Abnormal meteorological detection method based on phased array radar

Abstract

The invention relates to an abnormal weather detection method based on a phased array radar, and belongs to the technical field of data processing and analysis. According to the method, based on the working principle of the phased array radar, any kind of scanning parameter data in the current scanning period of the phased array radar is used as data to be analyzed, and the suspected anomaly degree of the meteorological type of the scanning parameter is determined according to the situation that the whole data to be analyzed exceeds the normal range; then, according to the difference between the data characteristics of the to-be-analyzed data under the real abnormal weather condition and the data characteristics under the noise influence, the credibility about the abnormal weather is determined, and finally the real abnormal degree is determined in combination with the suspected abnormal degree and the credibility. According to the method, the real anomaly degree under the period is determined by taking the whole scanning period as consideration, abnormal weather judgment errors caused by noise interference due to comparison between a data value and a threshold value at a single data point are eliminated, and the accuracy of abnormal weather detection based on the phased array radar is remarkably improved.

Description

Technical Field

[0001] This invention relates to the field of data processing and analysis technology, and in particular to an abnormal weather detection method based on phased array radar. Background Technology

[0002] In the field of meteorological observation, meteorological radar is widely used. Meteorological radar includes traditional rotating meteorological radar and phased array radar. Phased array radar achieves electronic scanning by controlling the phase of the array antenna elements, avoiding the structural limitations of the mechanical rotation of traditional radar, and greatly improving detection efficiency and response speed. Therefore, based on its advantages such as high precision, high resolution, and fast scanning, phased array radar has gradually become one of the core components of meteorological monitoring systems.

[0003] Abnormal weather refers to unusual phenomena observed in meteorological observations, typically manifested as deviations of key meteorological parameters such as temperature, air pressure, humidity, and wind speed from their normal ranges. While traditional threshold detection methods can quickly and easily identify abnormal weather data within an observation period and issue alerts, they are sensitive to noise and prone to misinterpreting noisy data as abnormal weather data. Although phased array radar has improved detection efficiency and response speed based on structural principles, it still cannot avoid the misinterpretation of abnormal weather by traditional threshold detection methods, resulting in reduced accuracy and a higher probability of false alarms.

[0004] In other words, current abnormal weather detection methods based on phased array radar suffer from insufficient detection accuracy. Summary of the Invention

[0005] In view of this, the present invention provides an abnormal weather detection method based on phased array radar to solve the technical problem of insufficient detection accuracy of current phased array radar for abnormal weather detection.

[0006] The present invention provides an abnormal weather detection method based on phased array radar, comprising:

[0007] During meteorological monitoring, the data sequence of any scanning parameter of the phased array radar in the current scanning cycle is acquired as the data to be analyzed.

[0008] The degree of suspected anomaly in the data to be analyzed is determined based on the number of data points in the data to be analyzed that exceed the normal range of any one of the scanning parameters and the extent of the exceedance.

[0009] The first confidence level of the data to be analyzed is determined based on the overall volatility and data completeness of the data to be analyzed; the second confidence level of the data to be analyzed is determined based on the local stability and local persistence of the local changes of the data to be analyzed; and the comprehensive confidence level is determined based on the first confidence level and the second confidence level.

[0010] The true anomaly of the data to be analyzed is determined based on the suspected anomaly and the overall confidence level, and anomaly detection for the meteorological category corresponding to any one of the scanning parameters is completed based on the true anomaly.

[0011] Furthermore, determining the suspected anomaly degree of the data to be analyzed includes:

[0012] The normalized value of the number of data points in the data to be analyzed that exceed the normal range of any one of the scanning parameters is calculated and denoted as the first suspected anomaly characterization term;

[0013] The normalized value of the sum of the absolute values ​​of the portions of all data points in the data to be analyzed that exceed the normal range of any one of the scanning parameters is recorded as the second suspected anomaly characterization term.

[0014] The suspected anomaly score of the data to be analyzed is constructed based on the first suspected anomaly score and the second suspected anomaly score, wherein the suspected anomaly score is proportional to both the first and second suspected anomaly scores.

[0015] Furthermore, a first confidence level of the data to be analyzed is determined based on the signal-to-noise ratio of the data to be analyzed.

