A Targeted Lightning Grading Early Warning Method and System Based on Multi-Source Detection Data

Abstract

The present invention discloses a targeted lightning grading warning method and system based on multi-source detection data, which generates lightning level warning information through real-time detection data; obtains the lightning storm area by using the real-time radar reflectivity CAPPI data obtained based on the real-time vertical temperature stratification data through a preset method, and continuously updates the lightning storm area; calculates the optical flow vector field corresponding to the current moment; predicts multiple positions that the lightning storm area corresponding to the current moment will pass through within a preset time period through the optical flow vector field; determines whether there is a position corresponding to a lightning storm area covering the bull's-eye among the passed positions, and if so, obtains the time when the forefront of the lightning storm area reaches the bull's-eye area as the estimated impact time; generates lightning approaching warning information through the lightning level warning information and / or the estimated impact time; that is, the present invention provides progressive lightning warning based on the user's location, realizing targeted and customizable warning services.

Description

Technical Field

[0001] The present invention relates to the field of lightning warning, and particularly to a targeted lightning classification warning method and system based on multi-source detection data. Background Art

[0002] Due to its powerful current, scorching high temperature, strong electromagnetic radiation, violent shock wave and other physical effects, lightning can cause huge destructive effects instantly, resulting in lightning disasters. With the global climate change and the development of social economy, the economic losses and social impacts caused by lightning disasters are increasing, and the affected areas involve various industries. The domestic and foreign lightning nowcasting schemes mainly include:

[0003] I. Using an atmospheric electric field meter alone for lightning monitoring and warning, with limited ranging and inability to accurately locate lightning;

[0004] II. Nowcasting based on radar and lightning location data, analyzing echo products corresponding to lightning data. Generally, it is considered that when the echo with an intensity of 40 dBZ develops to the height of the -10°C layer, it can be used as the best factor for predicting the initial lightning occurrence. This scheme improves the accuracy and timeliness of warning by analyzing the relationship between specific echo intensity and lightning activity, but there are still deficiencies in the prediction affected by thunderstorm clouds;

[0005] III. Based on the integrated application of multiple data, such as the Lightning Nowcasting System and the Lightning Nowcasting Warning System (CAMS-LNWS). This scheme demonstrates the potential of integrating multiple data to improve the warning effect, but still needs to deeply explore data integration, model optimization and personalized services.

[0006] In view of the limitations and uncertainties of single data application, it is an inevitable trend to use multi-source monitoring data for lightning short-term warning. In recent years, there have been certain technical and method reserves for radar three-dimensional data mining and analysis, including applications in storm tracking, extrapolation, etc., but only for decision-making services and public applications in large areas. And the public lightning warning is for the entire territorial scope, and it is insufficient in refinement and pertinence for users in specific locations. Summary of the Invention

[0007] In order to fully carry out the research and application of lightning multi-source detection data, transform the existing technical reserves, integrate industry application scenarios, provide customized targeted lightning nowcasting classification warning for users in specific regions, concisely inform users of the estimated storm impact time, and meet users' needs for refined lightning short-term warning, the present invention proposes a targeted lightning classification warning method based on multi-source detection data, including:

[0008] Taking the location of the user as the center, set multiple warning areas of different levels; the higher the level, the closer the outer edge of the corresponding warning area is to the center; the warning area includes the center area;

[0009] Obtain real-time detection data, and determine the lightning warning target level based on the multiple warning areas corresponding to the user and the real-time detection data, and generate lightning level warning information; the real-time detection data includes three-dimensional lightning data and atmospheric electric field data;

[0010] Obtain real-time vertical temperature stratification data, take the height of the temperature layer corresponding to a specific temperature in the real-time vertical temperature stratification data as the target height, and obtain the real-time radar reflectivity CAPPI data at the target height;

[0011] Use the real-time radar reflectivity CAPPI data to obtain the lightning storm area through a preset method, and continuously update the lightning storm area;

[0012] Calculate the optical flow vector field corresponding to the current moment in real time through the reflectivity factor data of the radar at the target height at the current moment and the previous moment, obtain the moving direction and moving speed of the current radar echo through the optical flow vector field, and predict the multiple positions that the lightning storm area will pass through within a preset time period through the moving direction and moving speed;

[0013] Judge whether there is a position among the passed positions where the corresponding lightning storm area covers the center. If so, obtain the time when the forefront of the lightning storm area reaches the center area as the estimated impact time;

[0014] Generate lightning approaching warning information through the lightning level warning information and / or the estimated impact time.

[0015] Further, obtain real-time detection data, and determine the lightning warning target level based on the multiple warning areas corresponding to the user and the real-time detection data; specifically:

[0016] Obtain the three-dimensional lightning data detected in real time by multiple lightning locators in the lightning location system, determine the location where the cloud-to-ground flash occurs based on the obtained three-dimensional lightning data, and determine the first pre-used lightning warning level based on the warning area where the cloud-to-ground flash occurs;

[0017] Obtain the atmospheric electric field data in real time through the atmospheric electric field instrument, calculate the field strength jump through the atmospheric electric field strength in the atmospheric electric field data, and obtain the comprehensive atmospheric electric field strength level through the atmospheric electric field strength and its corresponding field strength jump; determine the second pre-used lightning warning level based on the comprehensive atmospheric electric field strength level and the warning area where the atmospheric electric field instrument is set;

[0018] Determine the lightning warning target level through the first pre-used lightning warning level and the second pre-used lightning warning level.

[0019] Further, determine the location where cloud-to-ground lightning occurs based on the acquired three-dimensional lightning data, and determine the first preliminary lightning warning level based on the warning area where the cloud-to-ground lightning occurs. Specifically:

[0020] Determine the location where cloud-to-ground lightning occurs based on the acquired three-dimensional lightning data;

[0021] Set a preset range corresponding to each cloud-to-ground lightning with the real-time cloud-to-ground lightning location detected by the lightning locator as the center, and acquire the radar data within each preset range;

[0022] Judge whether there is an echo with an echo intensity greater than or equal to the preset value (30 dBZ) within the preset range (within 10 kilometers) centered on the cloud-to-ground lightning location in the radar data. If so, determine the first preliminary lightning warning level based on the warning area where the cloud-to-ground lightning occurs.

[0023] Further, the calculation formula for the field strength jump is:

[0024]

[0025] In the formula, t1 and t2 represent any two adjacent moments detected by the atmospheric electric field instrument, with an interval of 1 s; Δt represents the interval duration between t1 and t2; E represents the atmospheric electric field intensity, represents the atmospheric electric field intensity corresponding to the moment t1, represents the atmospheric electric field intensity corresponding to the moment t2.

