Lightning high-density region generation method

Abstract

The invention relates to a lightning high-density data generation method, and belongs to the technical field of lightning nowcasting. The method comprises the following steps: firstly, establishing a ground-based wide-area lightning positioning network; the lightning positioning point is regarded as an area with probability of occurrence of lightning, and a probability distribution expression of occurrence of lightning is constructed; and the lightning occurrence density corresponding to the lightning positioning point in one time step is superposed, and data obtained after superposition represents the number of times of possible lightning in statistics in one time step of the region regarded by the lightning positioning point. And carrying out hybrid calculation on data obtained by superposing the current time step length and data obtained by superposing the first n-1 time step lengths by utilizing a weight method to obtain the final mixed lightning density of the region regarded as the lightning positioning point, the lightning density image of the to-be-monitored range in the time period and the lightning high-density region. According to the method, the problems of overlarge grids, rough resolution, serious numerical sudden change and the like of the traditional lightning frequency statistical data are optimized, the problem of lack of proper data forms is finally solved, and the method has a good application prospect in the field of lightning nowcasting.

Description

Technical Field

[0001] The invention belongs to the technical field of lightning nowcasting, and in particular relates to a method for generating a lightning high-density area. Background Art

[0002] Nowcasting is a weather forecasting method used to predict weather changes within the next few hours. It is based on the movement and evolution trend of weather systems and inputs past observation data into the forecast model to obtain forecast results. Among them, the extrapolation of high-density lightning areas is a very important part.

[0003] A deep neural network is a machine learning model that can learn and extract complex features from input data through connections and weight adjustments between multiple layers of neurons. Although deep neural networks have been widely used in nowcasting, under existing technical conditions, the use of deep neural networks for extrapolation of high-density lightning areas still lacks a good data form. Existing deep neural networks used for nowcasting all accept grid data as input, which is incompatible with scattered lightning location data. Therefore, in order to use artificial intelligence technology to apply discrete lightning location data to nowcasting warning forecasts, lightning data must be gridded. Summary of the invention

[0004] The purpose of the present invention is to overcome the shortcomings of the prior art, apply discrete lightning location data to near-warning forecasting, and propose a method for generating a lightning high-density area.

[0005] To achieve the above purpose, the technical solution adopted by the present invention is as follows:

[0006] A method for generating a lightning high-density area comprises the following steps:

[0007] Step (1), establishing a ground-based wide-area lightning locating network within the range to be monitored, wherein the ground-based wide-area lightning locating network includes a plurality of base stations, each base station includes a very low frequency vertically polarized antenna for locating and a fast electric field change detector;

[0008] Step (2), based on the data collected by establishing a ground-based wide-area lightning location network, the lightning location point is regarded as an area where lightning is likely to occur, and a spatial distribution expression of lightning density is constructed;

[0009] Step (3), superimposing the lightning density corresponding to the lightning location point in a time step, the data obtained after superposition represents the number of lightning that may statistically occur in the area regarded as the lightning location point within a time step; using the threshold of the number of lightning occurrences, judging according to the data obtained after superposition, to obtain the lightning high density area in this time step;

[0010] Step (4), for a time period to be detected containing n time steps, the data obtained by superimposing the current time step is mixed and calculated with the data obtained by superimposing the previous n-1 time steps using the weighting method, so as to obtain the final mixed lightning density of the area regarded as the lightning location point; based on the final mixed lightning density of the area regarded as each lightning location point, the lightning density image within the time period of the range to be monitored is obtained, thereby obtaining the lightning high density area from the lightning density image within the range to be monitored.

[0011] Further, preferably, the specific method of step (2) is:

[0012] For any lightning location, there is a probability distribution of lightning with the lightning location as the center. A two-dimensional Gaussian pulse function is used as the spatial distribution of lightning density.

[0013] The peak value of the Gaussian distribution is set to 1 and the mean value is set to 0; the following spatial distribution expression of lightning density is obtained:

[0014] r = x 2 +y 2

[0015]

[0016] Among them, the location where lightning occurs is the coordinate origin, r represents the distance from any point to the coordinate origin, x and y represent the two-dimensional direct coordinates of the point, σ represents the Gaussian pulse width, and G(x,y) is the lightning density value.

[0017] Further, preferably, r is less than 50 km.

[0018] Further, preferably, the specific method of step (3) is:

[0019] The lightning densities corresponding to the lightning location points within a time step are superimposed. The superimposed data represent the number of times lightning may statistically occur within a time step in the area regarded as the lightning location point.