[0016] Further, determining the first confidence level of the data to be analyzed includes:

[0017] The data points in the data to be analyzed whose trend changes are recorded as change-sign data points. The absolute value of the data difference between any change-sign data point and the previous data point is recorded as the data change amount of any change-sign data point. The normalized value of the product of the mean of the data change amounts of all the change-sign data points and the number of change-sign data points is calculated and recorded as the overall fluctuation characterization term of the data to be analyzed.

[0018] Multiply the ratio of the number of missing data points in the data to be analyzed to the number of data points that should normally be collected in the data to be analyzed by the maximum number of consecutively missing data points in the data to be analyzed, and record the normalized value of the resulting product as the missing data characteristic term of the data to be analyzed.

[0019] The normalized value of the reciprocal of the signal-to-noise ratio of the data to be analyzed is denoted as the disturbance level characterization term of the data to be analyzed.

[0020] The first confidence level of the data to be analyzed is determined based on the overall fluctuation characterization, the missing characterization, and the disturbance degree characterization. The first confidence level is inversely proportional to the overall fluctuation characterization, the missing characterization, and the disturbance degree characterization.

[0021] Furthermore, a second confidence level of the data to be analyzed is determined based on the degree of difference between the data to be analyzed and the data sequence of any scanning parameter acquired by other phased array radars in the same area that acquired the data to be analyzed under the same scanning period.

[0022] Further, determining the second confidence level of the data to be analyzed includes:

[0023] Calculate the standard deviation of the first derivative of each data point in the data to be analyzed, and denote the reciprocal of the standard deviation of the first derivative as the local stationary characterization term of the data to be analyzed.

[0024] Calculate the average duration of each single trend of change within the data to be analyzed, and denote it as the local change persistence characterization term of the data to be analyzed;

[0025] The phased array radar that acquires the data to be analyzed is denoted as the current radar, and other phased array radars in the same area as the current radar are denoted as comparison radars. The data sequence of any scanning parameter acquired by each comparison radar in the current scanning cycle is denoted as the comparison sequence. The average difference between the data to be analyzed and each comparison sequence is calculated, and the reciprocal of the average difference is denoted as the extensive characterization term of the change of the data to be analyzed.

[0026] A second confidence level of the data to be analyzed is determined based on the local change stationary characterization term, the local change persistent characterization term, and the change widespread characterization term. The second confidence level is proportional to the local change stationary characterization term, the local change persistent characterization term, and the change widespread characterization term.

[0027] Furthermore, determining the true anomaly degree of the data to be analyzed includes:

[0028] The product of the first confidence level and the second confidence level is taken as the comprehensive confidence level, and the product of the comprehensive confidence level and the suspected anomaly level is taken as the true anomaly level.

[0029] Furthermore, based on the true anomaly degree, anomaly detection is performed on the meteorological category corresponding to any of the scanning parameters, including:

[0030] Multiple anomaly thresholds are set according to the anomaly level of the abnormal weather corresponding to any of the scanning parameters. The actual anomaly is compared with the anomaly threshold, and an abnormal weather alarm of the corresponding warning level is issued based on the comparison result between the actual anomaly and the anomaly threshold.

[0031] The advantages of this invention compared to the prior art are:

[0032] Based on the working principle of phased array radar, this invention uses any scanning parameter data within the current scanning cycle of the phased array radar as the data to be analyzed. The degree of suspected anomaly for the meteorological type related to that scanning parameter is determined based on the overall deviation of the analyzed data from the normal range. Then, the credibility of the abnormal meteorological situation is determined based on the difference between the data characteristics under actual abnormal meteorological conditions and the data characteristics under noise influence. Finally, the degree of suspected anomaly and the credibility are combined to determine the degree of actual anomaly. This invention considers the entire scanning cycle to determine the degree of actual anomaly within that cycle, eliminating the anomaly judgment error caused by noise interference when comparing data values ​​to thresholds at individual data points. This significantly improves the accuracy of abnormal meteorological detection based on phased array radar. Attached Figure Description

[0033] To more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0034] Figure 1 This is a flowchart illustrating an abnormal weather detection method based on phased array radar provided in Embodiment 1 of the present invention. Detailed Implementation

[0035] The overall concept of this invention is as follows:

[0036] This invention improves upon the traditional threshold detection method by determining whether a suspected anomaly is genuine based on the comprehensive anomaly level of each suspected anomaly data point within any scanning cycle of a phased array radar during meteorological detection. This eliminates noise interference and improves the accuracy of anomaly detection results. Specifically, firstly, the suspected anomaly level of a particular type of scanning parameter within a scanning cycle is obtained based on the number and degree of data points exceeding its normal range. Then, the anomaly confidence level is obtained by combining the signal quality and multi-source anomaly characteristics of the scanning parameter sequence within the scanning cycle. Finally, the comprehensive anomaly level of that type of scanning parameter within the scanning cycle is obtained based on the suspected anomaly level and the anomaly confidence level. The comprehensive anomaly level is then used to determine whether the meteorological type corresponding to that scanning parameter is abnormal in the current scanning cycle. This process is repeated for all meteorological types to detect abnormal weather conditions.

