Radar real-time monitoring and early warning method and device for weather target in three-dimensional system

Abstract

The invention discloses a radar real-time monitoring and early warning method and device for a weather target in a three-dimensional system, and relates to the technical field of radar monitoring and early warning, and the method comprises the following steps: S1, extracting early warning features; s2, a multi-elevation automatic identification method; s3, identifying a strong convective cloud cluster; s4, analyzing infrared channel parameters; and S5, weather target early warning. According to the method, space-time matching is carried out on satellite data and radar data in a built three-dimensional scene, so that the severe convection contours are recognized and extracted, statistical analysis is carried out on various parameters of cloud clusters and echoes within the overlapping range of the recognized convection contours, the severe convection echoes are extracted through a multi-elevation automatic recognition method, and the recognition accuracy of the severe convection contours is improved. And then radar parameter feature recognition is carried out, a foundation is laid for subsequent severe convective weather early warning, a cloud cluster main body can be effectively approached by recognizing the severe convective cloud cluster and analyzing infrared channel parameters, the feature parameters of the severe convective cloud cluster are efficiently extracted, and the weather target recognition precision is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of radar monitoring and early warning, and specifically provides a method and device for real-time radar monitoring and early warning of weather targets in a three-dimensional system. Background Technique

[0002] Severe convective precipitation systems pose a serious threat to people's production, life, and the safety of their lives and property. Therefore, accurate forecasting of convective systems plays a very important role in short-term weather forecasting and meteorological disaster prevention and mitigation. During the entire life cycle of the occurrence, development, and extinction of the entire convective precipitation system, the monitoring of convection is conducive to better early warning of possible supercells or severe convective systems. In the Chinese patent with the application number 202211487100.2, there is disclosed "a method for monitoring the initiation of convection based on ground-based cloud radar and satellite data, collecting data, calculating the echo structure texture, horizontal and vertical scales, and gradients of weather radar observation data, and automatically identifying the central area of the echo; screening out convection cases and dividing them into different levels; according to the time, location, and level of the screened convection cases, extracting cloud radar, satellite visible light, and infrared channel observation data to form a training set of cloud radar, satellite, and convection cases; using the stepwise regression method to select predictors from the cloud radar observation data, satellite visible light, and infrared detection data to determine the predictors for monitoring the initiation of convection and its level".

[0003] The above document can use a convection monitoring model to monitor convection in data, realizing the monitoring and early warning of the initiation and intensity level of convection. However, the above document only predicts the outflow intensity during the data collection stage, and does not specifically implement the monitoring and early warning in what way. After obtaining the monitoring data, although it is judged as severe convection, it cannot accurately analyze what type of convection it is, and cannot predict the specific weather changes and occurrence conditions caused by the convection, which will affect the early warning effect. In addition, since most severe convective weather with weak precipitation is not covered by severe convective cloud clusters before it occurs, but the edges of the cloud clusters are mostly close to the places where severe convection occurs, identifying and tracking severe convective cloud clusters plays an important role in timely capturing weather target data. Summary of the Invention

[0004] The purpose of the present invention is to provide a method and device for real-time radar monitoring and early warning of weather targets in a three-dimensional system to solve the problems raised in the above background technique.

[0005] To achieve the above purpose, the present invention provides the following technical solutions: A method and device for real-time radar monitoring and early warning of weather targets in a three-dimensional system, including the following steps:

[0006] S1. Extract warning features: Spatially and temporally match satellite and radar data in the constructed three-dimensional scene. Respectively extract severe convective contours by analyzing infrared channel parameters and using the multi-elevation angle automatic recognition method, and conduct statistical analysis of warning features on convective parameters within the overlapping range of the contours;

[0007] S2. Multi-elevation angle automatic recognition method: First, adopt multiple parameter recognition conditions for simultaneously judging the basic reflectivity factor, composite reflectivity factor, calculated values of radial velocity convergence at each elevation angle, and spectral width of the radar to extract severe convective echoes, and then conduct radar parameter feature recognition;

[0008] S3. Identify severe convective cloud clusters: Use the cloud top brightness temperature threshold of the acquired satellite severe convective cloud clusters as the initial judgment condition for severe convective cloud clusters. Use the brightness temperature data of the infrared channel to extract the cloud cluster contour of the severe convective cloud clusters, and calculate the correlation coefficient of cloud clusters with a centroid distance of less than 100 kilometers between cloud cluster centroids on adjacent time cloud images. Consider the cloud clusters that simultaneously meet the shortest centroid distance and the maximum correlation coefficient as the best-matched cloud clusters to identify and track severe convective cloud clusters;

[0009] S4. Analyze infrared channel parameters: Calculate and analyze severe convective cloud cluster parameters, where the severe convective cloud cluster parameters include radiation parameters and spatial parameters;

[0010] S5. Weather target warning: Select the convective parameters corresponding to severe convective echoes and cloud clusters. Detect severe convective cloud clusters through satellites and identify severe convective clouds through radar echo features, and then distinguish between hail and non-hail severe convective clouds, and further judge the conditions and probabilities of hail occurrence, so as to give warning prompts.

