Gamma distribution-based ground lightning density statistical model and lightning stroke risk assessment method thereof

Abstract

The invention is suitable for the technical field of lightning stroke risk assessment, and provides a cloud-to-ground lightning density statistical model based on gamma distribution and a lightning stroke risk assessment method thereof, and the method comprises the steps: collecting historical cloud-to-ground lightning data in a preset time range of a target region; the target area is divided into a plurality of grid units, and a year-by-year cloud-to-ground lightning density data set of each grid unit is determined based on historical cloud-to-ground lightning data; performing gamma distribution fitting on the year-by-year cloud-to-ground lightning density data set by adopting a maximum likelihood estimation method, and constructing a cloud-to-ground lightning density statistical model; calculating a plurality of preset probability quantiles based on a cloud-to-ground lightning density statistical model; and dividing the cloud-to-ground lightning density into a plurality of continuous risk intervals according to the probability quantile, wherein each risk interval corresponds to one lightning risk level. According to the method, through gamma distribution modeling and dynamic correction modes, the problems of poor adaptability of a distribution model, risk assessment threshold solidification and the like in traditional lightning stroke risk assessment are solved, and a reliable risk quantification decision basis is provided for power grid lightning protection transformation.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of lightning risk assessment, and particularly relates to a ground flash density statistical model based on a gamma distribution and a lightning risk assessment method thereof. BACKGROUND

[0002] A ground flash refers to an atmospheric discharge phenomenon between a cloud body and the earth or ground, and a ground flash density refers to an average ground flash number recorded by a lightning positioning system per unit area and per unit time. The parameter reflects the frequency of discharge of a thundercloud to the ground and is a key parameter for evaluating lightning risk of a power grid. An accurate ground flash density distribution rule of a transmission line can effectively guide differentiated lightning protection of the line and power grid planning and design, and improve stability of power grid operation.

[0003] In the prior art, ground flash density modeling mainly relies on a normal distribution or a Poisson distribution. However, because the ground flash density data has a right bias and a non-negative characteristic, the symmetry assumption of the normal distribution leads to distortion of a tail part of a high-density lightning area, so that the normal distribution has insufficient adaptability. In addition, because lightning activity is affected by coupling of multiple factors such as terrain, climate and atmospheric electric field, the independent and uniform assumption of the Poisson distribution is seriously deviated, so that an annual average ground flash density evaluation method based on the Poisson model cannot represent dynamic characteristics of lightning activity. It can be seen that the above modeling methods have problems of poor model adaptability and fixed risk evaluation threshold, which easily lead to misjudgment of lightning risk evaluation and mismatch of protection strategies.

[0004] In view of this, a ground flash density statistical model based on a gamma distribution and a lightning risk assessment method thereof are provided. SUMMARY

[0005] The present application provides a ground flash density statistical model based on a gamma distribution and a lightning risk assessment method thereof, which are used to solve the problem of misjudgment of lightning risk evaluation.

[0006] The present application provides a ground flash density statistical model based on a gamma distribution and a lightning risk assessment method thereof, which include:

[0007] Collect historical ground flash data in a preset time range of a target area;

[0008] Divide the target area into a plurality of grid units according to a preset spatial resolution, and determine a yearly ground flash density data set of each grid unit based on the historical ground flash data;

[0009] Perform gamma distribution fitting on the yearly ground flash density data set by using a maximum likelihood estimation method, determine a shape parameter and a scale parameter representing statistical characteristics of the ground flash density of the target area, and construct a ground flash density statistical model;

[0010] calculate a plurality of preset probability quantile values based on the ground flash density statistical model, the probability quantile values being used for dividing probability distribution intervals of the ground flash density;

[0011] divide the ground flash density into a plurality of continuous risk intervals according to the probability quantile values, each risk interval corresponding to a lightning risk level, the lightning risk level including an extremely low risk, a low risk, a medium risk, a high risk, and an extremely high risk.