[0037] To further illustrate the technical solution of the present invention, specific embodiments are described below.

[0038] References to "one embodiment" or "some embodiments" as described in this specification mean that one or more embodiments of the invention include a particular feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. Furthermore, a particular feature, structure, or characteristic in one or more embodiments may be combined in any suitable form, and the terms "comprising," "including," "having," and variations thereof mean "including, but not limited to," unless otherwise specifically emphasized.

[0039] It should be understood that the sequence number of each step in the following embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0040] Method Implementation Examples:

[0041] See Figure 1 This is a flowchart illustrating an abnormal weather detection method based on phased array radar provided in Embodiment 1 of the present invention. Figure 1 As shown, the detection method may include the following steps:

[0042] S101, during meteorological monitoring, acquires the data sequence of any scanning parameter of the phased array radar in the current scanning cycle as the data to be analyzed.

[0043] Phased array radar achieves efficient meteorological data acquisition through electronic scanning technology. Its core principle is to control the beam direction by utilizing the phase difference of a large number of radiating elements in the antenna array, allowing for rapid scanning of target areas without mechanical rotation. During data acquisition, phased array radar transmits electromagnetic waves and receives echo signals from target objects (such as precipitation particles, clouds, or weather balloons), simultaneously acquiring multi-dimensional meteorological information. The meteorological data from phased array radar is continuously generated and typically updated at fixed scanning cycles (e.g., 5-10 minutes). Therefore, the detection of abnormal weather conditions also needs to be completed within each scanning cycle.

[0044] This embodiment detects and analyzes the scanning parameters collected by the phased array radar, such as echo power (reflecting precipitation intensity), radial velocity (analyzing wind field structure), and velocity spectrum width (characterizing turbulence intensity), to determine whether there are any abnormalities in the meteorological type corresponding to each parameter.

[0045] For the various scanning parameters acquired on the phased array radar, this embodiment analyzes each scanning parameter one by one according to the scanning cycle. The preferred scanning cycle is the complete scanning cycle closest to the current time, which is recorded as the current scanning cycle. Then, the single scanning parameter is analyzed in turn. For example, the echo power data is analyzed first, and the echo power data sequence under the current scanning cycle is used as the data to be analyzed.

[0046] S102, determine the suspected anomaly degree of the data to be analyzed based on the number of data points in the data to be analyzed that exceed the normal range of any one of the scanning parameters and the degree of exceeding the normal range.

[0047] Because phased array radar is subject to interference from natural phenomena such as ground clutter and biological interference when collecting relevant meteorological data, the data points in the scanning parameters used for meteorological anomaly detection that exceed the normal fluctuation range contain not only data corresponding to actual abnormal weather but also noisy data, leading to misjudgments of whether or not there is a meteorological anomaly. Therefore, to eliminate misjudgments caused by this noise, this embodiment selects to analyze the overall characteristics of the data in a single scanning cycle, that is, to treat a scanning cycle as an entire analysis, in order to determine the meteorological anomalies in the scanning cycle, thereby eliminating the noise error caused by individual data points.

[0048] Therefore, it is first necessary to analyze the degree to which the data in the current scanning cycle exceeds the normal range, so as to reflect the suspected anomaly level of the data in the current scanning cycle through the data values. Specifically, when the weather is normal, the relevant scanning parameters generally change normally within the rated range, while when anomalies occur, they generally deviate from the normal range. Therefore, the normal range of the data to be analyzed can be obtained first, and then, based on the number of data points exceeding the normal range in each data point of the data to be analyzed in the current scanning cycle, and the degree to which each data point exceeds the normal range, the suspected anomaly level of the current scanning cycle can be obtained, including:

[0049] The normalized value of the number of data points in the data to be analyzed that exceed the normal range of any one of the scanning parameters is calculated and denoted as the first suspected anomaly characterization term;

[0050] The normalized value of the sum of the absolute values ​​of the portions of all data points in the data to be analyzed that exceed the normal range of any one of the scanning parameters is recorded as the second suspected anomaly characterization term.