[0011] Preferably, step S2 includes the following steps:

[0012] S201. Define critical conditions for radar parameter recognition: Set the critical values of the basic reflectivity factor at each elevation angle of the radar, the critical value of the composite reflectivity factor, the critical value of the radar spectral width, and the critical value of the calculated value of radial velocity convergence at each elevation angle;

[0013] S202. When two consecutive elevation angles reach the critical conditions, it belongs to severe convective echoes, and extract severe convective echoes and output echo contours;

[0014] S203. Calculate various radar echo parameters for analyzing severe convective radar echo features. The specific calculated parameters include: basic reflectivity factor at elevation angle, standard deviation of basic reflectivity, maximum rotation speed, maximum spectral width, mean intensity and position of centroid at each elevation angle, distance of centroid at each elevation angle deviating from the mean centroid, height where the mean centroid is located, convective echo top height, maximum echo intensity, height of the strongest echo, convective thickness, maximum value and position of vertically integrated liquid water content, maximum value and position of vertically integrated liquid water content density;

[0015] S204. Determine severe convective clouds based on the characteristics of severe convective radar echoes, and then distinguish between hail - falling and non - hail - falling severe convective clouds to conduct subsequent early warnings for severe convective weather.

[0016] Preferably, in step S2, the radar warning range is determined by using the following two methods:

[0017] Method 1: Draw and identify the area of concern through human - machine interaction, that is, manually draw the selected area on the radar echo image and then give a warning.

[0018] Method 2: Automatic identification of rectangular area units for business operation, that is, use the rectangular area as the search unit and then automatically give a warning for each rectangular unit.

[0019] Preferably, in step S3, calculate the centroid positions (X CG , Y CG ) of the cloud clusters at adjacent time intervals and the correlation coefficient r. The calculation formulas are as follows:

[0020]

[0021] Among them, T ij represents the pixel temperature. To facilitate the extraction of the centroid, when T ij > 32 °C, let T ij = 0, N p represents the total number of all pixels within the convective cloud cluster, X i and Y j respectively represent the pixel longitude and latitude, T f (i, j) and T g (i, j) respectively represent the pixel cloud - top temperatures of cloud cluster f and the matching cloud cluster g, and respectively represent the average cloud - top temperatures of cloud cluster f and the matching cloud cluster g. Taking the centroid of the cloud cluster as the center, m and n respectively represent the number of pixels of the rectangle covering the cloud cluster in the longitude direction and the latitude direction, and i and j represent the pixel serial numbers.

[0022] Preferably, in step S4, the radiation parameters and spatial parameters are as follows:

[0023] Radiation parameters: The cloud - top temperature gradient of the convective cloud cluster and the deep - convection index of the centroid;

[0024] Spatial parameters: The distances between the centroid position, the lowest - temperature position, the maximum - temperature - gradient position of the convective cloud cluster and the station position.

[0025] Preferably, step S5 includes the following steps:

[0026] S501. Determine the short-time heavy precipitation phenomenon using the data obtained from the satellite infrared channels 1 and 3;

[0027] S502. Perform an OR operation on the severe convective cloud clusters and echoes within a distance of 15 km from the edge of the convective contour. If the lowest brightness temperature of the cloud cluster is not higher than the centroid brightness temperature and the centroid brightness temperature is not higher than the cloud top brightness temperature threshold, it is determined as a severe precipitation convective cloud cluster. Among them, if the scale of the cloud cluster in adjacent time steps exceeds 100 kilometers, no warning is issued. For the severe convective radar echo contour, use two methods to determine the radar warning range and optimize the warning threshold method to judge the probability of hail occurrence, so as to give a warning prompt.