[0012] Further, the target area is divided into a plurality of grid cells according to a preset spatial resolution, and a yearly ground flash density data set of each grid cell is determined based on the historical ground flash data, including:

[0013] count the total number of ground flashes in each year in each grid cell, calculate the annual ground flash frequency per unit area based on the geographical area of the grid cell, and generate a yearly ground flash density data sequence of each grid cell.

[0014] Further, the yearly ground flash density data set is fitted with a gamma distribution by using a maximum likelihood estimation method, shape parameters and scale parameters representing statistical characteristics of the ground flash density of the target area are determined, and a ground flash density statistical model is constructed, including:

[0015] perform parameter estimation on the yearly ground flash density data set according to a probability density function of the gamma distribution, to obtain the shape parameters and the scale parameters representing the statistical characteristics of the ground flash density, the probability density function being used to describe the probability distribution characteristics of the annual ground flash frequency per unit area in the target area.

[0016] Further, the probability density function includes:

[0017]

[0018] wherein x is the ground flash density, k is the shape parameter, θ is the scale parameter, and Γ(k) is the gamma function.

[0019] Further, before the plurality of preset probability quantile values are calculated based on the ground flash density statistical model, the method further includes:

[0020] verify the matching degree between the yearly ground flash density data set and the fitted gamma distribution by using a statistical consistency test method;

[0021] if the test is passed, it is determined that the gamma distribution model conforms to the statistical law of the historical data.

[0022] Further, the matching degree between the yearly ground flash density data set and the fitted gamma distribution is verified by using the statistical consistency test method, including:

[0023] If the test fails, data transformation is performed on the annual lightning density data set;

[0024] The gamma distribution is refitted based on the transformed data to obtain corrected shape and scale parameters;

[0025] The matching degree of the corrected gamma distribution and the transformed data is verified by a statistical consistency test method.

[0026] Further, the method further comprises:

[0027] According to the fitted gamma distribution, the corresponding probability quantile value is calculated by the inverse function of the cumulative distribution function, and the probability quantile value is a threshold value of lightning density under a preset cumulative probability.

[0028] Further, the preset plurality of probability quantile values include 10%, 30%, 70%, and 90% quantile values.

[0029] Further, the method further comprises:

[0030] An interval in which the lightning density is lower than the 10% quantile value is defined as an extremely low risk.

[0031] An interval in which the lightning density is between the 10% and 30% quantile values is defined as a low risk.

[0032] An interval in which the lightning density is between the 30% and 70% quantile values is defined as a medium risk.

[0033] An interval in which the lightning density is between the 70% and 90% quantile values is defined as a high risk.

[0034] An interval in which the lightning density is higher than the 90% quantile value is defined as an extremely high risk.

[0035] Further, after the lightning density is divided into a plurality of continuous risk intervals according to the probability quantile values, each risk interval corresponds to a lightning risk level, the method further comprises:

[0036] A lightning protection strategy corresponding to the lightning risk level is matched; wherein:

[0037] An extremely low risk level corresponds to installation of basic lightning protection facilities and surge protectors.

[0038] A low risk level corresponds to addition of overhead ground wires and coordinated configuration of multi-stage surge protection.

[0039] A medium risk level corresponds to installation of zinc oxide arresters and deployment of electromagnetic shielding layers.

[0040] The high-risk level corresponds to the differentiated configuration of the multi-loop line and double protection device of the important section;

[0041] The very high-risk level corresponds to the underground cable laying and intelligent online monitoring system integration.