[0020] Set the threshold of the number of lightning occurrences. If the data obtained after superposition exceeds the threshold, the threshold represents the number of lightning that may statistically occur in the area regarded as the lightning location point within a time step, and the area regarded as the lightning location point is the high-density lightning area within this time step;

[0021] If the threshold is not exceeded, the superimposed data represents the number of lightning strikes that are statistically likely to occur within one time step in the area considered as the lightning location point.

[0022] Further, preferably, the length of the time step is set to 60 minutes, every 6 minutes is regarded as the same lightning density, and the threshold value of the number of lightning occurrences is 25.

[0023] Further, preferably, the specific method of step (4) is:

[0024] For the data of a certain time step, get n-1 time steps forward, a total of n time steps for weighted mixing, the calculation method is as follows:

[0025] (4.1) The decay function is set to e -x ;

[0026] (4.2) According to the time order, take the current time step as the first time step, normalize each time step, and then multiply it by 10 as the intermediate variable k i ; i is 1, 2, 3…n; that is, the intermediate variable of the current time step is k1, the intermediate variable of the previous time step is k2, the intermediate variable of the previous time step is k3, and so on;

[0027] (4.3) For the weight coefficient w of the i-th time step i for:

[0028]

[0029] Where n represents the total number of time steps required for mixing, and w i represents the weight of the i-th time period during mixing, k i is the intermediate variable, P i It represents the data obtained after superposition of the i-th time step in the area regarded as the lightning location point; P represents the final mixed lightning density in the area regarded as the lightning location point.

[0030] Furthermore, preferably, the final mixed lightning density of the area regarded as each lightning location point corresponds to a corresponding color value, the greater the density, the darker the color, and is displayed on a map, thereby obtaining a lightning density image of the range to be monitored within the time period.

[0031] The present invention also provides a method for forecasting lightning levels, which uses a deep neural network model for prediction. The lightning density image of the monitored range within the time period obtained by the above method is used as the input of the deep neural network model, and the regional lightning level regarded as the corresponding lightning location point is used as the output for training, thereby obtaining a deep neural network model for forecasting lightning levels.

[0032] In step (1) of the present invention, multiple base stations are evenly arranged within the range to be monitored, but are not limited to this and may also be arranged according to actual needs; the present invention has no specific limitation on the range to be monitored, which may be the entire country, but may also be a certain region, a certain province, etc., and the present invention does not impose any limitation on this.

[0033] When superimposing in step (3) of the present invention, the obtained superimposed value can be directly displayed, and the superimposed value can be displayed, or the superimposed value can be rounded off and then the corresponding integer value can be displayed.

[0034] The present invention replaces the lightning location point with a two-dimensional Gaussian pulse function. The center position of the Gaussian pulse function (ie, the location point) represents the place with the highest lightning density. The farther from the center position, the smaller the lightning density.

[0035] The present invention constructs a weight function that decays with time, so that the intensity of the Gaussian pulse function corresponding to each lightning location point gradually decreases with time. The present invention takes into account the temporal nature of lightning data, that is, the lightning position may change significantly over time. Therefore, the present invention proposes a temporal mixing processing method. Each discrete lightning location point is expressed by a Gaussian pulse function, and a weight function that decays with time is constructed, so that the intensity of the Gaussian pulse function corresponding to each lightning location point gradually decreases with time. In this way, the spatiotemporal continuity of the lightning density distribution can be achieved.

[0036] The Gaussian pulse functions corresponding to all lightning location points in a period of time, which take into account the gradual attenuation over time, are superimposed to obtain the total lightning high-density distribution in this period of time.

[0037] In the present invention, r can be set to be less than 50 km, and σ is also set to be a constant according to actual conditions. The larger σ is, the smaller the attenuation of lightning density distribution as the distance increases; conversely, the larger σ is.

[0038] The purpose of the present invention (4.2) is to adjust n to a range of 1 to 10.