[0026] Further, the method for obtaining the comprehensive atmospheric electric field intensity level through the atmospheric electric field intensity and its corresponding field strength jump is specifically: obtain the warning level corresponding to the atmospheric electric field intensity and the warning level corresponding to the field strength jump; calculate the comprehensive atmospheric electric field intensity level according to the warning levels corresponding to the atmospheric electric field intensity and the field strength jump;

[0027] The calculation formula for the comprehensive atmospheric electric field intensity level is:

[0028] G total = W1G E + W2G ΔE ;

[0029] In the formula, W1 and W2 respectively represent the weights of the atmospheric electric field intensity and the field strength jump; G E and G ΔE are respectively the warning levels of the atmospheric electric field intensity and the field strength jump; G total represents the comprehensive atmospheric electric field intensity level.

[0030] Further, the preset method is specifically:

[0031] Extract strong echo grid points with echo intensity greater than or equal to a preset value (40 dBZ) from real-time radar reflectivity CAPPI data, merge spatially adjacent strong echo grid points, and remove isolated points to obtain a strong echo area, i.e., a thunderstorm area; the isolated points are strong echo grid points far from the merged points (strong echo grid points at a preset distance from the nearest merged point).

[0032] Further, the warning area further includes: a warning area and a concern area;

[0033] The distance between the outer edge of the bull's-eye area and the bull's-eye < the distance between the outer edge of the warning area and the bull's-eye < the distance between the outer edge of the concern area and the bull's-eye.

[0034] Further, the range of the bull's-eye area is: the area within a radius of [0, 10] km centered on the bull's-eye; the range of the warning area is: the area within a radius of (10, 30] km centered on the bull's-eye; the range of the concern area is: the area within a radius of (30, 60] km centered on the bull's-eye.

[0035] The present invention also provides a targeted lightning grading warning system based on multi-source detection data, including:

[0036] A warning area module for setting multiple warning areas of different levels with the user's location as the bull's-eye; the higher the level, the closer the outer edge of the corresponding warning area is to the bull's-eye; the warning area includes a bull's-eye area;

[0037] A real-time data processing module for obtaining real-time detection data and determining the lightning warning target level based on the multiple warning areas corresponding to the user and the real-time detection data, and generating lightning level warning information; the real-time detection data includes three-dimensional lightning data and atmospheric electric field data;

[0038] A radar data acquisition module for obtaining real-time radar reflectivity CAPPI data at the target height of the real-time vertical temperature layer; the target height of the vertical temperature layer is the height of the -10°C layer detected by the microwave radiometer;

[0039] A thunderstorm area identification module for obtaining a thunderstorm area by using real-time radar reflectivity CAPPI data through a preset method and continuously updating the thunderstorm area;

[0040] A thunderstorm area extrapolation module for calculating the optical flow vector field corresponding to the current moment in real time through the reflectivity factor data of the radar at the target height at the current moment and the previous moment, obtaining the moving direction and moving speed of the current radar echo through the optical flow vector field, and predicting multiple positions that the thunderstorm area will pass through within a preset time period through the moving direction and moving speed;

[0041] An estimation module, configured to obtain the time when the forefront of the lightning storm area reaches the target area as the estimated impact time when there is a position corresponding to the lightning storm area covering the bull's-eye among the positions passed through;

[0042] A warning information generation module, configured to generate a lightning approaching warning information through the lightning level warning information and / or the estimated impact time.

[0043] Compared with the prior art, the present invention has at least the following beneficial effects:

[0044] (1) The present invention generates lightning level warning information through real-time detection data; obtains the lightning storm area by using the real-time radar reflectivity CAPPI data obtained based on the real-time vertical temperature stratification data through a preset method, and continuously updates the lightning storm area; calculates the optical flow vector field corresponding to the current moment; predicts multiple positions that the lightning storm area corresponding to the current moment will pass through within a preset time period through the optical flow vector field; determines whether there is a position corresponding to the lightning storm area covering the bull's-eye among the positions passed through, and if so, obtains the time when the forefront of the lightning storm area reaches the target area as the estimated impact time; generates a lightning approaching warning information through the lightning level warning information and / or the estimated impact time; that is, the present invention provides a progressive lightning warning based on the user's position, realizes targeted and customizable warning services, and improves the lightning monitoring and warning ability;

[0045] (2) In the present invention, the height of the temperature layer corresponding to a specific temperature in the real-time vertical temperature stratification data is obtained as the target height, and the real-time radar reflectivity CAPPI data at the target height is obtained; the lightning storm area is obtained by using the real-time radar reflectivity CAPPI data; the optical flow vector field corresponding to the current moment is calculated in real time through the reflectivity factor data of the radar at the target height at the current moment and the previous moment; the lightning storm area is extrapolated for a preset time period through the moving direction and moving speed of the optical flow vector field; that is, the present invention improves the accuracy of extrapolation of the lightning storm area by fusing the real-time detected vertical temperature stratification data, realizes precise lightning warning for a specific area or user, and solves the problem that customized warnings cannot be effectively issued in the prior art. Description of the Drawings

[0046] Figure 1 It is a flow chart of a targeted lightning classification warning method based on multi-source detection data;

[0047] Figure 2 It is a screenshot of a case review of a user's lightning yellow warning;

[0048] Figure 3 It is a screenshot of a case review of a user's lightning orange warning;

[0049] Figure 4It is a module diagram of a targeted lightning grading early warning system based on multi-source detection data. Detailed implementation manners

[0050] The following are specific embodiments of the present invention and, in combination with the accompanying drawings, further describe the technical solutions of the present invention, but the present invention is not limited to these embodiments.

[0051] Embodiment 1

[0052] In order to provide customized targeted lightning nowcasting grading early warning for users in a specific area, concisely inform users of the estimated storm impact time, and meet users' needs for refined lightning short-term nowcasting early warning, as Figure 1 shown, the present invention proposes a targeted lightning grading early warning method based on multi-source detection data, including:

[0053] Taking the location of the user as the bull's-eye, setting multiple warning areas with different levels; the higher the level, the closer the outer edge of the corresponding warning area is to the bull's-eye; the warning area includes the bull's-eye area (i.e., this area);

[0054] The warning area also includes: a warning area and a concerned area;

[0055] The distance between the outer edge of the bull's-eye area and the bull's-eye < the distance between the outer edge of the warning area and the bull's-eye < the distance between the outer edge of the concerned area and the bull's-eye.

[0056] The range of the bull's-eye area is: the area within a radius of [0, 10] km centered on the bull's-eye; the range of the warning area is: the area within a radius of (10, 30] km centered on the bull's-eye; the range of the concerned area is: the area within a radius of (30, 60] km centered on the bull's-eye.