[0042] From the above technical solutions, the present application has the following advantages:

[0043] The present application can accurately match the high-frequency low-value concentrated distribution and low-frequency high-value long-tail extension characteristics of lightning activity through the fitting of the ground flash density corresponding to the ground flash data of the target area and the gamma distribution, and through the dynamic boundary division mode of the quantile value of the gamma distribution, a five-level risk interval system is constructed, which can automatically adapt to the lightning intensity fluctuation of different regions and different climate periods according to the distribution intrinsic characteristics, realize the adaptive adjustment of the risk boundary, effectively improve the accuracy of the model risk assessment, make the resource investment proportional to the risk level, and ensure that the protection strategy is more accurate and efficient. BRIEF DESCRIPTION OF DRAWINGS

[0044] Figure 1 It is an embodiment flow schematic diagram of a ground flash density statistical model based on a gamma distribution and a lightning stroke risk assessment method in the present application;

[0045] Figure 2 It is a ground flash density probability distribution histogram in the present application;

[0046] Figure 3 It is a five-level lightning stroke risk interval schematic diagram based on a gamma distribution division in the present application;

[0047] Figure 4 It is a lightning stroke risk level schematic diagram of a certain region in the present application. DETAILED DESCRIPTION

[0048] The terms "first", "second", "third", "fourth" and the like in the specification of this application and in the above drawings (if any) are used to distinguish similar objects, and do not necessarily have to describe a particular order or sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "include" and "correspond to" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device including a series of steps or units does not have to be limited to those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0049] Embodiment one

[0050] The method implemented in the embodiment can be implemented in a system, and can be implemented in a server or a terminal, and the specific implementation is not limited. From the perspective of system implementation, the statistical model of cloud-to-ground lightning density based on gamma distribution and the lightning risk assessment method thereof in the present application will be introduced. Please refer to Figure 1 The method provided in the embodiment of the present application includes the following steps:

[0051] S11. Collect historical cloud-to-ground lightning data in a target area within a preset time range;

[0052] In the embodiment, the preset time range is set to 6 years, that is, 6 years of historical cloud-to-ground lightning data in the target area are collected, and the data is obtained from a lightning location system. Each record contains the timestamp, geographic coordinates including latitude and longitude, and current intensity of a cloud-to-ground lightning event. Only effective data jointly located by at least three detection stations are retained to exclude records with large single-station positioning errors and ensure spatial accuracy of the data. For example, the cloud-to-ground lightning data in a certain region of Guangxi from 2009 to 2014 is selected, the range is 20.875°N-22.625°N, 107.375°E-108.625°E, and a high-quality data set is formed after screening.

[0053] S12. Divide the target area into a plurality of grid cells according to a preset spatial resolution, and determine a yearly cloud-to-ground lightning density data set for each grid cell based on the historical cloud-to-ground lightning data;

[0054] In the embodiment, the determination of the cloud-to-ground lightning density data set includes the following:

[0055] 1. Statistics the total number of cloud-to-ground lightning in each grid cell per year, calculates the annual cloud-to-ground lightning frequency per unit area based on the geographic area of the grid cell, and generates a yearly cloud-to-ground lightning density data sequence for each grid cell.

[0056] Specifically, the preset spatial resolution is set to 0.25°x0.25°, about 715km 2 , the target area is divided into uniform grid cells according to the grid size, and the resolution balances the spatial fineness and calculation efficiency, ensuring that the cloud-to-ground lightning density in each grid has statistical significance. Each grid is uniquely identified by latitude and longitude coordinates (i,j), where i∈[1,M] and j∈[1,N], and MxN is the total number of grids.

[0057] For each grid cell, the total number of cloud-to-ground lightning is counted per year. For example, in the embodiment, the number of cloud-to-ground lightning in each grid of a certain region of Guangxi from 2009 to 2014 is counted per year, and the annual cloud-to-ground lightning frequency per unit area (times / km 2 ) is calculated according to the ratio of the total number of cloud-to-ground lightning to the area of the grid. The formula is: This index eliminates the influence of grid size difference on frequency statistics and realizes the standardization of spatial density.

[0058] Finally, the ground flash density values ​​are organized by year to generate a yearly dataset X based on grid units. i,j,t , where t is the year, such as t∈[2009,2014], and is stored in a matrix or geographic information system database to provide a data basis for subsequent statistical analysis and model fitting.