[0028] Preferably, step S5 further includes the following steps:

[0029] S503. The warning threshold optimization method is specifically as follows:

[0030] Determine the judgment condition for hail occurrence; H 45 ≥H 0 +H th ;

[0031] Among them, H 45 represents the echo top height of the single cell at 45 dBZ, H 0 represents the height of the 0°C isotherm of the single cell, H th represents the threshold;

[0032] Define SHI as the vertical integral of the thermal weight of the reflectivity factor profile of the storm cell through the difference prediction formula. The difference prediction formula is specifically as follows:

[0033]

[0034] Among them, H T represents the echo top height of the single cell, H 0 represents the height of the melting layer, E represents the hail kinetic energy flux, determined according to the relationship between hail and the reflectivity factor, W T (H) represents the temperature weight function, and H represents the reflectivity factor of the storm cell;

[0035] Calculate the hail parameter according to the relationship between the vertical integral value SHI of the thermal weight of the reflectivity factor profile of the storm cell and the height of the 0°C layer;

[0036]

[0037] Among them, P represents the probability of hail formation, SHI represents the vertical integral of the thermal weight of the reflectivity factor profile of the storm cell, WT represents the warning threshold function, calculated according to the height of the 0°C layer. If WT < 20, the warning threshold is 20, otherwise, WT = 57.5H 0 -121.

[0038] A radar real-time monitoring and early warning device for weather targets in a three-dimensional system, comprising the following modules:

[0039] A feature analysis module, which performs spatio-temporal matching of satellite and radar data in a constructed three-dimensional scene, and conducts statistical analysis of early warning features of convective parameters within the overlapping range of contours by extracting severe convective contours.

[0040] An echo extraction module, which first uses multiple parameter recognition conditions for simultaneously judging the basic reflectivity factor, composite reflectivity factor, calculated values of radial velocity convergence at each elevation angle of the radar, and spectrum width to extract severe convective echoes, and then conducts radar parameter feature recognition.

[0041] A cloud cluster recognition module, which takes the cloud top brightness temperature threshold of the satellite severe convective cloud cluster obtained as the initial judgment condition for the severe convective cloud cluster, uses the brightness temperature data of the infrared channel to extract the cloud cluster contour of the severe convective cloud cluster, calculates the correlation coefficient of cloud clusters with a centroid distance of less than 100 kilometers between cloud cluster centroids on adjacent time cloud images, and regards the cloud cluster that simultaneously satisfies the shortest centroid distance and the maximum correlation coefficient as the best matching cloud cluster to identify and track the severe convective cloud cluster.

[0042] A parameter analysis module, which calculates and analyzes severe convective cloud cluster parameters, where the severe convective cloud cluster parameters include radiation parameters and spatial parameters.

[0043] A target early warning module, which selects the convective parameters corresponding to the severe convective echo and the cloud cluster, monitors the severe convective cloud cluster through the satellite, judges the severe convective cloud through the radar echo characteristics, then differentiates and judges the hail and non-hail severe convective clouds, and further judges the conditions and probabilities of hail occurrence, so as to give early warning prompts.

[0044] Compared with the prior art, the beneficial effects of the present invention are:

[0045] 1. The present invention performs spatio-temporal matching of satellite and radar data in a constructed three-dimensional scene, thereby identifying and extracting severe convective contours, conducting statistical analysis on various parameters of cloud clusters and echoes within the overlapping range of the identified convective contours, performing an OR operation on severe convective cloud clusters and echoes with a distance of less than 15 km between the edges of the convective contours to quickly identify and track severe convective cloud clusters, solving the phenomenon that most severe convective weather with weak precipitation has no severe convective cloud cluster coverage before occurrence, but the edges of the cloud clusters are mostly close to the severe convective occurrence areas. Extracting severe convective echoes through the multi-elevation automatic recognition method and then conducting radar parameter feature recognition lays a foundation for subsequent severe convective weather early warning. By identifying severe convective cloud clusters and analyzing infrared channel parameters, it can effectively approach the main body of the cloud cluster and efficiently extract the characteristic parameters of the severe convective cloud cluster, improving the accuracy of weather target recognition.

[0046] 2. The present invention determines severe convective clouds based on the characteristics of severe convective radar echoes, then differentiates between hail - producing and non - hail - producing severe convective clouds, and thus conducts subsequent severe convective weather warnings. By formulating a method for determining the radar warning range, it reduces the calculation time for searching and identifying convective echoes within the entire radar image due to the large amount of calculation in the multi - elevation - angle automatic recognition method, improving the recognition efficiency. By selecting convective parameters corresponding to severe convective echoes and cloud clusters, it conducts severe convective warnings, and judges the probability of hail occurrence according to the alarm threshold optimization method. When the judgment condition for determining hail occurrence is reached, a warning is triggered, and the probability of hail formation is calculated for accurate warning prompts. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 It is a schematic diagram of the overall method flow provided by an embodiment of the present invention;