[0057] In this embodiment, the range of the warning area is set according to the speed of thunderstorm movement (generally 10 - 60 km / h).

[0058] Obtain real-time detection data, and determine the lightning warning target level based on the multiple warning areas corresponding to the user and the real-time detection data, and generate lightning level warning information; the real-time detection data includes three-dimensional lightning data and atmospheric electric field data;

[0059] Obtain real-time detection data, and determine the lightning warning target level based on the multiple warning areas corresponding to the user and the real-time detection data; specifically:

[0060] Obtain the three-dimensional lightning data detected in real time by multiple lightning locators in the DDW1 type lightning location system, determine the location of cloud-to-ground lightning based on the obtained three-dimensional lightning data, and determine the first preliminary lightning warning level based on the warning area where the cloud-to-ground lightning occurs;

[0061] In this embodiment, a DDW1 type lightning location system is adopted. This system comprehensively uses the multi-station time difference location method and the magnetic orientation method to detect three-dimensional lightning data, and the positioning error (median) is about 450 m. In addition, the present invention uses three-dimensional lightning data detected by three or more lightning locators to determine the occurrence location of cloud-to-ground lightning, and obtains relatively accurate positioning data.

[0062] Based on the acquired three-dimensional lightning data, determine the occurrence location of cloud-to-ground lightning, and based on the warning area where the cloud-to-ground lightning occurs, determine the first pre-used lightning warning level, specifically:

[0063] Based on the acquired three-dimensional lightning data, determine the occurrence location of cloud-to-ground lightning;

[0064] Set a preset range corresponding to each cloud-to-ground lightning with the real-time cloud-to-ground lightning location detected by the lightning locator as the center, and obtain the radar data within each preset range;

[0065] Judge whether there is an echo with an echo intensity greater than or equal to the preset value (30 dBZ) within the preset range (within a radius of 10 km) centered on the cloud-to-ground lightning occurrence location in the radar data. If so, determine the first pre-used lightning warning level based on the warning area where the cloud-to-ground lightning occurs.

[0066] In this embodiment, the lightning warning levels are distinguished according to the occurrence location of the actual cloud-to-ground lightning (including a total of 3 levels: Level 1 - Yellow Warning, Level 2 - Orange Warning, Level 3 - Red Warning). When the actual cloud-to-ground lightning occurs in the concerned area, a yellow lightning warning is triggered; when it occurs in the warning area, an orange warning is triggered; when it occurs in this area, a red warning is triggered. At the same time, considering that there may be a certain detection error in the lightning locator, which is prone to false alarms, therefore, during the process of determining the first pre-used lightning warning level, radar echo is introduced as a quality control standard, and it is judged whether there is an echo with an echo intensity reaching above 30 dBZ within a radius of 10 km centered on the cloud-to-ground lightning occurrence location, so as to improve the reliability of the warning. For example, when the actual cloud-to-ground lightning occurs in this area and there is an echo with an echo intensity reaching above 30 dBZ within a radius of 10 km centered on the cloud-to-ground lightning occurrence location, the red warning is set as the first pre-used lightning warning level.

[0067] Real-time atmospheric electric field data is obtained through an atmospheric electric field meter of the Pre-storm2.0 model, the field strength jump is calculated through the atmospheric electric field strength in the atmospheric electric field data, and the comprehensive atmospheric electric field strength level is obtained through the atmospheric electric field strength and its corresponding field strength jump; based on the comprehensive atmospheric electric field strength level and the warning area where the atmospheric electric field meter is set, determine the second pre-used lightning warning level;

[0068] The calculation formula for the field strength jump is:

[0069]

[0070] Wherein, t1 and t2 represent any two adjacent moments detected by the atmospheric electric field instrument, with an interval of 1 s; Δt represents the interval duration between t1 and t2; E represents the atmospheric electric field intensity, represents the atmospheric electric field intensity corresponding to the moment t1, represents the atmospheric electric field intensity corresponding to the moment t2.

[0071] The method for obtaining the comprehensive intensity level of the atmospheric electric field through the atmospheric electric field intensity and its corresponding field strength jump is as follows: obtaining the warning level corresponding to the atmospheric electric field intensity and the warning level corresponding to the field strength jump; calculating the comprehensive intensity level of the atmospheric electric field according to the warning levels corresponding to the atmospheric electric field intensity and the field strength jump;

[0072] It should be noted that the warning levels corresponding to the atmospheric electric field intensity and the field strength jump are given in Table 1 below:

[0073]

[0074] In Table 1, the grade thresholds of the absolute value E of the atmospheric electric field intensity are respectively taken as: 1 kV / m, 3 kV / m, 6 kV / m, and 9 kV / m according to the data analysis of 57 thunderstorm weather processes in 2019 and the recommended thresholds of the atmospheric electric field instrument manufacturer.

[0075] The grade thresholds of the field strength jump (differential) ΔE are respectively taken as: 0.1 kV / m, 0.5 kV / m, 1 kV / m, and 2 kV / m according to the analysis of the differential atmospheric electric field in sunny days and the differential atmospheric electric field in the above thunderstorm weather processes. The calculations of the above atmospheric electric field intensity and its differential are all in minutes. Calculate the maximum atmospheric electric field intensity and differential within each minute to participate in the division of the threshold levels and assign weights, as the basis for the atmospheric electric field instrument to trigger the lightning approaching warning. It should be noted that the above grade thresholds need to be adjusted accordingly during application due to different manufacturers and different locations.

[0076] The comprehensive intensity level of the atmospheric electric field is calculated by weighting the maximum atmospheric electric field intensity and the maximum field strength jump warning level within each minute: The calculation formula for the comprehensive intensity level of the atmospheric electric field is:

[0077] G total = W1G E + W2G ΔE ;

[0078] Wherein, W1 and W2 respectively represent the weights of the atmospheric electric field intensity and the field strength jump, both taking the value of 0.5; G E and G ΔE are respectively the warning levels of the atmospheric electric field intensity and the field strength jump; G total represents the comprehensive intensity level of the atmospheric electric field.