[0051] The suspected anomaly score of the data to be analyzed is constructed based on the first suspected anomaly score and the second suspected anomaly score, wherein the suspected anomaly score is proportional to both the first and second suspected anomaly scores.

[0052] Furthermore, as a preferred embodiment, the suspected anomaly level is:

[0053]

[0054] in, This indicates the degree of suspected anomaly in the data to be analyzed during the i-th scan cycle of the phased array radar meteorological detection process. This refers to normalization, such as linear normalization, norm normalization, etc. This represents the number of data points in the data to be analyzed that exceed its normal range during the i-th scan cycle. This represents the absolute value of the excess portion of the j-th data point that exceeds the normal range in the data to be analyzed during the i-th scan cycle. This represents the sum of the absolute values ​​of the out-of-range values ​​among all data points in the data to be analyzed that are outside the normal range during the i-th scan cycle. The more data points that exceed the normal range in the current scan cycle, and the greater the degree of exceedance, the higher the suspected anomaly level of that scan cycle.

[0055] S103, determine the first confidence level of the data to be analyzed based on the overall fluctuation level and data completeness of the data to be analyzed, determine the second confidence level of the data to be analyzed based on the local change stability and local change persistence of the data to be analyzed, and determine the comprehensive confidence level based on the first confidence level and the second confidence level.

[0056] Since not only the scanning parameter data corresponding to the actual abnormal weather will exceed the normal range, but some noise data will also exceed the normal range, the suspected anomaly degree obtained in the previous step will still have errors due to noise interference. Therefore, it is necessary to combine the characteristics of each part of the data to be analyzed to obtain its signal quality and obtain the corresponding confidence level to correct the suspected anomaly degree used in the previous step to characterize the degree of abnormal weather.

[0057] Specifically, noise firstly affects the overall data quality of the data to be analyzed during the scanning cycle, causing changes in both the overall fluctuation frequency and amplitude of the data. Simultaneously, noise interference can also lead to data loss at data points during the acquisition process. Therefore, the first confidence level of the data to be analyzed can be determined based on the overall fluctuation level and data integrity. Furthermore, considering that the phased array radar system also supports the automatic acquisition of information such as the signal-to-noise ratio (SNR) of its acquired data, and that the SNR directly reflects the signal quality of the data to be analyzed, the first confidence level of the data to be analyzed is also determined based on the SNR. Therefore, determining the first confidence level of the data to be analyzed includes:

[0058] The data points in the data to be analyzed whose trend changes are recorded as change-sign data points. The absolute value of the data difference between any change-sign data point and the previous data point is recorded as the data change amount of any change-sign data point. The normalized value of the product of the mean of the data change amounts of all the change-sign data points and the number of change-sign data points is calculated and recorded as the overall fluctuation characterization term of the data to be analyzed.

[0059] Multiply the ratio of the number of missing data points in the data to be analyzed to the number of data points that should normally be collected in the data to be analyzed by the maximum number of consecutively missing data points in the data to be analyzed, and record the normalized value of the resulting product as the missing data characteristic term of the data to be analyzed.

[0060] The normalized value of the reciprocal of the signal-to-noise ratio of the data to be analyzed is denoted as the disturbance level characterization term of the data to be analyzed.

[0061] The first confidence level of the data to be analyzed is determined based on the overall fluctuation characterization, the missing characterization, and the disturbance degree characterization. The first confidence level is inversely proportional to the overall fluctuation characterization, the missing characterization, and the disturbance degree characterization.