[0048] Figure 2 It is a schematic diagram of the device structure provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0049] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0050] Please refer to Figures 1 to 2 , the present invention provides a technical solution: a method and device for real - time radar monitoring and warning of weather targets in a three - dimensional system, including the following modules:

[0051] S1. Extract warning features: Match satellite and radar data in the constructed three - dimensional scene in terms of time and space. Respectively extract severe convective contours by analyzing infrared channel parameters and using the multi - elevation - angle automatic recognition method, and conduct statistical analysis of warning features for convective parameters within the overlapping range of the contours;

[0052] S2. Multi - elevation - angle automatic recognition method: First, use multiple parameter recognition conditions for simultaneously judging the basic reflectivity factor, composite reflectivity factor, calculated values of radial velocity convergence at each elevation angle, and spectrum width of the radar at each elevation angle to extract severe convective echoes, and then conduct radar parameter feature recognition;

[0053] S3. Identify severe convective cloud clusters: Use the cloud top brightness temperature threshold of the satellite severe convective cloud clusters obtained as the initial judgment condition for severe convective cloud clusters. Extract the cloud cluster contours of severe convective cloud clusters using the brightness temperature data of the infrared channel, and calculate the correlation coefficient of the cloud clusters within 100 kilometers of the centroid distance between cloud clusters on adjacent time cloud images. Consider the cloud clusters that simultaneously meet the shortest centroid distance and the maximum correlation coefficient as the best-matched cloud clusters to identify and track severe convective cloud clusters;

[0054] S4. Analyze infrared channel parameters: Calculate and analyze severe convective cloud cluster parameters, where the severe convective cloud cluster parameters include radiation parameters and spatial parameters;

[0055] S5. Weather target warning: Select the convective parameters corresponding to severe convective echoes and cloud clusters. Detect severe convective cloud clusters through satellite monitoring and identify severe convective clouds through radar echo characteristics. Then, distinguish between hail-producing and non-hail-producing severe convective clouds, and further judge the conditions and probabilities of hail occurrence, so as to give warning prompts.

[0056] The step S2 includes the following steps:

[0057] S201. Define the critical conditions for radar parameter identification: Set the critical values of the basic reflectivity factor at each elevation angle of the radar, the critical value of the composite reflectivity factor, the critical value of the radar spectral width, and the critical value of the calculated value of the radial velocity convergence at each elevation angle;

[0058] S202. When two consecutive elevation angles reach the critical conditions, it belongs to severe convective echoes, and extract the severe convective echoes and output the echo contours;

[0059] S203. Calculate various radar echo parameters for analyzing the characteristics of severe convective radar echoes. The specific calculation parameters include: the basic reflectivity factor at the elevation angle, the standard deviation of the basic reflectivity, the maximum rotational velocity, the maximum spectral width, the mean intensity and position of the centroid at each elevation angle, the distance of the centroid at each elevation angle from the mean centroid, the height where the mean centroid is located, the convective echo top height, the maximum echo intensity, the height of the strongest echo, the convective thickness, the maximum value and position of the vertically integrated liquid water content, the maximum value and position of the vertically integrated liquid water content density;

[0060] S204. Identify severe convective clouds based on the characteristics of severe convective radar echoes, and then distinguish between hail-producing and non-hail-producing severe convective clouds, so as to carry out subsequent severe convective weather warnings;

[0061] When the convection reaches the 32 dBZ echo level and the horizontal scale is 5 to 25 kilometers, and the vertically integrated liquid water content reaches 2 kg / m 2 and the echo top height is 4.5 kilometers away from the radar station, then the convective cloud belongs to ordinary convective cloud, and convective ordinary precipitation is generated;

[0062] When the convective echo reaches a level of 40 dBZ, with a horizontal scale of 5 to 25 kilometers and a vertically integrated liquid water content reaching 5 kg / m 2 , and the convective echo top height is 7 kilometers away from the radar antenna, then the convective cloud belongs to a severe convective cloud;

[0063] When the 45 dBZ echo top height of the severe convective cloud reaches more than half of the 18.3 dBZ echo top height, and the strongest echo height still reaches more than half of the 18.3 dBZ echo top height, it belongs to a hail-producing severe convective cloud, accompanied by hail;