[0079] Comprehensively calculate the comprehensive intensity level of the atmospheric electric field and convert it into the corresponding lightning warning level, as shown in Table 2. In Table 2 are the levels corresponding to the data of the atmospheric electric field instrument in this area. If it is the value detected by the atmospheric electric field instrument in the warning area, the warning levels of the electric field intensity and the field intensity jump will be correspondingly reduced by one level to participate in the calculation of the comprehensive intensity level of the atmospheric electric field. Specifically:

[0080] 1. The levels corresponding to the data of the atmospheric electric field instrument in this area: This means that when the atmospheric electric field instrument is deployed within the user-defined this area (within 0 - 10 km from the user center point), the directly detected electric field intensity and the field intensity jump levels will be directly used for calculating the lightning warning level according to the classification in Table 1. For example, if the detected electric field intensity and the field intensity jump levels meet a certain level in Table 1, then it will directly correspond to the corresponding color warning (yellow, orange, red) in the lightning warning level Table 2.

[0081] 2. The values detected by the atmospheric electric field instrument in the warning area: If the atmospheric electric field instrument is located in the warning area (within 10 - 30 km from the user center point), then the detected electric field intensity and the field intensity jump levels need to be reduced by one level each when calculating the total level. In other words, the data that might reach a certain warning level in this area can only be counted as a lower warning level in the warning area. This is because the warning area is relatively farther from this area and poses a slightly smaller direct threat to the user, so the warning level needs to be adjusted accordingly.

[0082] 3. By analogy for the data of the atmospheric electric field instrument in the concerned area: For the detected data of the atmospheric electric field instrument located in the concerned area (within 30 - 60 km from the user center point), the electric field intensity and the field intensity jump levels also need to be further reduced when calculating the total level, one level lower than that in the warning area. This reflects that as the distance increases, the potential risk to the user gradually decreases, and the warning level is reduced accordingly to more reasonably reflect the actual risk level.

[0083] In summary, this dynamic adjustment mechanism ensures the pertinence and practicality of lightning warnings, and gives the user the most appropriate warning information according to the actual risk levels in different areas.

[0084] In addition, considering that the atmospheric electric field instrument is extremely sensitive to the external environment and may be interfered by the surrounding environment, weather, season, altitude and other different factors of the instrument layout area, and occasionally produce detection outliers. Therefore, in the application, the same radar echo quality control method as that of the lightning location instrument is adopted: comparing the detection results of the atmospheric electric field instrument with the radar echo data at the same moment. Radar echoes can provide the position, intensity and dynamic changes of thunderstorm clouds. If the radar shows weak or no thunderstorm activity, while the atmospheric electric field instrument records a high-intensity electric field, this may be an abnormal detection. Then, the data consistency of multiple electric field instruments can be analyzed. If the readings of a certain instrument are significantly different from those of other instruments, the data of this instrument may be locally interfered or there may be a fault, and further investigation is needed. Through this cross-verification, the accuracy of early warning can be improved.

[0085]

[0086] In this embodiment, assuming that the calculated value of the comprehensive intensity level of the atmospheric electric field is 2, then the second preliminary lightning warning level is a yellow warning.

[0087] Determine the lightning warning target level through the first preliminary lightning warning level and the second preliminary lightning warning level.

[0088] To sum up, a targeted lightning classification early warning is constructed based on the lightning location instrument and the atmospheric electric field instrument. The former has a detection accuracy of milliseconds, while the latter calculates the comprehensive intensity level of the atmospheric electric field once every minute. If the early warnings triggered by the two instruments are of the same level, then the first preliminary lightning warning level triggered by the lightning location instrument is taken as the lightning warning target level, and the corresponding lightning level early warning information is generated. If the two instruments trigger early warnings of different levels, the higher level is given priority. For example, if the level triggered by the atmospheric electric field instrument is higher than that of the lightning location instrument, then the second preliminary lightning warning level is set as the lightning warning target level, and the corresponding lightning level early warning information is generated.

[0089] Obtain the real-time vertical temperature stratification data through the microwave radiometer, obtain the height of the temperature layer corresponding to a specific temperature (-10 °C) in the real-time vertical temperature stratification data as the target height, and obtain the real-time radar reflectivity CAPPI data at the target height;

[0090] It should be noted that in the application of meteorological radar, the -10°C altitude is an important reference level because near this altitude, ice crystals and supercooled water droplets coexist and are often associated with the development of severe thunderstorms. Currently, there are two types of microwave radiometers in Ningbo. One is the RPG-HATPRO-G5 produced in Germany (one unit), and the other is the domestic KYW1 (two units). Both instruments can continuously monitor in real time information such as the temperature, humidity, liquid water profile, and total amount in the atmospheric boundary layer and troposphere within the vertical range of 0 to 10 km. To obtain the data of the -10°C layer in real time, in the application, the average of the -10°C altitude layer data detected by the three microwave radiometers in the Ningbo area is taken.

[0091] Use the real-time radar reflectivity CAPPI data through a preset method to obtain the thunderstorm area and continuously update the thunderstorm area;

[0092] The specific preset method is as follows:

[0093] Extract the strong echo grid points with an echo intensity greater than or equal to the preset value (40 dBZ) in the real-time radar reflectivity CAPPI data, merge the spatially adjacent strong echo grid points, and eliminate the isolated points to obtain the strong echo area, that is, the thunderstorm area; the isolated points are the strong echo grid points far from the merged points (the strong echo grid points at a preset distance from the nearest merged point).

[0094] It should be explained that 40 dBZ is a specific reflectivity factor value, which is often used to indicate strong convective activities.

[0095] Calculate the optical flow vector field corresponding to the current moment in real time through the reflectivity factor data of the radar at the target altitude at the current moment and the previous moment. Obtain the moving direction and moving speed of the current radar echo through the optical flow vector field, and predict multiple positions (10 positions in this embodiment, one every 6 minutes) that the thunderstorm area at the current moment will pass through within a preset time period (within 1 hour);

[0096] Specifically, in this embodiment, the optical flow method is used to judge the moving trend of the thunderstorm area in the next 1 hour and estimate the time when the storm affects the location of the target center. The optical flow extrapolation method is a motion estimation technology based on image sequences, which predicts the change trend of the radar echo in the next period of time by analyzing the motion between two consecutive frames or multiple frames of images. First, use the reflectivity factor data of the radar at the -10°C altitude layer at the current moment and the previous moment to calculate the optical flow vector field and obtain the moving direction and moving speed of the current echo; then use the optical flow vector field to extrapolate the thunderstorm area at the -10°C layer altitude at the current moment for 1 hour.

[0097] Determine whether there is a lightning storm area corresponding to a certain position among the passed positions that covers the bull's-eye. If so, obtain the time when the forefront of the lightning storm area reaches the bull's-eye area as the estimated impact time;

[0098] It should be noted that by using the radar data at the current moment and the next new moment to calculate the optical flow vector field again, and so on, the optical flow vector field is continuously updated in this way so that the system can track the latest position of the lightning storm area in real time.