[0059] S13. Use the maximum likelihood estimation method to fit the gamma distribution of the annual CG flash density dataset, determine the shape and scale parameters that characterize the statistical characteristics of CG flash density in the target area, and construct a CG flash density statistical model.

[0060] In this embodiment, the maximum likelihood estimation method finds the shape and scale parameters of the gamma distribution so that the joint probability density of the historical ground flash density dataset under this distribution reaches the maximum value. In other words, it selects the parameter combination that is most likely to cause the observed data to appear, as follows:

[0061] 1. Parameters of the annual CG flash density dataset were estimated based on the probability density function of the gamma distribution. The shape and scale parameters that characterize the statistical characteristics of the CG flash density were obtained. The probability density function is used to describe the probability distribution characteristics of the annual CG flash frequency per unit area within the target region.

[0062] Input annual flash density dataset X={x1,x2,…,x n}, where x i is the annual ground-to-ground lightning density of the i-th grid cell; construct the likelihood function of the gamma distribution: f(x i ,k,θ) is the probability density function of the gamma distribution; finally, a numerical optimization algorithm is used to solve k and θ that maximize L(k,θ).

[0063] Among them, k reflects the skewness of the distribution of ground-to-ground lightning density. When k < 1, the ground-to-ground lightning density is highly skewed, indicating the presence of sporadic high-density lightning strikes in the region, such as in mountainous areas or areas of strong convection. When k > 1, the distribution is nearly symmetrical, indicating a relatively uniform ground-to-ground lightning density, such as in plains. θ reflects the dispersion of ground-to-ground lightning density. A larger value of θ indicates a wider fluctuation range in ground-to-ground lightning density and an increased probability of extreme high-density events. For a region in Guangxi, for example, the maximum likelihood estimation method yielded k = 3.79 and θ = 3.60. These parameters indicate that the ground-to-ground lightning density distribution in this region is nearly symmetrical (k > 1) with a moderate fluctuation range, consistent with the lightning activity characteristic of the region's mountainous terrain.

[0064] Specifically, the probability density function of the gamma distribution is:

[0065]

[0066] Where: x is the density of ground-to-ground flashes, k is the shape parameter, θ is the scale parameter, and Γ(k) is the gamma function.

[0067] The gamma distribution is only defined in the interval x>0, which fully matches the physical characteristics of ground-to-ground lightning density (non-negative values). Figure 2 and Figure 3 As shown in Figure 2, the goodness of fit of the model to historical data is verified by comparing the probability distribution histogram with the gamma distribution curve.

[0068] Figure 2 The histogram construction includes the following:

[0069] Calculation of statistical interval width: Among them, x max and x min are the maximum and minimum ground-to-ground flash density values ​​in the data set, and n is the total amount of data. Taking a certain area in Guangxi as an example, the ground-to-ground flash density ranges from 0 to 40 times / km 2 The total amount of data is n = 210 (35 grids × 6 years), and the calculated interval width is h = 2.5 times / km 2 The histogram can visually display the distribution characteristics of the CG flash density, such as right skewness and peak location, and assist in determining whether the gamma distribution assumption is suitable. By comparing the histogram with the fitted gamma distribution curve, it is possible to qualitatively assess whether the model accurately represents the data distribution.

[0070] S14. Calculate multiple preset probability quantile values ​​based on the statistical model of ground-to-crystal flash density, and the probability quantile values ​​are used to divide the probability distribution interval of the ground-to-crystal flash density;

[0071] In this embodiment, before step S14, the following is also included:

[0072] 1. Verify the matching degree between the annual flash density dataset and the fitted gamma distribution through statistical consistency test methods;

[0073] If the test passes, it is determined that the gamma distribution model conforms to the statistical laws of historical data.

[0074] If the test fails, the annual flash density dataset is transformed; the gamma distribution is refitted based on the transformed data to obtain the corrected shape parameters and scale parameters; and the matching degree between the corrected gamma distribution and the transformed data is verified by a statistical consistency test method.