[0064] As the hail-producing severe convective cloud continues to develop, when the maximum echo intensity reaches or exceeds 48 dBZ, the maximum vertically integrated liquid water content reaches or exceeds 15 kg / m 2 , the 18.3 dBZ echo top height from the radar antenna reaches or exceeds 8 kilometers, the change in the maximum vertically integrated liquid water content between adjacent time intervals reaches or exceeds 15 kg / m 2 , and the density of the maximum vertically integrated liquid water content in this time interval exceeds 2.2 g / m 3 , then the convection has developed to the point where hail is about to fall, and a hail warning is issued;

[0065] The radar warning characteristics of a hail-producing cell are as follows: When the convective cell simultaneously has the characteristics that the maximum echo intensity is greater than or equal to 50 dBZ, the height where the strongest echo is located is greater than 3 kilometers, the height where the 45 dBZ echo is located is greater than or equal to half of the 18 dBZ precipitation echo top height, and the vertically integrated liquid water content above the 0 °C layer is greater than 0, it belongs to a hail-producing cell;

[0066] When the 45 dBZ precipitation echo top height exceeds the 0 °C layer height by 0.5 kilometers or more, and the vertically integrated liquid water content above the 0 °C layer is greater than or equal to 3.8 kg / m 2 , short-term heavy precipitation accompanied by hail or other convective precipitation weather will occur within the next 30 minutes;

[0067] When the jump increment of the vertically integrated liquid water content of the hail-producing cell is greater than or equal to 6 kg / m 2 and the density is greater than or equal to 2.2 g / m 3 , if a second jump in the vertically integrated liquid water content of the hail-producing cell occurs, hail weather will be produced within the next 30 minutes under the influence of the hail-producing cell;

[0068] The radar warning characteristics of a non-hail-producing cell: When the height where the strongest echo is located is less than 3.5 kilometers, and the height where the 45 dBZ echo is located is less than half of the 18 dBZ precipitation echo top height, it belongs to a non-hail-producing cell, and thunderstorms or heavy precipitation weather will occur within the next 30 minutes;

[0069] In step S2, the radar warning range is determined by using the following two methods:

[0070] Method 1: Draw the area of interest through human-computer interaction and perform recognition, that is, manually draw the selected area on the radar echo image and then give an early warning;

[0071] Method 2: Automatic recognition of rectangular area units for business operation, that is, use the rectangular area as the search unit and then automatically give an early warning for each rectangular unit;

[0072] Since the calculation amount of the multi-elevation angle automatic recognition method is large, in order to reduce the calculation time for searching and recognizing convective echoes in the entire radar image and improve the recognition efficiency, a method is set to determine the radar early warning range;

[0073] In step S3, calculate the centroid positions (X CG , Y CG ) and the correlation coefficient r of cloud clusters at adjacent times. The calculation formulas are as follows:

[0074]

[0075] Among them, T ij represents the pixel temperature. To facilitate the extraction of the centroid, when T ij > 32 °C, let T ij = 0. N p represents the total number of pixels in the convective cloud cluster. X i and Y j respectively represent the longitude and latitude of the pixel. T f (i, j) and T g (i, j) respectively represent the cloud top temperatures of the pixels of cloud cluster f and the matching cloud cluster g. and respectively represent the average cloud top temperatures of cloud cluster f and the matching cloud cluster g. Taking the centroid of the cloud cluster as the center, m and n respectively represent the number of pixels of the rectangle covering the cloud cluster in the longitude direction and the latitude direction, and i and j represent the pixel numbers;

[0076] In step S4, the radiation parameters and spatial parameters are as follows:

[0077] Radiation parameters: The cloud top temperature gradient of the convective cloud cluster and the deep convection index of the centroid;

[0078] The calculation formula of the temperature gradient is:

[0079]

[0080] Among them, T 12 and T 13 respectively represent the differences in infrared multi-spectral bands at the centroid position. G represents the temperature gradient, T represents the cloud top temperature of the convective cloud cluster, i and j are pixel numbers, T CG1 , T CG2 and T CG3They respectively represent the cloud top temperatures of infrared 1, infrared 2, and infrared 3 of the centroid, and the peak value of the average cloud top temperature of the infrared channel is greater than or equal to 10 degrees Celsius;

[0081] The calculation formula for the deep convection index of the centroid is as follows:

[0082]

[0083] where T CG represents the cloud top temperature at the centroid position. The temperature of the atmosphere near 400 hPa is 250 K. The DCI value represents the degree of deep convection clouds with cloud tops equal to or higher than 400 hPa. The probability of developing into strong convection clouds before the onset of strong convection increases as the DCI value increases;

[0084] Spatial parameters: the distances between the centroid position of the convective cloud cluster, the position of the lowest temperature, the position of the maximum temperature gradient, and the station position;