[0099] In this embodiment, the calculation of the optical flow vector field is explained as follows:

[0100] The optical flow vector field is a vector field that characterizes the velocity of a rigid-like object. The velocity vector is generally represented by two components, u and v, in the meteorological field; among them, u represents the component of the optical flow in the horizontal direction (X-axis), that is, the moving speed of the pixel points in the image along the horizontal direction; v represents the component of the optical flow in the vertical direction (Y-axis), corresponding to the moving speed of the pixel points along the vertical direction. (u, v) not only represents the moving speed but also implies the moving direction; by analyzing these vectors in the optical flow field, the change trend of each point in the image over time can be inferred, and thus the motion information in the scene can be understood;

[0101] Suppose there is a point (x, y) on the Cartesian coordinate plane, which represents the projection of a certain point (x, y, z) in three-dimensional space on the image plane. The gray value (the numerical value of the regional reflectivity factor representing 40 dBZ at the -10°C height in the present invention) of this point at time t is I(x, y, t), where:

[0102] I represents the intensity value of the image, that is, the above-mentioned gray value;

[0103] x and y are spatial coordinates, representing the pixel positions in the image, that is, a certain point in two-dimensional space;

[0104] t is the time coordinate, indicating that this gray value is measured or calculated at time t, which is applicable to dynamic or time series data, such as a sequence of radar images;

[0105] The numerical value of the regional reflectivity factor representing 40 dBZ at the -10°C height: It is used to express the special meaning of the gray value. Through the gray value I(x, y, t), it intuitively shows that at a certain time t, at the coordinate position (x, y), it corresponds to the area in the atmosphere with a reflectivity factor of 40 dBZ at the -10°C height, that is, the area of strong meteorological activities;

[0106] Assume that this point moves to (x + Δx, y + Δy) at (t + Δt), and the gray value remains unchanged within a certain time interval Δt, that is:

[0107] I(x + uΔt, y + vΔt, t + Δt) = I(x, y, t) (1)

[0108] In the formula, u and v are the components of the optical flow of this point in the x and y directions respectively; calculating according to the Taylor expansion formula, the optical flow constraint equation can be obtained:

[0109] I x u + I y v + I t = 0 (2)

[0110] In the formula,

[0111] Let be the spatial gradient of the grayscale of the image point; I t is the change rate of the grayscale with time; (u, v) is called the optical flow, and the optical flow of all points in the image constitutes the optical flow vector field; represents the small change rate of the image in the horizontal direction, that is, the partial derivative of the grayscale value of each pixel point in the image with respect to the x coordinate (abscissa), which reflects the edge or change information of the image along the x-axis; then represents the small change rate of the image in the vertical direction, that is, the partial derivative of the grayscale value of each pixel point in the image with respect to the y coordinate (ordinate); it reflects the edge or change information of the image along the y-axis.

[0112] Since there are two unknown variables u and v in the optical flow, and equation (2) has only one optical flow constraint equation, further constraint conditions need to be introduced to completely determine the optical flow (u, v). The present invention adopts the Lucas-Kanade local constraint method as the constraint condition for calculating the optical flow:

[0113] Assume that the optical flow of N points in a sufficiently small area centered on P is the same, and within a sufficiently short time interval, the movement between two frames of images can be approximately regarded as linear. Different points in the area are given different weights, and the closer to point P, the higher the weight. Then the calculation of the optical flow can be transformed into:

[0114]

[0115] In formula (3), θ is a small area centered on P; x represents the points in area θ; W 2 (x) is the window function, representing the weights of each point in the area; V = (u, v) T is the optical flow of point P; is the gradient vector of I at position x and time t, representing the change rate of the image brightness in space; I t (x, t) is the partial derivative of I with respect to time t, that is, the change rate of the image with time; Let:

[0116]

[0117] W = diag(W(x1), W(x2), …, W(x n ));

[0118] b = -(I t (x1), I t (x2), …, I t (x n )); T ;

[0119] n represents the total number of points within the region θ;

[0120] Then the solution of the equation is:

[0121] V = (A T W 2 A) -1 A T W 2 b, and A T W 2 A is actually a 2×2 matrix:

[0122]

[0123] Although the radar echo reflectivity between two adjacent time instances changes with time, within the 6 - minute time interval between two radar volume scans, this change can be considered relatively weak. Therefore, it can be approximately considered that the radar image satisfies the optical flow constraint equation (2). In the algorithm, first - order differences are used to replace the derivatives of the reflectivity in each direction

[0124]

[0125] In the formula, i and j are spatial coordinates, corresponding to the columns (horizontal positions) and rows (vertical positions) of the image respectively, similar to the pixel positions in a two - dimensional image; t is the time coordinate, representing the frame number or time point, which makes I an image sequence that changes with time; I i,j,t represents the intensity or gray - scale value of the pixel located at the image coordinates (i, j) at time t.

[0126] According to the above formula, the optical flow vector field can be calculated. From the above formula, it can be seen that when calculating the optical flow vector field, both the change of consecutive - time - instance echoes and the change of adjacent - position echoes are considered.

[0127] Generate lightning nowcasting warning information based on lightning level warning information and / or estimated impact time and send this information to users.

[0128] Lightning level warning information, examples are as follows:

[0129] Example 1: "Thunder and lightning yellow warning: At xx:xx on xx / xx, the atmospheric electric field intensity detected by the atmospheric electric field instrument 15.2 KM in the northwest direction of 'xx user' in the past 1 minute is 7 kV / m, the field strength jump is 0.8 kV / m, and the comprehensive intensity level of the atmospheric electric field is yellow. Please pay attention."

[0130] Example 2: "Thunder and lightning orange warning: At xx:xx on xx / xx, there have been xx ground flashes in the warning area of 'xx user' in the past 1 hour. The nearest ground flash is located 25.3 KM in the west-southwest direction. There may be thunder and lightning activities in this area. Please pay attention."

[0131] The above thunder and lightning approaching warning information mainly covers targeted thunder and lightning grading warnings (thunder and lightning level warning information) centered on each user point and the estimated impact time. If it is estimated that the thunderstorm area will not affect the user's location within 1 hour, the thunder and lightning approaching warning information only includes the targeted thunder and lightning grading warning centered on each user point, that is, the thunder and lightning level warning information. The thunder and lightning approaching warning information is released using a reasonable release strategy, obtaining detection data in real time, triggering the first warning based on the warning area where the ground flash occurs or the detection result of the atmospheric electric field instrument, and at the same time, rolling and updating the thunder and lightning level warning information in real time through the real-time detection data. When the target level of the thunder and lightning warning triggered in real time is upgraded or downgraded compared with the previous time, the thunder and lightning level warning information is updated; when the target level of the thunder and lightning warning triggered in real time has no level change compared with the previous time, the thunder and lightning warning target level is not updated (after the thunder and lightning level warning information is updated, a new thunder and lightning approaching warning information is sent to the user again); when the targeted thunder and lightning approaching warning standard is not met (that is, the comprehensive intensity of the atmospheric electric field is at a non-alarming warning level, and the actual ground flash does not occur in the bull's-eye area, warning area, and attention area), and no thunderstorm is identified within the user's attention area, warning area, and this area, and the thunderstorm will not affect the above three areas within the next hour, the thunder and lightning approaching warning is lifted.