[0075] The Kolmogorov-Smirnov (KS) test method is used here to test whether the annual flash density dataset obeys the fitted gamma distribution. It compares the maximum difference between the empirical distribution function of the dataset and the theoretical cumulative distribution function of the gamma distribution to determine whether the two are statistically consistent.

[0076] First, sort the annual lightning density dataset, calculate the cumulative probability of each data point; second, based on the fitted gamma distribution parameters, calculate the theoretical cumulative probability of each data point; calculate the maximum absolute difference between the empirical distribution function and the theoretical distribution function at all data points; finally, according to the maximum absolute difference and the sample size, calculate the corresponding p value through the K-S test table or numerical algorithm, if the p value is greater than 0.05, accept the original hypothesis, consider that the data obeys the gamma distribution; if the p value is less than or equal to 0.05, reject the original hypothesis, and the data needs to be modified.

[0077] Specifically, when the K-S test passes, it indicates that the fitted gamma distribution can accurately represent the statistical law of historical lightning density, so the current gamma distribution parameters are retained as the final result of the lightning density statistical model; based on the model, the quantile value is calculated, and the subsequent lightning risk assessment step is executed.

[0078] If the K-S test does not pass, it indicates that there is a significant deviation between the data distribution and the gamma distribution assumption, and the following modification steps need to be performed: power transformation is performed on the original lightning density data, and the formula is: Where λ is optimized by maximum likelihood method, so that the transformed data is closer to the gamma distribution characteristics; the transformation purpose is to eliminate data skewness or heteroscedasticity, and enhance the applicability of the distribution assumption.

[0079] Based on the transformed dataset, the maximum likelihood estimation method is applied again to obtain the modified parameters, and the K-S test is performed again on the modified gamma distribution; if it passes, the modified parameters are used to construct the model; if it still does not pass, the distribution assumption needs to be adjusted or the data preprocessing process needs to be optimized.

[0080] In this embodiment, the calculation of the plurality of preset probability quantile values based on the lightning density statistical model includes the following:

[0081] 1. According to the fitted gamma distribution, the corresponding probability quantile value is calculated through the inverse function of the cumulative distribution function, and the probability quantile value is the threshold of the lightning density at the preset cumulative probability; wherein the plurality of preset probability quantile values include 10%, 30%, 70% and 90% quantile values.

[0082] The calculation principle of quantile value: the inverse function of the cumulative distribution function of the fitted gamma distribution is used to determine the lightning density threshold corresponding to the preset cumulative probability. For example, the 90% quantile value means that the cumulative probability of lightning density below this threshold is 90%, and exceeding this threshold corresponds to the highest 10% of extreme cases. The predefined quantile values are 10%, 30%, 70%, and 90%. These quantile values divide the lightning density into five consecutive risk intervals (very low, low, medium, high, and very high risk). Through this method, the classification of risk levels is completely based on data-driven statistical rules, avoiding the bias of subjective threshold setting, while ensuring the objectivity and interpretability of the risk intervals. For example, the area with lightning density below the 10% quantile value is defined as very low risk, and the area with lightning density above the 90% quantile value is classified as very high risk. The intermediate quantile values define the low, medium, and high risk levels in turn, forming a progressive risk assessment system.

[0083] S15. Dividing the lightning density into a plurality of consecutive risk intervals according to the probability quantile value, each risk interval corresponding to a lightning risk level, the lightning risk level including very low risk, low risk, medium risk, high risk, and very high risk.

[0084] Based on the above step S14, the plurality of consecutive risk intervals in this embodiment include:

[0085] The interval with lightning density below the 10% quantile value is defined as very low risk; the interval with lightning density between the 10% and 30% quantile values is defined as low risk; the interval with lightning density between the 30% and 70% quantile values is defined as medium risk; the interval with lightning density between the 70% and 90% quantile values is defined as high risk; and the interval with lightning density above the 90% quantile value is defined as very high risk.