[0085] The step S5 includes the following steps:

[0086] S501. Use the data obtained from the satellite infrared 1 and infrared 3 channels to judge the short-term heavy precipitation phenomenon;

[0087] The peak value of the average cloud top temperature gradient of the infrared 1 and infrared 3 channels is greater than or equal to 10 °C. For the minimum value of the cloud top temperature T min1 and the maximum value of the cloud top temperature gradient GT max1 , the deep convection index DCI of the infrared 1 channel, and the minimum value of the cloud top temperature T min3 of the infrared 3 channel. If T min1 < -20 °C, GT max1 > 8 °C, DCI > 10, and T min3 < -30 °C are satisfied simultaneously, then the short-term heavy precipitation phenomenon occurs;

[0088] S502. Since most strong convective weather with weak precipitation is not covered by strong convective cloud clusters before it occurs, but the edges of the cloud clusters are mostly close to the places where strong convection occurs. Therefore, perform an OR operation on the strong convective cloud clusters and echoes within 15 km of the convective contour edge. If the lowest bright temperature of the cloud cluster is not higher than the bright temperature of the centroid and the bright temperature of the centroid is not higher than the cloud top bright temperature threshold, then it is determined as a strong precipitation convective cloud cluster. If the scale of the cloud cluster in adjacent time steps exceeds 100 kilometers, no warning is issued. For the strong convective radar echo contour, use two methods to determine the radar warning range and judge the probability of hail occurrence according to the warning threshold optimization method, so as to give a warning prompt;

[0089] The step S5 further includes the following steps:

[0090] The warning threshold optimization method is specifically as follows:

[0091] Determine the judgment conditions for hail occurrence; H 45 ≥H 0 +H th ;

[0092] Among them, H 45 represents the echo top height of the single cell at 45 dBZ, and H 0 represents the height of the 0°C isotherm of the single cell, and H th represents the threshold value;

[0093] Define SHI as the vertical integral of the thermal weight of the reflectivity factor profile of the storm cell through the difference prediction formula. The specific difference prediction formula is:

[0094]

[0095]

[0096] Among them, H T represents the echo top height of the single cell, and H 0 represents the height of the melting layer. E represents the hail kinetic energy flux, which is determined according to the relationship between hail and the reflectivity factor. W T (H) represents the temperature weight function, and H represents the reflectivity factor of the storm cell;

[0097] Calculate the hail parameters according to the relationship between the vertical integral value SHI of the thermal weight of the reflectivity factor profile of the storm cell and the height of the 0°C layer;

[0098]

[0099] Among them, P represents the probability of hail generation, SHI represents the vertical integral of the thermal weight of the reflectivity factor profile of the storm cell, and WT represents the alarm threshold function, which is calculated according to the height of the 0°C layer. If WT < 20, the alarm threshold is 20. Otherwise, WT = 57.5H 0 -121;

[0100] A radar real-time monitoring and warning device for weather targets in a three-dimensional system, including the following modules:

[0101] Feature analysis module. The feature analysis module performs spatio-temporal matching on satellite and radar data in the constructed three-dimensional scene, and conducts statistical analysis of warning features of convection parameters within the overlapping range of the contours by extracting severe convection contours;

[0102] Echo extraction module. The echo extraction module first uses multiple parameter identification conditions for simultaneously judging the basic reflectivity factor, composite reflectivity factor, calculated values of radial velocity convergence at each elevation angle, and spectral width of the radar to extract severe convection echoes, and then conducts radar parameter feature identification;

[0103] Cloud cluster recognition module. The cloud cluster recognition module uses the cloud top brightness temperature threshold of the satellite severe convective cloud cluster obtained as the initial judgment condition for the severe convective cloud cluster, extracts the cloud cluster contour of the severe convective cloud cluster using the brightness temperature data of the infrared channel, and calculates the correlation coefficient of the cloud clusters within 100 kilometers between the cloud cluster centroids on the cloud images at adjacent time intervals. The cloud cluster that simultaneously satisfies the shortest centroid distance and the maximum correlation coefficient is considered the best matching cloud cluster to identify and track the severe convective cloud cluster;

[0104] Parameter analysis module. The parameter analysis module calculates and analyzes the severe convective cloud cluster parameters. Among them, the severe convective cloud cluster parameters include radiation parameters and spatial parameters;

[0105] Target warning module. The target warning module selects the convective parameters corresponding to the severe convective echo and the cloud cluster, monitors the severe convective cloud cluster through the satellite and judges the severe convective cloud through the radar echo characteristics, then differentiates and judges the hail and non-hail severe convective clouds, and further judges the conditions and probabilities of hail occurrence, so as to give a warning prompt.

