo type identification device.

[0097] In this embodiment, the weather radar echo type identification device includes an acquisition module 10, a calculation module 20, a construction module 30, and an identification module 40.

[0098] The acquisition module 10 is used to acquire preset polarization radar observation data.

[0099] As a preferred embodiment of Embodiment 2, the acquisition of preset polarization radar observation data specifically includes:

[0100] Acquire initial polarization radar observation data for a preset time period, wherein the preset time period is 30 days;

[0101] Based on the time, location, intensity, and Doppler characteristics of the ocean echoes, the preset polarization radar observation data is obtained by filtering from the initial polarization radar observation data.

[0102] More specifically, the steps for collecting and organizing ocean clutter examples are as follows: Ocean clutter examples are selected based on their temporal, spatial distribution, intensity, and other characteristics. Figure 2 This paper presents a case study of ocean clutter observed by the Zhanjiang polarization radar at 01:30 on December 5, 2019. The ocean clutter was mainly distributed to the east of the radar, with an average reflectivity intensity between 0 and 30 dBZ. Figure 2 The data was obtained from the Zhanjiang polarization radar at 01:30 (UTC) on December 5, 2019, at an elevation angle of 0.5°. (a) is reflectivity; (b) is differential reflectivity; (c) is radial velocity; and (d) is correlation coefficient.

[0103] The radar station observation data described in this embodiment comes from the Guangdong Provincial Operational Observation System, which has 11 S-band polarization radars throughout the province.

[0104] In this preferred embodiment, this application obtains high-quality pre-defined polarization radar observation data by acquiring initial polarization radar observation data at a preset time and filtering it based on the time, location, intensity, and Doppler characteristics of ocean echoes. This process first ensures the timeliness and representativeness of the data. By limiting the time frame, the collected data reflects the meteorological characteristics within a specific time period, providing an accurate temporal background for subsequent analysis. Next, through a filtering step, observation data related to ocean echoes is accurately extracted from a large amount of initial data using the characteristic parameters of ocean echoes. This filtering mechanism effectively removes irrelevant data, reduces data noise, and improves the purity and relevance of the data. The filtered data not only reduces the computational load of subsequent processing but also improves the efficiency and accuracy of data processing. Therefore, this processing step of the present invention significantly improves the quality of polarization radar observation data, providing a solid data foundation for subsequent echo type identification and classification, thereby improving the application effect and reliability of weather radar in meteorological monitoring and early warning.

[0105] The calculation module 20 is used to calculate the distribution characteristics of ocean echoes from the polarization radar observation data based on the polarization radar observation data and statistical methods.

[0106] As a preferred embodiment of Example 2, the calculation of the distribution characteristics of the ocean echoes from the polarization radar observation data based on the polarization radar observation data and statistical methods specifically includes:

[0107] Based on statistical methods, the one-dimensional probability density distributions of the differential phase standard deviation, differential reflectivity, radial velocity, and signal-to-noise ratio standard deviation of the polarization radar observation data are calculated.

[0108] The formula for calculating the standard deviation of the differential phase is as follows:

[0109]

[0110] In the formula, σ X The standard deviation of the differential phase is σ, where σ is the standard deviation and X represents the polarization radar observation difference differential phase KDP or signal-to-noise ratio SNR. The value represents the mean; N represents the standard deviation and the number of statistical samples; in this case, it is taken as 8 consecutive ranges from the radar radial observation.

[0111] In this preferred embodiment, this application calculates the ocean echo distribution characteristics of polarization radar observation data using statistical methods, and constructs membership functions based on these characteristics, significantly improving the accuracy and reliability of ocean echo identification. Specifically, firstly, statistical methods are used to calculate the one-dimensional probability density distributions of the differential phase standard deviation, differential reflectivity, radial velocity, and signal-to-noise ratio standard deviation of the polarization radar observation data. These distribution characteristics provide a scientific basis for the quantitative analysis of ocean echoes, enabling the fuzziness of ocean echo characteristics to be quantified. By calculating the differential phase standard deviation, the characteristic description of ocean echoes is further refined, improving the distinguishability of features. Next, membership functions are constructed based on these one-dimensional probability density distributions. The construction of membership functions enables fuzzy logic algorithms to more effectively handle the uncertainty and fuzziness of data, improving the accuracy and robustness of identification. Therefore, through this series of scientifically sound steps, this invention not only improves the accuracy of ocean echo identification but also provides a solid foundation for subsequent echo type classification, significantly enhancing the application effect and reliability of weather radar in marine meteorological monitoring.