[0086] Please refer to Figure 4 The differentiated lightning protection strategy according to the lightning risk level is as follows:

[0087] The very low risk level corresponds to the installation of basic lightning protection facilities and surge protectors; the low risk level corresponds to the addition of overhead ground wires and the coordinated configuration of multiple surge protection levels; the medium risk level corresponds to the installation of zinc oxide arresters and the deployment of electromagnetic shielding layers; the high risk level corresponds to the differentiated configuration of multiple parallel lines and double protection devices in important sections; and the very high risk level corresponds to the laying of underground cables and the integration of intelligent online monitoring systems.

[0088] The probability distribution characteristics of the quantized lightning density by using the quantile value ensure that the division of risk levels completely follows the statistical rules driven by data. For example, the area below the 10% quantile value (extremely low risk) corresponds to the 10% probability events of the most sparse lightning activity in the historical data, while the area above the 90% quantile value (extremely high risk) represents the 10% extreme cases of the highest lightning frequency, and the intermediate quantile values are sequentially progressive, forming an objective and continuous risk progressive system. On this basis, different lightning protection strategies are matched for different risk levels: basic protection (such as surge protectors) is used for extremely low risk, overhead ground wires and multi-stage protection coordination are added for low risk, zinc oxide arresters and electromagnetic shields are deployed for medium risk, multi-loop lines and double protection devices are configured for high risk, and underground cables and intelligent monitoring systems are integrated for extremely high risk.

[0089] The above embodiments not only avoid the subjectivity of the traditional threshold method, but also accurately map the risk level through the quantile value, making the resource allocation of lightning protection measures more scientific, while taking into account the reinforcement protection of high-risk areas and the cost optimization of low-risk areas, significantly improving the safety and reliability of power grid lightning protection engineering.

[0090] It can be understood that those skilled in the art can combine various embodiments in the above embodiments to obtain technical solutions of various embodiments.

[0091] The above only describes the preferred embodiments of the present application and is not intended to limit the present application. Any modifications, equivalent replacements and improvements made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A statistical model of ground-to-ground lightning density based on gamma distribution and a lightning strike risk assessment method thereof, characterized in that: include: Collect historical ground-to-ground lightning data within a preset time range in the target area; Dividing the target area into a plurality of grid cells according to a preset spatial resolution, and determining an annual ground-to-ground flash density dataset for each grid cell based on the historical ground-to-ground flash data; The maximum likelihood estimation method is used to fit the gamma distribution of the annual CG flash density dataset, determine the shape parameters and scale parameters that characterize the statistical characteristics of the CG flash density in the target area, and construct a CG flash density statistical model; Calculating a plurality of preset probability quantile values ​​based on the statistical model of ground-to-ground lightning density, wherein the probability quantile values ​​are used to divide the probability distribution interval of ground-to-ground lightning density; The ground-to-ground lightning density is divided into multiple continuous risk intervals according to the probability quantile value, each risk interval corresponds to a lightning risk level, and the lightning risk levels include very low risk, low risk, medium risk, high risk and very high risk.

2. The gamma distribution-based ground-to-ground lightning density statistical model and the lightning strike risk assessment method thereof according to claim 1, characterized in that: The step of dividing the target area into a plurality of grid cells according to a preset spatial resolution and determining an annual ground-to-ground flash density dataset for each grid cell based on the historical ground-to-ground flash data includes: The total number of ground-to-ground flashes in each grid cell per year is counted, and the annual ground-to-ground flash frequency per unit area is calculated based on the geographical area of ​​the grid cell to generate a year-by-year ground-to-ground flash density data series for each grid cell.