[0106] It should be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not expressly listed, or also includes elements inherent to such process, method, article or device.

[0107] Although the embodiments of the present invention have been shown and described, for those of ordinary skill in the art, it can be understood that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A radar real-time monitoring and early warning method for weather targets in a three-dimensional system, characterized in that: The method comprises the following steps: S1. Extract warning features: perform spatiotemporal matching of satellite and radar data in the constructed three-dimensional scene, extract strong convective contours by analyzing infrared channel parameters and using multi-elevation angle automatic recognition method, and perform warning feature statistical analysis on convective parameters within the overlapping range of contours; S2, multi-elevation angle automatic identification method: First, the basic reflectivity factor of each elevation angle of the radar, the combined reflectivity factor, the calculated value of the radial velocity convergence at each elevation angle, and the spectrum width are used to simultaneously determine multiple parameter identification conditions, extract the strong convective echo, and then perform radar parameter feature identification; S3. Identify strong convective clouds: The cloud top brightness temperature threshold of strong convective clouds acquired from satellites is used as the initial identification condition for strong convective clouds. The cloud outlines of strong convective clouds are extracted using infrared channel brightness temperature data. The correlation coefficients of clouds with a distance of less than 100 km between the centroids of clouds on adjacent cloud images are calculated. The cloud that satisfies both the shortest centroid distance and the maximum correlation coefficient is considered the best matching cloud, so as to identify and track the strong convective clouds. S4. Analyze infrared channel parameters: calculate and analyze the parameters of severe convective cloud clusters, wherein the parameters of severe convective cloud clusters include radiation parameters and space parameters; S5. Weather target warning: Select the convective parameters corresponding to the strong convective echoes and cloud clusters, monitor the strong convective cloud clusters through satellites, and determine the strong convective clouds through radar echo characteristics. Then distinguish between hail and non-hail strong convective clouds, and then determine the conditions and probability of hail, so as to issue early warning prompts.

2. The radar real-time monitoring and early warning method for weather targets in a three-dimensional system according to claim 1 is characterized in that: The step S2 comprises the following steps: S201, defining critical conditions for radar parameter identification: setting critical values ​​of basic reflectivity factors and combined reflectivity factors at each elevation angle of the radar, and setting critical values ​​of radar spectrum width and calculated values ​​of radial velocity convergence at each elevation angle; S202, when two consecutive elevation angles reach a critical condition, it is a strong convective echo, and the strong convective echo is extracted and the echo profile is output; S203, calculating multiple radar echo parameters to analyze the strong convection radar echo characteristics, the specific calculation parameters include: elevation angle basic reflectivity factor, basic reflectivity standard deviation, maximum rotation speed, maximum spectrum width, the mean intensity and position of the centroid at each elevation angle, the distance of the centroid at each elevation angle from the mean centroid, the height of the mean centroid, the top height of the convection echo, the maximum echo intensity, the strongest echo height, the convection thickness, the maximum value and position of the vertical cumulative liquid water content, and the maximum value and position of the vertical cumulative liquid water content density; S204. Determine the strong convective clouds according to the strong convective radar echo characteristics, and then distinguish between hail-prone and non-hail-prone strong convective clouds, so as to issue subsequent severe convective weather warnings.

3. The radar real-time monitoring and early warning method for weather targets in a three-dimensional system according to claim 1 is characterized in that: In step S2, the radar warning range is determined by using the following two methods: Method 1: Draw the area of ​​concern and identify it through human-computer interaction, that is, manually draw the selected area on the radar echo image and then issue an early warning; Method 2: Automatically identify rectangular area units used for business operations, that is, use rectangular areas as search units, and then automatically issue warnings for each rectangular unit.

4. The radar real-time monitoring and early warning method for weather targets in a three-dimensional system according to claim 1, characterized in that: In step S3, the position of the cloud mass center (X CG ,Y CG ) and the correlation coefficient r, calculated as: Among them, T ij Represents the pixel temperature. To facilitate the extraction of the centroid, when T ij >32℃, set T ij =0,N p represents the number of pixels in the convective cloud cluster, X i and Y j Represent the longitude and latitude of the pixel, T f (i,j) and T g (i, j) represent the cloud top temperatures of the cloud cluster f and the matching cloud cluster g, respectively. and T g They represent the average cloud top temperatures of cloud cluster f and matching cloud cluster g, respectively, with the cloud cluster centroid as the center. m and n represent the number of pixels in the longitude and latitude directions of the rectangle covering the cloud cluster, respectively. i and j represent the pixel numbers.