[0039] Compared with the prior art, the present invention has the following beneficial effects:

[0040] The present invention provides a method for generating lightning high-density data. In the work of extrapolating lightning in high-density areas, the present invention overcomes the problem that lightning location point data cannot be directly input into the prediction model, optimizes the problems of traditional lightning frequency statistical data such as too large grid, rough resolution, serious numerical mutation, etc., and finally solves the problem of lack of suitable data form. The present invention has a good application prospect in the field of lightning nowcasting. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1It is a schematic flow chart of the method for generating a lightning high-density area according to the present invention;

[0042] Figure 2 A schematic diagram of the weight function used in the present invention;

[0043] Figure 3 Four-hour lightning density images generated for the application example of the present invention; (a) is from 7:00 to 8:00 on November 28, 2022; (b) is from 8:00 to 9:00 on November 28, 2022; (c) is from 9:00 to 10:00 on November 28, 2022; (d) is from 10:00 to 11:00 on November 28, 2022;

[0044] Figure 4 The lightning density image of lightning density on November 27, 2022 was generated for the application example of the present invention. DETAILED DESCRIPTION

[0045] The present invention is further described in detail below in conjunction with embodiments.

[0046] Those skilled in the art will appreciate that the following examples are only used to illustrate the present invention and should not be considered to limit the scope of the present invention. If no specific techniques or conditions are specified in the examples, the techniques or conditions described in the literature in the art or the product specifications are used. If the manufacturer of the materials or equipment used is not specified, they are all conventional products that can be purchased.

[0047] like Figure 1 As shown, a method for generating a lightning high-density area includes the following steps:

[0048] Step (1), establishing a ground-based wide-area lightning locating network within the range to be monitored, wherein the ground-based wide-area lightning locating network includes a plurality of base stations, each base station includes a very low frequency vertically polarized antenna for locating and a fast electric field change detector;

[0049] Step (2), based on the data collected by establishing a ground-based wide-area lightning location network, the lightning location point is regarded as an area where lightning is likely to occur, and a spatial distribution expression of lightning density is constructed;

[0050] Step (3), superimposing the lightning density corresponding to the lightning location point in a time step, the data obtained after superposition represents the number of lightning that may statistically occur in the area regarded as the lightning location point within a time step; using the threshold of the number of lightning occurrences, judging according to the data obtained after superposition, to obtain the lightning high density area in this time step;

[0051] Step (4), for a time period to be detected containing n time steps, the data obtained by superimposing the current time step is mixed and calculated with the data obtained by superimposing the previous n-1 time steps using the weighting method, so as to obtain the final mixed lightning density of the area regarded as the lightning location point; based on the final mixed lightning density of the area regarded as each lightning location point, the lightning density image within the time period of the range to be monitored is obtained, thereby obtaining the lightning high density area from the lightning density image within the range to be monitored.

[0052] The specific method of step (2) is as follows:

[0053] For any lightning location, there is a probability distribution of lightning with the lightning location as the center. A two-dimensional Gaussian pulse function is used as the spatial distribution of lightning density.

[0054] The peak value of the Gaussian distribution is set to 1 and the mean value is set to 0; the following spatial distribution expression of lightning density is obtained:

[0055] r = x 2 +y 2

[0056]

[0057] Among them, the location where lightning occurs is the coordinate origin, r represents the distance from any point to the coordinate origin, x and y represent the two-dimensional direct coordinates of the point, σ represents the Gaussian pulse width, and G(x,y) is the lightning density value.

[0058] Preferably, r is less than 50 km.

[0059] The specific method of step (3) is:

[0060] The lightning densities corresponding to the lightning location points within a time step are superimposed. The superimposed data represent the number of times lightning may statistically occur within a time step in the area regarded as the lightning location point.

[0061] Set the threshold of the number of lightning occurrences. If the data obtained after superposition exceeds the threshold, the threshold represents the number of lightning that may statistically occur in the area regarded as the lightning location point within a time step, and the area regarded as the lightning location point is the high-density lightning area within this time step;

[0062] If the threshold is not exceeded, the superimposed data represents the number of lightning strikes that are statistically likely to occur within one time step in the area considered as the lightning location point.

[0063] The length of the time step is set to 60 minutes, every 6 minutes is considered to be the same lightning density, and the threshold for the number of lightning occurrences is 25.

[0064] The specific method of step (4) is:

[0065] For the data of a certain time step, get n-1 time steps forward, a total of n time steps for weighted mixing, the calculation method is as follows:

[0066] (4.1) The decay function is set to e -x ;

[0067] (4.2) According to the time order, take the current time step as the first time step, normalize each time step, and then multiply it by 10 as the intermediate variable k i ; i is 1, 2, 3…n; that is, the intermediate variable of the current time step is k1, the intermediate variable of the previous time step is k2, the intermediate variable of the previous time step is k3, and so on;

[0068] (4.3) For the weight coefficient w of the i-th time step i for:

[0069]

[0070] Where n represents the total number of time steps required for mixing, and w i represents the weight of the i-th time period during mixing, k i is the intermediate variable, P i It represents the data obtained after superposition of the i-th time step in the area regarded as the lightning location point; P represents the final mixed lightning density in the area regarded as the lightning location point.