[0112] The construction module 30 is used to construct a membership function based on the distribution characteristics of the ocean echo.

[0113] As a preferred embodiment of Example 2, the step of constructing a membership function based on the distribution characteristics of the ocean echo is specifically as follows:

[0114] Membership functions are the core of fuzzy logic methods, used to quantify the fuzziness of oceanic clutter characteristics. In this case, a trapezoidal function is used to represent membership function values. The membership function is as follows:

[0115]

[0116] In the formula, P(X) is the membership function value, X1 is 5% of the one-dimensional probability density distribution of the differential phase standard deviation, X2 is 25% of the one-dimensional probability density distribution of the differential reflectivity, X3 is 75% of the one-dimensional probability density distribution of the radial velocity, and X4 is 95% of the one-dimensional probability density distribution of the signal-to-noise ratio standard deviation.

[0117] The steps for statistically analyzing membership functions are as follows: The probability density distribution of several ocean clutter cases is statistically analyzed using polarization radar observations; and the threshold values ​​for the membership functions are given using the 5%, 25%, 75%, and 95% quantiles. The results are as follows: Figure 3 As shown, Figure 3 In the table, (a) represents the standard deviation of the differential phase; (b) represents the differential reflectivity; (c) represents the radial velocity; and (d) represents the standard deviation of the signal-to-noise ratio.

[0118] The identification module 40 is used to process the polarization radar observation data according to the fuzzy logic algorithm and the membership function to obtain the membership degree of each pixel of the polarization radar observation data belonging to the preset echo category.

[0119] As a preferred embodiment of Embodiment 2, the step of processing the polarization radar observation data according to the fuzzy logic algorithm and the membership function to obtain the membership degree of each pixel of the polarization radar observation data belonging to a preset echo category is as follows:

[0120] The system uses an equal-weighted principle to infer the membership degree of each pixel to a specified echo category. Finally, the echo category (meteorological echo, ground clutter, biological echo, and ocean clutter) is determined based on the inference results and output.

[0121] In this preferred embodiment, this application processes polarization radar observation data by applying fuzzy logic algorithms and membership functions, combined with a preset equal-weight principle, to calculate the membership degree of each pixel belonging to a preset echo category. This process first utilizes fuzzy logic algorithms to handle the uncertainty and ambiguity of the data, enabling the algorithm to handle complex meteorological data more flexibly. The preset equal-weight principle ensures that each feature parameter contributes equally to the final result when calculating the membership degree, avoiding excessive influence from any single feature parameter and improving the fairness and reliability of the results. Through membership functions, the features of the polarization radar observation data are transformed into specific membership degree values, making the classification of each pixel into meteorological echoes, ground clutter, biological echoes, and oceanic clutter more explicit and scientific. Ultimately, this method not only improves the accuracy of echo type identification but also enhances the robustness of the algorithm, enabling weather radar to accurately identify and classify echo types under different meteorological conditions, providing higher-quality data support for meteorological monitoring and early warning.

[0122] The identification module 40 is also used to determine each echo category of the polarization radar observation data based on the membership degree of each pixel to a preset echo category.

[0123] As a preferred embodiment of Embodiment 2, the step of determining the echo categories of the polarization radar observation data based on the membership degree of each pixel to a preset echo category specifically involves:

[0124] The ocean clutter identification algorithm based on fuzzy logic effectively distinguishes between ocean clutter, meteorological echoes, ground object echoes, and biological echoes, as shown in the results. Figure 4 As shown, at an elevation angle of 0.5°, a total of 14,937 pixels were identified as ocean clutter. Figure 4 The results are as follows: Marine clutter identification results from Zhanjiang polarization radar at 01:30 (UTC) on December 5, 2019: SC: Marine clutter; GC: Ground clutter; BS: Biological echo; ME: Meteorological echo.

[0125] As a preferred embodiment of Embodiment 2, after determining the echo categories of the polarization radar observation data based on the membership degree of each pixel to a preset echo category, the method further includes:

[0126] Obtain the underlying surface type information;

[0127] Based on the underlying surface type information, echoes that do not conform to the underlying surface type in each echo category of the polarization radar observation data are removed.