5. The radar real-time monitoring and early warning method for weather targets in a three-dimensional system according to claim 1, characterized in that: In step S4, the radiation parameters and space parameters are as follows: Radiation parameters: temperature gradient of cloud top and deep convection index of mass center of convective cloud cluster; Spatial parameters: the distance between the centroid of the convective cloud cluster, the lowest temperature position, the maximum temperature gradient position and the measuring station.

6. The radar real-time monitoring and early warning method for weather targets in a three-dimensional system according to claim 1, characterized in that: The step S5 comprises the following steps: S501, using the data obtained from the satellite infrared 1 and infrared 3 channels to determine the short-term heavy rainfall phenomenon; S502. Perform an OR operation on the strong convective cloud cluster and the echo within 15 km of the edge of the convective contour. If the lowest brightness temperature of the cloud cluster is not higher than the centroid brightness temperature and the centroid brightness temperature is not higher than the cloud top brightness temperature threshold, it is determined to be a heavy precipitation convective cloud cluster. If the scale of the cloud clusters at adjacent times exceeds 100 kilometers, no warning will be issued. For the strong convective radar echo contour, two methods for determining the radar warning range and the probability of hail occurrence based on the alarm threshold optimization method are used to determine the probability of hail, so as to issue a warning prompt.

7. The radar real-time monitoring and early warning method for weather targets in a three-dimensional system according to claim 6, characterized in that: The step S5 further comprises the following steps: S503: The specific method for optimizing the alarm threshold is as follows: Determine the conditions for judging the occurrence of hail; 45 ≥H0+H th ; Among them, H 45 It indicates the top height of the single 45dBZ echo, H0 indicates the height of the single 0℃ isotherm, H th Indicates the threshold value; The difference prediction formula is used to define SHI as the thermal weighted vertical integral of the reflectivity factor profile of the storm cell. The difference prediction formula is as follows: Among them, H T represents the height of the monomer echo top, H0 represents the height of the melting layer, E represents the hail kinetic energy flux, which is determined according to the relationship between hail and reflectivity factor, and W T (H) represents the temperature weight function, and H represents the reflectivity factor of the storm cell; The hail parameters are calculated based on the relationship between the thermal weight vertical integral value SHI of the storm cell reflectivity factor profile and the 0℃ layer height. Where P represents the probability of hail formation, SHI represents the thermal weighted vertical integral of the reflectivity factor profile of the storm cell, and WT represents the alarm threshold function, which is calculated based on the height of the 0℃ layer. If WT is less than 20, the alarm threshold is 20, otherwise, WT = 57.5H0-121.

8. A radar real-time monitoring and early warning device for weather targets in a three-dimensional system, characterized in that: The radar real-time monitoring and early warning device is applicable to a radar real-time monitoring and early warning method for weather targets in a three-dimensional system according to any one of claims 1 to 7, and comprises the following modules: A feature analysis module, which performs spatiotemporal matching of satellite and radar data in a constructed three-dimensional scene, extracts strong convective contours, and performs early warning feature statistical analysis on convective parameters within the overlapping range of contours; The echo extraction module first extracts strong convective echoes by simultaneously judging the basic reflectivity factor of each elevation angle of the radar, the combined reflectivity factor, the calculated value of the radial velocity convergence at each elevation angle, and the spectrum width, and then performs radar parameter feature recognition; A cloud cluster identification module, which uses the cloud top brightness temperature threshold of the satellite strong convective cloud cluster as the initial identification condition of the strong convective cloud cluster, uses the infrared channel brightness temperature data to extract the cloud cluster contour of the strong convective cloud cluster, and calculates the correlation coefficient of the cloud clusters within 100 kilometers between the cloud cluster mass centers on the adjacent time cloud images, and considers the cloud cluster that satisfies both the shortest mass center distance and the maximum correlation coefficient as the best matching cloud cluster, so as to identify and track the strong convective cloud cluster; A parameter analysis module, wherein the parameter analysis module calculates and analyzes the parameters of the severe convective cloud cluster, wherein the parameters of the severe convective cloud cluster include radiation parameters and space parameters; A target warning module selects convective parameters corresponding to strong convective echoes and cloud clusters, monitors strong convective cloud clusters through satellites, and determines strong convective clouds through radar echo characteristics, then distinguishes between hail and non-hail strong convective clouds, and further determines the conditions and probability of hail occurrence, thereby providing early warning prompts.

Images