[0071] like Figure 2 The figure shows a time series mixed weight function, where the horizontal axis represents the dimension of time and the vertical axis represents the attenuation coefficient. Multiplying this time series weight function with the Gaussian pulse function can achieve the continuity of the spatiotemporal distribution of lightning density. Multiplying this time series weight function with the Gaussian pulse function expresses that the density cloud corresponding to each lightning gradually decays over time.

[0072] The present invention can set a certain threshold value according to the lightning density image obtained within the time period of the monitored range, and the area exceeding the threshold value is a lightning high density area. The present invention does not make any special limitation on the threshold value, and it can be set according to actual needs.

[0073] The final mixed lightning density of the area regarded as each lightning location point corresponds to a corresponding color value. The greater the density, the darker the color. It is displayed on the map, thereby obtaining an image of the lightning density within the time period of the monitored range.

[0074] A method for forecasting lightning levels, using a deep neural network model for prediction, wherein the lightning density image of the monitored range within the time period obtained by the above-mentioned lightning high-density area generation method is used as the input of the deep neural network model, and the regional lightning level regarded as the corresponding lightning location point is used as the output for training, thereby obtaining a deep neural network model for forecasting lightning levels.

[0075] Application Examples

[0076] 1. Data Source

[0077] The lightning location data used in this experiment comes from the VLF Lightning Location Network (VLF-LLN) deployed and gradually improved in China by Nanjing University of Information Science and Technology since September 2021. This experiment collected a total of 54,394 data from November 27 to 28, 2022. These data cover the key parameters of each lightning event, such as the time of occurrence and geographic location coordinates. Figure 2 Shown is the spatial distribution of lightning density.

[0078] 2. A method for generating a lightning high-density area, comprising the following steps:

[0079] Step (1), establishing a ground-based wide-area lightning locating network within the range to be monitored, wherein the ground-based wide-area lightning locating network includes a plurality of base stations, each base station includes a very low frequency vertically polarized antenna for locating and a fast electric field change detector;

[0080] Step (2), based on the data collected by establishing a ground-based wide-area lightning location network, the lightning location point is regarded as an area where lightning is likely to occur, and a spatial distribution expression of lightning density is constructed;

[0081] Step (3), superimposing the lightning density corresponding to the lightning location point in a time step, the data obtained after superposition represents the number of lightning that may statistically occur in the area regarded as the lightning location point within a time step; using the threshold of the number of lightning occurrences, judging according to the data obtained after superposition, to obtain the lightning high density area in this time step;

[0082] Step (4), for a time period to be detected containing n time steps, the data obtained by superimposing the current time step is mixed and calculated with the data obtained by superimposing the previous n-1 time steps using the weighting method, so as to obtain the final mixed lightning density of the area regarded as the lightning location point; based on the final mixed lightning density of the area regarded as each lightning location point, the lightning density image within the time period of the range to be monitored is obtained, thereby obtaining the lightning high density area from the lightning density image within the range to be monitored.

[0083] Step (1) specifically includes:

[0084] Build a nationwide ground-based wide-area lightning location network, with each base station including a very low frequency vertically polarized antenna (bandwidth within 30kHz) directly used for positioning. All base stations continuously collect data at a sampling rate of 1MHz, with a data record length of 1ms and a pre-trigger time of 300μs, and save the original waveform data greater than the minimum threshold (the minimum threshold is generally set according to the actual situation, mainly used for denoising). At present, the positioning quality of the ground-based wide-area lightning location network is good, and good positioning results are obtained for single-cell thunderstorms, multi-cell thunderstorms, squall lines and other processes at different times.

[0085] Step (2) specifically includes:

[0086] For any lightning location point, there is a Gaussian distribution centered on it. Since Gaussian distribution is the most common and basic probability distribution in nature, the Gaussian distribution is used as the probability distribution of lightning. Its one-dimensional form can be described by the following formula:

[0087]

[0088] Among them, σ represents the variance, which can also represent the concentration and peak value of the distribution, and μ represents the mean. For the convenience of calculation, when converting the positioning point into a Gaussian distribution, we set the peak value of the Gaussian distribution to 1 (that is, representing "1" lightning), the mean to 0, and use its two-dimensional form as follows:

[0089] r = x 2 +y 2

[0090]

[0091] Among them, the location where lightning occurs is the coordinate origin, r represents the distance from any point to the coordinate origin, x and y represent the two-dimensional direct coordinates of the point, σ represents the Gaussian pulse width, and G(x,y) is the lightning density value.