[0132] The present invention generates thunder and lightning level warning information through real-time detection data; obtains the thunderstorm area by using the real-time radar reflectivity CAPPI data obtained based on the real-time vertical temperature stratification data through a preset method, and continuously updates the thunderstorm area; calculates the optical flow vector field corresponding to the current moment; predicts multiple positions that the thunderstorm area at the current moment will pass through within a preset time through the optical flow vector field; determines whether there is a position corresponding to the thunderstorm area covering the bull's-eye among the passed positions. If so, the time when the forefront of the thunderstorm area reaches the bull's-eye area is obtained as the estimated impact time; generates thunder and lightning approaching warning information through the thunder and lightning level warning information and / or the estimated impact time; that is, the present invention provides a progressive thunder and lightning warning based on the user's location, realizes targeted and customizable warning services, and improves the warning ability of thunder and lightning monitoring.

[0133] Example 2

[0134] As Figure 4 shown, this embodiment also proposes a targeted lightning grading early warning system CTL based on multi-source detection data, including:

[0135] An early warning area module, used to set multiple early warning areas with different levels taking the user's location as the center; the higher the level, the closer the outer edge of the corresponding early warning area is to the center; the early warning area includes the center area;

[0136] A real-time data processing module, used to obtain real-time detection data, determine the lightning early warning target level based on the multiple early warning areas corresponding to the user and the real-time detection data, and generate lightning level early warning information; the real-time detection data includes three-dimensional lightning data and atmospheric electric field data;

[0137] A radar data acquisition module, used to obtain real-time radar reflectivity CAPPI data at the target height of the real-time vertical temperature layer; the target height of the vertical temperature layer is the height of the -10°C layer detected by the microwave radiometer;

[0138] A lightning storm area identification module, used to obtain the lightning storm area by using the real-time radar reflectivity CAPPI data through a preset method and continuously update the lightning storm area;

[0139] A lightning storm area extrapolation module, used to calculate the optical flow vector field corresponding to the current moment in real time through the reflectivity factor data of the radar at the target height at the current moment and the previous moment, obtain the moving direction and moving speed of the current radar echo through the optical flow vector field, and predict multiple positions that the lightning storm area at the current moment will pass through within a preset time period through the moving direction and moving speed;

[0140] An estimation module, used to obtain the time when the forefront of the lightning storm area reaches the center area as the estimated impact time when a certain position among the passed positions has the lightning storm area covering the center;

[0141] An early warning information generation module, used to generate lightning approaching early warning information through the lightning level early warning information and / or the estimated impact time.

[0142] Example 3

[0143] In order to further reflect the beneficial effects of the present invention, this embodiment gives the evaluation results of the targeted lightning grading early warning and the lightning storm area extrapolation algorithm, and provides the application situation and analysis description of the targeted lightning grading early warning method of the present invention in typical cases:

[0144] Now, the system lightning grading early warning inspection and evaluation of key users from 2021 to 2023 is carried out by using the inspection method mentioned in the "Medium- and Short-Term Weather Forecast Quality Inspection Measures" of the China Meteorological Administration.

[0145] TS score:

[0146] False negative rate:

[0147] False alarm rate:

[0148] Wherein, NA is the number of correct forecasts (warnings), NB is the number of false alarms in the forecasts (warnings), and NC is the number of missed forecasts in the forecasts (warnings). Among them, each time a lightning approaching warning message is sent to the user, it represents one warning.

[0149] In this embodiment, an evaluation is carried out on six key users to whom the method of the present invention is applied. Among them, users 1 to 4 have complete annual system usage data from 2021 to 2023, and users 5 and 6 only have complete annual system usage data in 2023. As can be seen from Table 3, the TS scores of the system warnings for the six key users generally exceed 80%. The average TS score of the users is 86.82%, slightly better than the evaluation results of relevant systems and methods in provinces such as Sichuan and Zhejiang before. The false alarm rate of the users is 7%-17%. This is mainly because there are still detection errors in the current lightning locator used. Therefore, the accuracy of the lightning warning triggered based on the three-dimensional lightning data is affected to a certain extent. At the same time, from the evaluation and inspection results, there is no missed warning phenomenon for five users, and only one user has missed one thunderstorm process. This process is short in time and occurs at night. The radar echo data is briefly missing during the thunderstorm impact period, and the targeted lightning grading warning needs to go through the quality control of the radar echo before triggering. Therefore, the system did not trigger a warning during this process.

[0150]

[0151]

[0152] To evaluate the credibility of the lightning storm area extrapolation algorithm, the S-band dual-polarization Doppler weather radar data in Ningbo is used to conduct an analysis of 17 thunderstorm weather processes from 2021 to 2023, and the hit rate of the extrapolation algorithm is statistically analyzed (see Table 4 for details).

[0153] Hit rate:

[0154] Wherein NA is the number of correct forecasts, that is, the cloud-to-ground flash appears in the lightning storm area extrapolated by the optical flow method for 1 hour, and NC is the number of missed forecasts, that is, the cloud-to-ground flash appears outside the lightning storm area extrapolated by the optical flow method for 1 hour.

[0155]

[0156]

[0157] The evaluation of the extrapolation algorithm for 1-hour thunderstorm cloud clusters found that the hit rate of individual cases ranged from 24.78% to 60.46%, and the average hit rate of 17 individual cases was 41.14%, which was better than the hit rate of lightning extrapolation forecasts mentioned in previous relevant literature. The hit rate in the relevant literature (Zhou Kanghui, Zheng Yongguang, Lan Yu. Thunderstorm Identification, Tracking and Extrapolation Method Based on Lightning Data [J]. Journal of Applied Meteorology, 2016, 27(2): 173-181.) ranged from 18% to 51%, with an average value of 34.50%.

[0158] From 2021 to 2023, the Ningbo Meteorological Service Center used the targeted lightning classification warning method (or the targeted lightning classification warning system constructed based on the targeted lightning classification warning method) to carry out services for hundreds of thunderstorm weather processes. Now, a case analysis is made on the application of two typical lightning weather processes during the business operation period.