[0062] As a further preferred option, the first confidence level is:

[0063]

[0064] in, This represents the first confidence level of the data to be analyzed in the i-th scan cycle. This refers to normalization, such as linear normalization, norm normalization, etc. This represents the number of data points in the data to be analyzed during the i-th scan period whose trend has changed, i.e., the number of data points whose signs have changed. The more times the trend changes, the higher the fluctuation frequency of the data to be analyzed in the current scan period. This represents the absolute value of the data difference between the j-th data point in the data to be analyzed where the trend of change has changed during the i-th scan period, and the preceding data point in the data to be analyzed. It is denoted as the data change amount of the j-th data point among the data points where the trend of change has changed. This represents the total number of data points in the data to be analyzed whose trend has changed during the i-th scan cycle. This represents the mean of the changes in the data points whose trend changed during the i-th scan period. The larger this mean is, the greater the fluctuation amplitude of the data at the point of change of sign, which is more consistent with the characteristic of large fluctuation intensity of noisy data points. This represents the number of missing data points in the data to be analyzed during the i-th scan cycle. This represents the number of data points that should normally be collected in the data to be analyzed in the i-th scan cycle. This represents the ratio of missing data points to the number of data points that should have been collected in the data to be analyzed in the i-th scan cycle. The larger this value is, the worse the data signal quality of the data to be analyzed in the scan cycle. This represents the maximum number of consecutively lost data points in the data to be analyzed during the i-th scan cycle. It characterizes the maximum degree of consecutive data loss. The more consecutively lost data points there are, the worse the signal continuity of the data, and correspondingly, the worse its signal quality. This represents the signal-to-noise ratio of the data to be analyzed in the i-th scan cycle. The higher the signal-to-noise ratio, the better the signal quality. The lower the signal-to-noise ratio, the lower the signal reliability and the worse the signal quality.

[0065] Since abnormal weather is a change in meteorological data caused by normal environmental changes, real meteorological anomalies (such as convective cells and fronts) often have a clear spatial structure and exhibit a certain degree of continuity, meaning the data changes are relatively uniform. Noise, on the other hand, is often more random in space. Therefore, different noise levels will lead to different degrees of local stability of the data under analysis in the current scan cycle, i.e., different degrees of consistency in the trend. Secondly, real meteorological anomalies are often accompanied by a sustained trend; that is, when an upward or downward trend occurs within a single scan cycle, the duration of the trend is relatively long. Noise data, however, is generally generated instantaneously and exhibits abrupt changes in the data under analysis. Therefore, different noise levels will also lead to different persistence of data changes. Thus, the second confidence level can be determined based on these two aspects of the data characteristics under analysis. Furthermore, considering that real meteorological anomalies often manifest as anomalies on multiple radar parameters simultaneously, i.e., the change trends of multiple radar data are highly similar and the changes in abnormal meteorological conditions are widespread, while the range of noise is significantly smaller than the range of meteorological changes, it generally only affects a single phased array radar. Therefore, a second confidence level of the data to be analyzed is determined based on the degree of difference between the data to be analyzed and the data sequences of any scanning parameter acquired by other phased array radars in the same area that acquired the data to be analyzed under the same scanning period.

[0066] Therefore, the second confidence level of the data to be analyzed can be determined, including:

[0067] Calculate the standard deviation of the first derivative of each data point in the data to be analyzed, and denote the reciprocal of the standard deviation of the first derivative as the local stationary characterization term of the data to be analyzed.

[0068] Calculate the average duration of each single trend of change within the data to be analyzed, and denote it as the local change persistence characterization term of the data to be analyzed;

[0069] The phased array radar that acquires the data to be analyzed is denoted as the current radar, and other phased array radars in the same area as the current radar are denoted as comparison radars. The data sequence of any scanning parameter acquired by each comparison radar in the current scanning cycle is denoted as the comparison sequence. The average difference between the data to be analyzed and each comparison sequence is calculated, and the reciprocal of the average difference is denoted as the extensive characterization term of the change of the data to be analyzed.

[0070] A second confidence level of the data to be analyzed is determined based on the local change stationary characterization term, the local change persistent characterization term, and the change widespread characterization term. The second confidence level is proportional to the local change stationary characterization term, the local change persistent characterization term, and the change widespread characterization term.

[0071] As a further preferred option, the second confidence level is:

[0072]