[0092] The lightning location point is regarded as an area with a probability of lightning occurrence. The closer the probability of regional lightning occurrence is to the center, the closer it is to 100%; according to experiments, r is set to be less than 25km.

[0093] Step (3) specifically includes:

[0094] The Gaussian distributions corresponding to all lightning locations within a time step are superimposed on a gridded data filled with 0s, and the length of the time step is set to six minutes. The upper limit of the value of each grid point is 25, which means that 25 lightning strikes may statistically occur in the area within 6 minutes. If the value exceeds 25, it will be reduced to 25. If the value of a grid point reaches 25, then the area where the grid point is located is a high-density lightning area. If not, the value can still express the density information of lightning. For thunderstorms of average intensity, the transition from high-density areas (value of 25) to no possibility of lightning (value of 0) is relatively smooth; for strong thunderstorms, this transition will be steeper because a large number of Gaussian distributions are superimposed on the same area, and then clipped with a maximum value of 25. Since the data input into the deep neural network must first be normalized, and the location and number of lightning are random, it is impossible to determine the upper limit of the number of lightning strikes in a certain area. Therefore, in addition to representing high-density areas, the clipped data can also prevent the problem that relatively weak thunderstorms are easily affected by the numerical value of the strongest thunderstorm after normalization. At the same time, the steepness of the edge can indirectly reflect the original strength of a high-density area, and information will not be lost.

[0095] Step (4) specifically includes:

[0096] For the data of a certain time period, the time series is mixed with the data of the previous time period according to a certain weight, and the earlier the time, the smaller the weight. The weight used is taken from the negative semi-axis of the natural exponential function. This is because the area between the negative semi-axis of the natural exponential function and the x-axis is 1. Taking values ​​from such a function can ensure that the sum of the weights is 1, that is, the numerical range of the data after time series mixing is still 100% of the original, such as Figure 2 As shown, the horizontal axis represents time information and the vertical axis represents intensity attenuation.

[0097] The specific algorithm of time series mixing is: for the data in a certain time period, n-1 time steps should be obtained forward, and a total of n time steps should be used for weighted mixing. The weight calculation method is shown in the following formula.

[0098] (4.1) The decay function is set to e -x ;

[0099] (42) According to the time order, the current time step is taken as the first time step. Each time step is normalized and then multiplied by 10 to be the intermediate variable k. i ; i is 1, 2, 3…n; that is, the intermediate variable of the current time step is k1, the intermediate variable of the previous time step is k2, the intermediate variable of the previous time step is k3, and so on;

[0100] (4.3) For the weight coefficient w of the i-th time stepi for:

[0101]

[0102] Where n represents the total number of time steps required for mixing, and w i represents the weight of the i-th time period during mixing, k i is the intermediate variable, P i It represents the data obtained after superposition of the i-th time step in the area regarded as the lightning location point; P represents the final mixed lightning density in the area regarded as the lightning location point.

[0103] k i is an intermediate variable, the purpose of which is to map 1 to n to 1 to 10, and e -10 =4×10 -5 ≈0, which can be convenient for integration in actual calculations.

[0104] 3. Test results

[0105] Using the lightning high-density area generation method, this experiment successfully identified four strong weather processes, namely 7:00 to 8:00, 8:00 to 9:00, 9:00 to 10:00, and 10:00 to 11:00 on November 28, 2022, as follows Figure 3 Through the intuitive display of the image, the temporal and spatial distribution characteristics of lightning activity in these four hours can be clearly observed. The depth of color represents the density and intensity changes of lightning events in different periods, effectively revealing the dynamic process of the evolution of the thunderstorm system.

[0106] At the same time, according to the above method, we can get Figure 4 The lightning density and high-density distribution of lightning on November 27, 2022 are shown in red. Each dot represents a lightning location point. The darker the color, the higher the lightning density in the area; the lighter the color, the lower the lightning density in the area. In the upper left corner of the picture, the lightning location points are more scattered, so the lightning density in this area is lower, so the color is lighter; in the center of the picture, the lightning location points are relatively concentrated, so the lightning density in this area is higher, and the color is dark red.