[0159] Case 1 (such as Figure 2As shown in the figure: Affected by the upper-air shallow trough at night on July 25, 2022, the north-south zonal echo moved from west to east and affected Ningbo. Obvious lightning weather processes occurred in most areas, accompanied by thunderstorm gales and short-term heavy precipitation. The number of cloud-to-ground lightning strikes in the whole city exceeded 1,000 times. Among them, the strongest impact period was from 20:00 to 21:00, and the hourly cloud-to-ground lightning count in the whole city reached 592 times, accounting for 47.06% of the total cloud-to-ground lightning strikes on that day. At night on that day, the hourly maximum wind force reached 8-10 levels at 45 stations, and the hourly rainfall intensity exceeded 50 mm at 3 stations. The "XX Daxie" user of the targeted lightning grading warning system is located on Daxie Island in Beilun District. At 19:33 on that day, the first cloud-to-ground lightning strike entered the attention area, and the system issued a yellow lightning warning. At 20:26, the system judged based on the lightning storm area identification and extrapolation results that the thunderstorm in the northwest by west direction would gradually move eastward and affect the factory area after 24 minutes. Therefore, the following comprehensive lightning approaching warning information was given: "At 20:26 on July 25, 29 lightning strikes have occurred in the 'XX Daxie' attention area in the past 1 hour! The nearest lightning strike is 39.7 KM in the west direction. It is expected to affect this area after 24 minutes." The actual data shows that at 20:59, the lightning storm moved directly above the factory area. At this moment, the nearest cloud-to-ground lightning strike was only 0.7 km away from the factory area. At the same time, it was noted that at 21:03, the instantaneous maximum wind speed of 17.4 m / s (level 8) was recorded at the Tunizui Station, a meteorological station near the target center where the factory area is located. In addition, a key container terminal in the Ningbo-Zhoushan Port is located on Meishan Island on the south side of the Beilun Peninsula. At 20:32 on that day, the system issued an orange lightning warning signal for the "Meishan Port Area" user and predicted that the storm would affect after 49 minutes. "At 20:32 on July 25, 1 lightning strike has occurred in the 'Meishan Port Area' warning area in the past 1 hour! The nearest lightning strike is 18.6 KM in the northwest by north direction. It is expected that there may be lightning activities after 49 minutes." The actual data shows that three cloud-to-ground lightning strikes occurred in this area of the user between 21:22 and 21:32, and the nearest cloud-to-ground lightning strike was 6.9 km away from the target center position. Generally speaking, with the formation and eastward movement and southward pressure of the storm, the CTL system accurately identified the lightning storm and predicted its movement and impact trend, providing effective approaching service products for users.

[0160] Case 2 (such as Figure 3As shown): In the afternoon of July 26, 2022, a multi-cell storm formed near Fenghua in the southwest of Ningbo, and then gradually moved eastward, with a slight weakening in intensity. After moving to the vicinity of the main urban area, three echoes merged to form a bow-shaped echo, which then significantly strengthened and quickly moved eastward, forming a mid-level convective storm with vertical wind shear. This storm brought significant lightning and thunderstorm gale processes that were rare in previous years to Ningbo Zhoushan Port. Ground lightning occurred mainly between 16:00 and 17:00 on that day, with a frequency of 407 ground lightnings per hour, accounting for 80.12% of the total number of ground lightnings that occurred on that day. Among them, the number of ground lightnings near Ningbo Zhoushan Port in the evening exceeded 30 times, and the instantaneous wind speed at Daxie Second Bridge Station reached 35.0m / s (level 12) at 18:02, the largest among all meteorological stations in the city on that day. The "XX Daxie" user of the CTL system was located on Daxie Island. At 15:27 on the same day, ground lightning entered the attention area. The system immediately activated the first yellow thunderstorm warning, reminding users to pay attention to changes in real time. "At 15:27 on July 26, "XX Daxie" had lightning strike once in the attention area in the past hour! The nearest lightning was located 56.2KM southwest, please pay attention." As the thunderstorm system moved eastward, from 17:40 to 18:13, high-level orange and red warnings were continuously issued, totaling 10 times. It is worth noting that at 17:40 on the same day, the system not only updated the yellow warning signal of lightning to the orange warning signal of lightning, but also identified the lightning storm area to the west of the user, and used the optical flow method to extrapolate the time of affecting the user in 18 minutes. Therefore, at 17:40, the following warning information combining targeted graded warning (lightning level warning information) and extrapolation results (estimated impact time) was given: "At 17:40 on July 26, 272 lightning strikes occurred in the warning area of 'XX Daxie' in the past hour! The latest lightning was located 20.4KM southwest, and it is expected to affect this area in the next 18 minutes." The actual data showed that at 17:56, the lightning storm affected the user's location, and ground lightning occurred 2.5km away from the factory area. The extrapolated forecast results of the lightning storm were consistent with the actual situation.

[0161] It should be noted that all directional indications in the embodiments of the present invention (such as up, down, left, right, front, back, etc.) are only used to explain the relative position relationship, movement status, etc. between the components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication will also change accordingly.

[0162] In addition, in the present invention, descriptions such as "first", "second", "one", etc. are only used for descriptive purposes and cannot be understood as indicating or implying their relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" or "second" may explicitly or implicitly include at least one of the features. In the description of the present invention, the meaning of "plurality" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0163] In the present invention, unless otherwise clearly specified or limited, terms such as "connection" and "fixation" shall be understood in a broad sense. For example, "fixation" may be a fixed connection, a detachable connection, or integrated; it may be a mechanical connection or an electrical connection; it may be directly connected or indirectly connected through an intermediate medium, and it may be the internal communication of two components or the interaction relationship between two components, unless otherwise clearly limited. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0164] In addition, the technical solutions between various embodiments of the present invention can be combined with each other, but it must be based on the ability of those of ordinary skill in the art to implement. When the combination of technical solutions results in contradictions or cannot be implemented, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

Claims

1. A targeted lightning grading warning method based on multi-source detection data, characterized in that Including: Taking the location of the user as the bull's-eye, setting multiple warning areas of different levels; The higher the level, the closer the outer edge of the corresponding warning area is to the bull's-eye; the warning area includes the bull's-eye area; Obtaining real-time detection data, and determining the lightning warning target level based on the multiple warning areas corresponding to the user and the real-time detection data, and generating lightning level warning information; The real-time detection data includes three-dimensional lightning data and atmospheric electric field data; Obtaining real-time vertical temperature stratification data, taking the height of the temperature layer corresponding to a specific temperature in the real-time vertical temperature stratification data as the target height, and obtaining the real-time radar reflectivity CAPPI data at the target height; Obtaining the lightning storm area by using the real-time radar reflectivity CAPPI data through a preset method, and continuously updating the lightning storm area; Calculating the optical flow vector field corresponding to the current moment in real time through the reflectivity factor data of the radar at the target height at the current moment and the previous moment, obtaining the moving direction and moving speed of the current radar echo through the optical flow vector field, and predicting the multiple positions that the lightning storm area will pass through within a preset time period at the current moment; Judging whether there is a position among the passed positions where the corresponding lightning storm area covers the bull's-eye. If so, obtaining the time when the forefront of the lightning storm area reaches the bull's-eye area as the estimated impact time; Generating lightning approaching warning information through the lightning level warning information and / or the estimated impact time.