[0128] In this preferred embodiment, after determining the various echo categories of the polarization radar observation data, this application further acquires underlying surface type information and eliminates echoes that do not conform to the underlying surface type based on this information. This process significantly improves the accuracy and reliability of echo identification. Specifically, by acquiring underlying surface type information, we can distinguish different surface features, such as ocean, land, and cities. Combining this information, we can more accurately identify and eliminate those unreasonable echo types on specific underlying surfaces, such as eliminating ocean clutter appearing on land. This step not only reduces misjudgments but also further purifies the radar data, improving the purity and quality of the data.

[0129] This device utilizes four modules that work in a coordinated manner to more accurately identify weather radar echo signals. By acquiring pre-defined polarization radar observation data, this application ensures the representativeness and timeliness of the data, providing a solid foundation for subsequent analysis. Then, statistical methods are used to calculate the ocean clutter distribution characteristics of the polarization radar observation data, providing a scientific basis for constructing membership functions. The construction of membership functions quantifies the characteristic ambiguity of ocean clutter, laying the foundation for the application of fuzzy logic algorithms. By combining fuzzy logic algorithms with membership functions, the polarization radar observation data is processed to obtain the membership degree of each pixel belonging to a pre-defined echo category. This process effectively handles the uncertainty and ambiguity of the data, improving the accuracy and robustness of identification. Finally, each echo category is determined based on the membership degree, ensuring the scientific nature and accuracy of the classification. This application solves the problem of inaccurate identification of weather radar echo types in existing technologies.

[0130] Example 3:

[0131] This application provides a computer-readable storage medium including a stored computer program, wherein the computer program, when running, controls the device where the computer-readable storage medium is located to execute the aforementioned weather radar echo type identification method.

[0132] The weather radar echo type identification method, if implemented as a software functional unit and used as an independent product, can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the above embodiments of the present invention can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc.

[0133] Example 4

[0134] This application provides a terminal device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements any of the weather radar echo type identification methods described in Embodiment 1.

[0135] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. In particular, it should be noted that any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention for those skilled in the art.

Claims

1. A method of identifying weather radar echo types, characterized by, The method comprises the following steps: acquiring preset polarimetric radar observation data; calculating distribution characteristics of ocean echoes of the polarimetric radar observation data according to the polarimetric radar observation data and a statistical method; constructing a membership function according to the distribution characteristics of the ocean echoes; the calculating of the distribution characteristics of the ocean echoes of the polarimetric radar observation data according to the polarimetric radar observation data and the statistical method, and the constructing of the membership function according to the distribution characteristics of the ocean echoes, specifically comprise: calculating respective one-dimensional probability density distributions of specific differential phase standard deviation, differential reflectivity, radial velocity and signal-to-noise ratio standard deviation of the polarimetric radar observation data according to the statistical method; wherein, the calculation formula of the specific differential phase standard deviation is: where σ X is the standard deviation of the differential phase, σ is the standard deviation, X represents the polarization radar observation, the differential phase KDP, or the signal-to-noise ratio SNR, represents the average value; N represents the number of standard deviation statistical samples; constructing the membership function according to the respective one-dimensional probability density distributions; the constructing of the membership function according to the respective one-dimensional probability density distributions, specifically comprises: the membership function is: wherein, P(X) is a membership function value, X1 is 5% of the one-dimensional probability density distribution of the specific differential phase standard deviation, X2 is 25% of the one-dimensional probability density distribution of the differential reflectivity, X3 is 75% of the one-dimensional probability density distribution of the radial velocity, and X4 is 95% of the one-dimensional probability density distribution of the signal-to-noise ratio standard deviation; processing the polarimetric radar observation data according to a fuzzy logic algorithm and the membership function to obtain a membership degree of each pixel of the polarimetric radar observation data belonging to a preset echo category; determining respective echo categories of the polarimetric radar observation data according to the membership degree of each pixel belonging to the preset echo category.

2. The weather radar echo type identification method of claim 1, wherein, after the determining of the respective echo categories of the polarimetric radar observation data according to the membership degree of each pixel belonging to the preset echo category, the method further comprises: acquiring underlying surface type information; eliminating each echo of the respective echo categories of the polarimetric radar observation data that does not conform to the underlying surface type according to the underlying surface type information.