[0073] in, This represents the second confidence level of the data to be analyzed in the i-th scan cycle. This refers to normalization, such as linear normalization, norm normalization, etc. It represents the standard deviation of the first derivative of each data point in the data to be analyzed in the i-th scan cycle. It reflects the stability of the change of the data in the data to be analyzed or the consistency of the change trend. The stronger the stability of the change, the more uniform the first derivative is, and the smaller the corresponding standard deviation is. This represents the dynamic time-warped distance between the m-th phased array radar (among other phased array radars in the same region as the phased array radar acquiring the data to be analyzed in the i-th scan period) and the data to be analyzed in the same scan period. This represents the total number of other phased array radars located in the same area as the phased array radar acquiring the data to be analyzed in the i-th scan period. The average DTW value, or average difference, represents the data collected by the current phased array radar in the same area and the data of the same type collected by another m phased array radars. The smaller the average difference value, the higher the credibility of the meteorological anomaly data detected by the current radar. This represents the duration of the nth single trend in the data to be analyzed during the i-th scan cycle. This represents the total number of single trends in the data to be analyzed during the i-th scan cycle. This represents the average duration of each upward or downward trend in the data to be analyzed. The longer the duration, the higher the credibility of it as a meteorological anomaly. The method for obtaining the duration of each single trend is as follows: take the time interval between any two data points that change signs (here, a change from 0 to positive or negative, or from positive or negative to 0, is also considered a change of sign) as the duration of the corresponding single trend.

[0074] The first and second confidence levels of the data to be analyzed in the current scanning cycle can be obtained from the above. The combined confidence level can be obtained by combining the two.

[0075]

[0076] in The overall confidence level of the data to be analyzed in the i-th scanning period (characterizing the credibility of real abnormal weather). This represents the first confidence level of the data to be analyzed in the i-th scan cycle; This represents the second confidence level of the data to be analyzed in the i-th scan cycle.

[0077] S104, determine the true anomaly degree of the data to be analyzed based on the suspected anomaly degree and the comprehensive confidence degree, and complete the anomaly detection for the meteorological category corresponding to any one of the scanning parameters based on the true anomaly degree.

[0078] Then, based on the overall confidence level and the suspected anomaly level of the data to be analyzed, its true anomaly level can be determined:

[0079]

[0080] in Let represent the true anomaly degree of the data to be analyzed in the i-th scan cycle. This indicates the degree of suspected anomaly in the data to be analyzed during the i-th scan cycle. Let be the overall confidence level of the data to be analyzed in the i-th scan cycle. This indicates that the higher the suspected anomaly level of the data to be analyzed, and the higher the overall confidence level, the higher its true anomaly level, and the more likely it is to be a real meteorological anomaly.

[0081] Based on the true anomaly degree, anomaly detection for the meteorological category corresponding to the data to be analyzed can be completed, including:

[0082] Multiple anomaly thresholds are set according to the anomaly level of the abnormal weather corresponding to any of the scanning parameters. The actual anomaly is compared with the anomaly threshold, and an abnormal weather alarm of the corresponding warning level is issued based on the comparison result between the actual anomaly and the anomaly threshold.

[0083] For example, anomaly thresholds of 0.6 and 0.7 are set. When the true anomaly of the data to be analyzed in the current scanning cycle exceeds 0.6, a yellow warning is issued for the meteorological type corresponding to the data to be analyzed, indicating that the meteorological type may soon become abnormal. When the true anomaly exceeds 0.7 and continues for 10 scanning cycles, a red warning is issued, indicating that the meteorological type has been continuously abnormal for a long time recently, and further analysis of the meteorological data is required. When there is a high probability of severe meteorological disasters, a notice to deal with extreme weather is issued.

[0084] This invention is based on the working principle of phased array radar. It takes a certain scanning parameter data under the current scanning cycle as the data to be analyzed. It determines the degree of suspected anomaly and the confidence level of the abnormal weather by considering the overall size of the data to be analyzed and the differences between the data characteristics under the actual abnormal weather conditions and the data characteristics under the influence of noise. In this way, the true degree of anomaly under the cycle is determined by considering the entire scanning cycle. This eliminates the abnormal weather discrimination error caused by noise interference when comparing data values ​​with thresholds at a single data point, and significantly improves the accuracy of abnormal weather detection based on phased array radar.

[0085] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A method for detecting abnormal weather conditions based on phased array radar, characterized in that, The method includes: During meteorological monitoring, the data sequence of any scanning parameter of the phased array radar in the current scanning cycle is acquired as the data to be analyzed. The degree of suspected anomaly in the data to be analyzed is determined based on the number of data points in the data to be analyzed that exceed the normal range of any one of the scanning parameters and the extent of the exceedance. The first confidence level of the data to be analyzed is determined based on the overall volatility and data completeness of the data to be analyzed; the second confidence level of the data to be analyzed is determined based on the local stability and local persistence of the local changes of the data to be analyzed; and the comprehensive confidence level is determined based on the first confidence level and the second confidence level. The true anomaly of the data to be analyzed is determined based on the suspected anomaly and the overall confidence level, and anomaly detection for the meteorological category corresponding to any one of the scanning parameters is completed based on the true anomaly.