2. The targeted lightning classification and early warning method based on multi-source detection data according to claim 1, wherein Obtaining real-time detection data, and determining the lightning warning target level based on the multiple warning areas corresponding to the user and the real-time detection data; specifically: Obtaining the three-dimensional lightning data detected in real time by multiple lightning locators in the lightning location system, determining the location of the cloud-to-ground flash based on the obtained three-dimensional lightning data, and determining the first preliminary lightning warning level based on the warning area where the cloud-to-ground flash occurs; Obtaining the atmospheric electric field data in real time through an atmospheric electric field instrument, calculating the field strength jump through the atmospheric electric field strength in the atmospheric electric field data, and obtaining the comprehensive atmospheric electric field intensity level through the atmospheric electric field strength and its corresponding field strength jump; Determining the second preliminary lightning warning level based on the comprehensive atmospheric electric field intensity level and the warning area where the atmospheric electric field instrument is set; Determining the lightning warning target level through the first preliminary lightning warning level and the second preliminary lightning warning level.

3. A targeted lightning classification and early warning method based on multi-source detection data according to claim 2, characterized in that, Determining the location of the cloud-to-ground flash based on the obtained three-dimensional lightning data, and determining the first preliminary lightning warning level based on the warning area where the obtained location is located; specifically: Determining the location of the cloud-to-ground flash based on the obtained three-dimensional lightning data; Setting a preset range corresponding to each cloud-to-ground flash with the real-time cloud-to-ground flash position detected by the lightning locator as the center, and obtaining the radar data within each preset range; Judging whether there is an echo with an echo intensity greater than or equal to the preset value within the preset range centered on the cloud-to-ground flash position in the radar data. If so, determining the first preliminary lightning warning level based on the warning area where the cloud-to-ground flash occurs.

4. The targeted lightning classification and early warning method based on multi-source detection data according to claim 2, wherein The calculation formula of the field strength jump is: Wherein, t1 and t2 represent any two adjacent moments detected by the atmospheric electric field instrument, with an interval of 1 s; Δt represents the interval duration between t1 and t2; E represents the atmospheric electric field strength, represents the atmospheric electric field strength corresponding to the moment t1, represents the atmospheric electric field strength corresponding to the moment t2.

5. The targeted lightning grading warning method based on multi-source detection data according to claim 4, characterized in that The method for obtaining the comprehensive intensity level of the atmospheric electric field based on the atmospheric electric field intensity and its corresponding field strength jump is as follows: obtain the warning level corresponding to the atmospheric electric field intensity and the warning level corresponding to the field strength jump; calculate the comprehensive intensity level of the atmospheric electric field according to the warning levels corresponding to the atmospheric electric field intensity and the field strength jump. The calculation formula for the comprehensive intensity level of the atmospheric electric field is: G total = W1G E + W2G ΔE ; where W1 and W2 respectively represent the weights of the atmospheric electric field intensity and the field intensity jump; G E and G ΔE are respectively the warning levels of the atmospheric electric field intensity and the field intensity jump; G total represents the comprehensive intensity level of the atmospheric electric field.

6. The targeted lightning classification and early warning method based on multi-source detection data according to claim 5, wherein The preset method is specifically: Extract the strong echo grid points with echo intensity greater than or equal to the preset value from the real-time radar reflectivity CAPPI data, merge the spatially adjacent strong echo grid points, and remove the isolated points to obtain the strong echo area, i.e., the thunderstorm area; the isolated points are the strong echo grid points far from the merged points.

7. A targeted lightning classification and early warning method based on multi-source detection data according to claim 1, characterized in that The warning area also includes: the warning area and the attention area; The distance between the outer edge of the bull's-eye area and the bull's-eye < the distance between the outer edge of the warning area and the bull's-eye < the distance between the outer edge of the attention area and the bull's-eye.

8. A targeted lightning grading warning method based on multi-source detection data according to claim 7, characterized in that, The range of the bull's-eye area is: the area within a radius of [0, 10] km centered on the bull's-eye; the range of the warning area is: the area within a radius of (10, 30] km centered on the bull's-eye; the range of the attention area is: the area within a radius of (30, 60] km centered on the bull's-eye.

9. A targeted lightning classification and early warning system based on multi-source detection data, characterized in that, It includes: A warning area module, which is used to set multiple warning areas with different levels with the user's location as the bull's-eye; The higher the level, the closer the outer edge of the corresponding warning area is to the bull's-eye; the warning area includes the bull's-eye area; A real-time data processing module, which is used to obtain real-time detection data, and determine the thunder warning target level based on the multiple warning areas corresponding to the user and the real-time detection data, and generate thunder level warning information; The real-time detection data includes three-dimensional lightning data and atmospheric electric field data; A radar data acquisition module, which is used to obtain the real-time radar reflectivity CAPPI data at the target height of the real-time vertical temperature layer; the target height of the vertical temperature layer is the height of the -10°C layer detected by the microwave radiometer; A thunderstorm area identification module, which is used to obtain the thunderstorm area by using the preset method with the real-time radar reflectivity CAPPI data, and continuously update the thunderstorm area; A thunderstorm area extrapolation module, which is used to calculate the optical flow vector field corresponding to the current moment in real time through the reflectivity factor data of the radar at the target height at the current moment and the previous moment, obtain the moving direction and moving speed of the current radar echo through the optical flow vector field, and predict the multiple positions that the thunderstorm area will pass through within a preset time period at the current moment; An estimation module, which is used to obtain the estimated impact time when the thunderstorm area corresponding to a certain position among the passed positions covers the bull's-eye, that is, the time when the forefront of the thunderstorm area reaches the bull's-eye area; A warning information generation module, which is used to generate thunder approaching warning information through the thunder level warning information and / or the estimated impact time.

Images