3. The weather radar echo type identification method of claim 1, wherein, the acquiring of the preset polarimetric radar observation data, specifically comprises: acquiring initial polarimetric radar observation data of a preset time; screening the preset polarimetric radar observation data from the initial polarimetric radar observation data according to time, position, intensity and Doppler characteristics of ocean echoes.

4. The weather radar echo type identification method of claim 1, wherein, the processing of the polarimetric radar observation data according to the fuzzy logic algorithm and the membership function to obtain the membership degree of each pixel of the polarimetric radar observation data belonging to the preset echo category, specifically comprises: processing the polarimetric radar observation data according to the fuzzy logic algorithm, a preset equal weight principle and the membership function to obtain the membership degree of each pixel of the polarimetric radar observation data belonging to the preset echo category; wherein, the preset echo category comprises meteorological echoes, ground object clutter, biological echoes and ocean clutter.

5. A device for identifying weather radar echo types, characterized in that The method comprises the following steps: an acquiring module, a calculating module, a constructing module and an identifying module; the acquiring module is used for acquiring preset polarimetric radar observation data; the calculating module is used for calculating distribution characteristics of ocean echoes of the polarimetric radar observation data according to the polarimetric radar observation data and a statistical method; The constructing module is configured to construct a membership function according to the distribution characteristics of the marine echo. The calculating module is configured to calculate the distribution characteristics of the marine echo of the polarimetric radar observation data according to the polarimetric radar observation data and a statistical method, and the constructing module is configured to construct a membership function according to the distribution characteristics of the marine echo, specifically as follows: The calculating module is configured to calculate each one-dimensional probability density distribution of the specific differential phase standard deviation, the differential reflectivity, the radial velocity, and the signal-to-noise ratio standard deviation of the polarimetric radar observation data according to a statistical method. where σ X is the standard deviation of the differential phase, σ is the standard deviation, X represents the polarization radar observation, the differential phase KDP, or the signal-to-noise ratio SNR, represents the average value; N represents the number of standard deviation statistical samples; The calculating module is configured to calculate each one-dimensional probability density distribution of the specific differential phase standard deviation, the differential reflectivity, the radial velocity, and the signal-to-noise ratio standard deviation of the polarimetric radar observation data according to a statistical method. The calculating module is configured to calculate each one-dimensional probability density distribution of the specific differential phase standard deviation, the differential reflectivity, the radial velocity, and the signal-to-noise ratio standard deviation of the polarimetric radar observation data according to a statistical method. The constructing module is configured to construct a membership function according to the distribution characteristics of the marine echo. The constructing module is configured to construct a membership function according to the distribution characteristics of the marine echo. The membership function is as follows: In the formula, P(X) is a membership function value, X1 is 5% of the one-dimensional probability density distribution of the specific differential phase standard deviation, X2 is 25% of the one-dimensional probability density distribution of the differential reflectivity, X3 is 75% of the one-dimensional probability density distribution of the radial velocity, and X4 is 95% of the one-dimensional probability density distribution of the signal-to-noise ratio standard deviation.

6. The weather radar echo type identification apparatus according to claim 5, characterized in that, The identifying module is configured to process the polarimetric radar observation data according to a fuzzy logic algorithm and the membership function to obtain a membership degree of each pixel of the polarimetric radar observation data belonging to a preset echo category. Each echo category of the polarimetric radar observation data is determined according to the membership degree of each pixel belonging to the preset echo category. After each echo category of the polarimetric radar observation data is determined according to the membership degree of each pixel belonging to the preset echo category, the following steps are further included.

7. A computer readable storage medium characterized in that, Obtain underlying surface type information.

8. A terminal device, comprising: According to the underlying surface type information, eliminate each echo of each echo category of the polarimetric radar observation data that does not conform to the underlying surface type. The computer readable storage medium includes a stored computer program, wherein the computer readable storage medium controls the device where the computer readable storage medium is located to execute the weather radar echo type identification method according to any one of claims 1 to 4 when the computer program is running. The computer readable storage medium includes a stored computer program, wherein the computer readable storage medium controls the device where the computer readable storage medium is located to execute the weather radar echo type identification method according to any one of claims 1 to 4 when the computer program is running.

Images