Claims

1. A method for generating a lightning high-density area, characterized in that: The steps include: Step (1), establishing a ground-based wide-area lightning locating network within the range to be monitored, wherein the ground-based wide-area lightning locating network includes a plurality of base stations, each base station includes a very low frequency vertically polarized antenna for locating and a fast electric field change detector; Step (2), based on the data collected by establishing a ground-based wide-area lightning location network, the lightning location point is regarded as an area where lightning is likely to occur, and a spatial distribution expression of lightning density is constructed; Step (3), superimposing the lightning density corresponding to the lightning location point in a time step, the data obtained after superposition represents the number of lightning that may statistically occur in the area regarded as the lightning location point within a time step; using the threshold of the number of lightning occurrences, judging according to the data obtained after superposition, to obtain the lightning high density area in this time step; Step (4), for a time period to be detected containing n time steps, the data obtained by superimposing the current time step is mixed and calculated with the data obtained by superimposing the previous n-1 time steps using the weighting method, so as to obtain the final mixed lightning density of the area regarded as the lightning location point; based on the final mixed lightning density of the area regarded as each lightning location point, the lightning density image within the time period of the range to be monitored is obtained, thereby obtaining the lightning high density area from the lightning density image within the range to be monitored.

2. The method for generating a lightning high-density area according to claim 1, characterized in that: The specific method of step (2) is: For any lightning location, there is a probability distribution of lightning with the lightning location as the center. A two-dimensional Gaussian pulse function is used as the spatial distribution of lightning density. The peak value of the Gaussian distribution is set to 1 and the mean value is set to 0; the following spatial distribution expression of lightning density is obtained: r=x 2 +y 2 Among them, the location where lightning occurs is the coordinate origin, r represents the distance from any point to the coordinate origin, x and y represent the two-dimensional direct coordinates of the point, σ represents the Gaussian pulse width, and G(x,y) is the lightning density value.

3. The method for generating a lightning high-density area according to claim 2, characterized in that: r is less than 50km.

4. The method for generating a lightning high-density area according to claim 3, characterized in that: The specific method of step (3) is: The lightning densities corresponding to the lightning location points within a time step are superimposed. The superimposed data represent the number of times lightning may statistically occur within a time step in the area regarded as the lightning location point. Set the threshold of the number of lightning occurrences. If the data obtained after superposition exceeds the threshold, the threshold represents the number of lightning that may statistically occur in the area regarded as the lightning location point within a time step, and the area regarded as the lightning location point is the high-density lightning area within this time step; If the threshold is not exceeded, the superimposed data represents the number of lightning strikes that are statistically likely to occur within one time step in the area considered as the lightning location point.

5. The method for generating a lightning high-density area according to claim 4, characterized in that: The length of the time step is set to 60 minutes, every 6 minutes is considered to be the same lightning density, and the threshold for the number of lightning occurrences is 25.

6. The method for generating a lightning high-density area according to claim 1, characterized in that: The specific method of step (4) is: For the data of a certain time step, get n-1 time steps forward, a total of n time steps for weighted mixing, the calculation method is as follows: (4.1) The decay function is set to e -x ; (4.2) According to the time order, take the current time step as the first time step, normalize each time step, and then multiply it by 10 as the intermediate variable k i ; i is 1, 2, 3…n; that is, the intermediate variable of the current time step is k1, the intermediate variable of the previous time step is k2, the intermediate variable of the previous time step is k3, and so on; (4.3) For the weight coefficient w of the i-th time step i for: Where n represents the total number of time steps required for mixing, and w i represents the weight of the i-th time period during mixing, k i is the intermediate variable, P i It represents the data obtained after superposition of the i-th time step in the area regarded as the lightning location point; P represents the final mixed lightning density in the area regarded as the lightning location point.

7. The method for generating a lightning high-density area according to claim 6, characterized in that: The final mixed lightning density of the area regarded as each lightning location point corresponds to a corresponding color value. The greater the density, the darker the color. It is displayed on the map, thereby obtaining an image of the lightning density within the time period of the monitored range.

8. A method for predicting lightning levels using a deep neural network model, characterized in that: The lightning density image within the time period of the monitored range obtained by the method for generating a high-density lightning area according to claim 1 is used as the input of the deep neural network model, and the regional lightning level regarded as the corresponding lightning location point is used as the output for training, thereby obtaining a deep neural network model for predicting lightning level.

Images