3. The gamma distribution-based ground-to-ground lightning density statistical model and the lightning strike risk assessment method thereof according to claim 1, characterized in that: The method of using the maximum likelihood estimation method to perform gamma distribution fitting on the annual CG flash density dataset, determining shape parameters and scale parameters characterizing statistical characteristics of CG flash density in the target area, and constructing a CG flash density statistical model includes: Parameters of the annual ground-to-ground flash density dataset are estimated based on the probability density function of the gamma distribution to obtain shape parameters and scale parameters that characterize the statistical characteristics of the ground-to-ground flash density. The probability density function is used to describe the probability distribution characteristics of the annual ground-to-ground flash frequency per unit area in the target area.

4. The gamma distribution-based ground-to-ground lightning density statistical model and the lightning strike risk assessment method thereof according to claim 3 are characterized in that: The probability density function includes: Where: x is the density of ground-to-ground flashes, k is the shape parameter, θ is the scale parameter, and Γ(k) is the gamma function.

5. The gamma distribution-based ground-to-ground lightning density statistical model and the lightning strike risk assessment method thereof according to claim 1, characterized in that: Before calculating a plurality of preset probability quantile values ​​based on the ground-to-ground flash density statistical model, the method further includes: The matching degree between the annual flash density dataset and the fitted gamma distribution is verified by a statistical consistency test method; If the test passes, it is determined that the gamma distribution model conforms to the statistical laws of historical data.

6. The gamma distribution-based ground-to-ground lightning density statistical model and the lightning strike risk assessment method thereof according to claim 5, characterized in that: Verifying the degree of match between the annual flash density dataset and the fitted gamma distribution by using a statistical consistency test method includes: If the test fails, data transformation is performed on the annual flash density dataset; Refit the gamma distribution based on the transformed data to obtain the corrected shape parameters and scale parameters; The matching degree between the modified gamma distribution and the transformed data was verified by statistical consistency test method.

7. The gamma distribution-based ground-to-ground lightning density statistical model and the lightning strike risk assessment method thereof according to claim 5, characterized in that: The calculating of a plurality of preset probability quantile values ​​based on the statistical model of the ground-to-ground flash density, wherein the probability quantile values ​​are used to divide the probability distribution interval of the ground-to-ground flash density, includes: According to the fitted gamma distribution, the corresponding probability quantile value is calculated by the inverse function of the cumulative distribution function. The probability quantile value is the threshold value of the ground-to-ground flash density under the preset cumulative probability.

8. The gamma distribution-based ground-to-ground lightning density statistical model and lightning strike risk assessment method thereof according to claim 7, characterized in that: The preset multiple probability quantile values ​​include 10%, 30%, 70% and 90% quantile values.

9. The gamma distribution-based ground-to-ground lightning density statistical model and the lightning strike risk assessment method thereof according to claim 8, characterized in that: The dividing the ground-to-ground flash density into a plurality of continuous risk intervals according to the probability quantile value includes: The interval where the ground-to-ground flash density is below the 10th percentile is defined as extremely low risk; The flash-to-ground flash density between the 10th and 30th percentiles is defined as low risk; The interval between the 30th and 70th percentile values ​​of ground-to-ground flash density is defined as medium risk; The flash-to-ground flash density range between the 70th and 90th percentiles is defined as high risk; The interval where the ground-to-ground flash density is above the 90th percentile is defined as extremely high risk.

10. The gamma distribution-based ground-to-ground lightning density statistical model and the lightning strike risk assessment method thereof according to claim 9, characterized in that: After dividing the ground-to-ground lightning density into a plurality of continuous risk intervals according to the probability quantile value, each risk interval corresponding to a lightning risk level, the method further includes: Match the corresponding lightning protection strategy according to the lightning risk level; among them: The extremely low risk level corresponds to the installation of basic lightning protection facilities and surge protectors; Low risk level corresponds to the addition of overhead ground wires and coordinated configuration of multi-level surge protection; The medium risk level corresponds to the installation of zinc oxide lightning arresters and the deployment of electromagnetic shielding layers; High-risk levels correspond to differentiated configurations of multiple parallel lines and double sets of protection devices in important sections; The extremely high risk level corresponds to underground cable laying and intelligent online monitoring system integration.

Images