2. The abnormal weather detection method based on phased array radar according to claim 1, characterized in that, Determining the suspected anomaly degree of the data to be analyzed includes: The normalized value of the number of data points in the data to be analyzed that exceed the normal range of any one of the scanning parameters is calculated and denoted as the first suspected anomaly characterization term; The normalized value of the sum of the absolute values ​​of the portions of all data points in the data to be analyzed that exceed the normal range of any one of the scanning parameters is recorded as the second suspected anomaly characterization term. The suspected anomaly score of the data to be analyzed is constructed based on the first suspected anomaly score and the second suspected anomaly score, wherein the suspected anomaly score is proportional to both the first and second suspected anomaly scores.

3. The abnormal weather detection method based on phased array radar according to claim 1, characterized in that, The first confidence level of the data to be analyzed is also determined based on the signal-to-noise ratio of the data to be analyzed.

4. The abnormal weather detection method based on phased array radar according to claim 3, characterized in that, Determining the first confidence level of the data to be analyzed includes: The data points in the data to be analyzed whose trend changes are recorded as change-sign data points. The absolute value of the data difference between any change-sign data point and the previous data point is recorded as the data change amount of any change-sign data point. The normalized value of the product of the mean of the data change amounts of all the change-sign data points and the number of change-sign data points is calculated and recorded as the overall fluctuation characterization term of the data to be analyzed. Multiply the ratio of the number of missing data points in the data to be analyzed to the number of data points that should normally be collected in the data to be analyzed by the maximum number of consecutively missing data points in the data to be analyzed, and record the normalized value of the resulting product as the missing data characteristic term of the data to be analyzed. The normalized value of the reciprocal of the signal-to-noise ratio of the data to be analyzed is denoted as the disturbance level characterization term of the data to be analyzed. The first confidence level of the data to be analyzed is determined based on the overall fluctuation characterization, the missing characterization, and the disturbance degree characterization. The first confidence level is inversely proportional to the overall fluctuation characterization, the missing characterization, and the disturbance degree characterization.

5. The abnormal weather detection method based on phased array radar according to claim 1, characterized in that, Furthermore, a second confidence level of the data to be analyzed is determined based on the degree of difference between the data to be analyzed and the data sequence of any scanning parameter acquired by other phased array radars in the same area that acquired the data to be analyzed under the same scanning period.

6. The abnormal weather detection method based on phased array radar according to claim 5, characterized in that, Determining the second confidence level of the data to be analyzed includes: Calculate the standard deviation of the first derivative of each data point in the data to be analyzed, and denote the reciprocal of the standard deviation of the first derivative as the local stationary characterization term of the data to be analyzed. Calculate the average duration of each single trend of change within the data to be analyzed, and denote it as the local change persistence characterization term of the data to be analyzed; The phased array radar that acquires the data to be analyzed is denoted as the current radar, and other phased array radars in the same area as the current radar are denoted as comparison radars. The data sequence of any scanning parameter acquired by each comparison radar in the current scanning cycle is denoted as the comparison sequence. The average difference between the data to be analyzed and each comparison sequence is calculated, and the reciprocal of the average difference is denoted as the extensive characterization term of the change of the data to be analyzed. A second confidence level of the data to be analyzed is determined based on the local change stationary characterization term, the local change persistent characterization term, and the change widespread characterization term. The second confidence level is proportional to the local change stationary characterization term, the local change persistent characterization term, and the change widespread characterization term.

7. The abnormal weather detection method based on phased array radar according to claim 1, characterized in that, Determining the true anomaly of the data to be analyzed includes: The product of the first confidence level and the second confidence level is taken as the comprehensive confidence level, and the product of the comprehensive confidence level and the suspected anomaly level is taken as the true anomaly level.

8. The abnormal weather detection method based on phased array radar according to any one of claims 1 to 6, characterized in that, Based on the true anomaly degree, anomaly detection is performed for the meteorological category corresponding to any of the scanning parameters, including: Multiple anomaly thresholds are set according to the anomaly level of the abnormal weather corresponding to any of the scanning parameters. The actual anomaly is compared with the anomaly threshold, and an abnormal weather alarm of the corresponding warning level is issued based on the comparison result between the actual anomaly and the anomaly threshold.

Images