A system for measuring the electrical parameters of a lightning activity path based on spatiotemporal clustering analysis

Abstract

This invention discloses a system for measuring electrical parameters along a lightning path based on spatiotemporal clustering analysis, belonging to the field of lightning path electrical parameter measurement technology. It includes a data acquisition module for collecting lightning data, including the time, longitude, and latitude of the lightning event; an electrical parameter measurement module for measuring electrical parameters along the lightning path, including electric field strength, current intensity, and voltage; and sensors for real-time monitoring of areas along the lightning path to obtain accurate electrical parameter data. The system uses sensors to measure electric field strength, current transformers to measure current intensity, and voltage sensors to measure voltage. This invention significantly improves the accuracy of electrical parameter measurement, addressing the problem of poor accuracy. When measuring the instantaneous large current of a lightning strike, the current transformer, with its suitable range and fast response, can accurately record changes in current intensity, providing reliable data for studying lightning current characteristics.

Description

Technical Field

[0001] This invention relates to the field of electrical parameter measurement technology for lightning activity paths, specifically a lightning activity path electrical parameter measurement system based on spatiotemporal clustering analysis. Background Technology

[0002] In lightning research, accurately measuring the electrical parameters along the path of lightning activity is crucial. As a powerful natural discharge phenomenon, lightning carries electrical parameters such as electric field strength, current intensity, and voltage, which play a key role in understanding the physical mechanisms of lightning and assessing its impact on various facilities and the environment.

[0003] In response, patent CN107271793B discloses an automated lightning warning system. This system uses thunderstorm clusters as the warning unit and predicts the location and time of the next occurrence of a thunderstorm cluster based on its historical movement path and time. This avoids the tedious process of individually calculating the movement of lightning points, simplifies the calculation workload, and improves prediction efficiency. By establishing a coordinate system and combining it with statistical analysis of the historical paths of thunderstorm clusters and prediction of their next occurrence locations, the patent enhances the accuracy and intuitiveness of the predicted paths and locations.

[0004] However, existing technologies have significant shortcomings. Some electric field strength measurement methods are susceptible to environmental interference, making it difficult to guarantee the accuracy of the measurement results. External factors can severely affect the performance of the measurement sensors, causing significant deviations in the measurement data. Regarding current intensity measurement, traditional equipment may be unable to accurately capture instantaneous current changes during ground flashovers due to range limitations or insufficient response speed when measuring large currents. For voltage measurement, due to the instantaneous and complex nature of ground flashover voltages, existing voltage measurement methods struggle to accurately measure their peak values ​​and changes.

[0005] To address the aforementioned issues, a system for measuring electrical parameters of ground flash activity paths based on spatiotemporal clustering analysis is proposed. Summary of the Invention

[0006] The purpose of this invention is to provide a system for measuring electrical parameters of ground flash activity paths based on spatiotemporal clustering analysis, which solves the problem of poor accuracy in electrical parameter measurement in the prior art.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a system for measuring electrical parameters of ground flashover activity paths based on spatiotemporal clustering analysis, comprising,

[0008] The data acquisition module is used to collect lightning data, which includes the time, longitude, and latitude information of the lightning event.

[0009] The electrical parameter measurement module is used to measure the electrical parameters along the path of the lightning strike, including electric field strength, current intensity, and voltage.

[0010] The sensor monitors the area in the path of the lightning strike in real time, obtains accurate electrical parameter data, and uses the sensor to measure the electric field strength, the current intensity to measure the current intensity through the current transformer, and the voltage to measure the voltage.

[0011] In the electrical parameter measurement module, a data fusion algorithm is used to measure the electric field strength, current intensity, and voltage value.

[0012] Preferably, the electrical parameter measurement system also includes a region division module for region division. The region division is determined by analyzing the altitude and topographic features, and the region is divided into three regions: mountainous, grassland and desert.

[0013] The region division module further divides the three regions based on slope and aspect;

[0014] The slope zones are divided into extremely steep zones, steep slope zones, medium slope zones, and gentle slope zones;

[0015] The spatiotemporal database establishment module is used to construct a three-dimensional spatiotemporal database from the collected lightning data; the dimensions of the three-dimensional spatiotemporal database are determined to be time, longitude and latitude, each lightning data record is converted into an object point, and all object points are entered into the database in chronological order;

[0016] The spatiotemporal clustering analysis module is used to perform spatiotemporal clustering analysis on data in the spatiotemporal database. It includes a threshold setting unit, a core object judgment unit, a neighbor point search unit, and a cluster improvement unit.

[0017] The threshold setting unit is used to set the time distance threshold, spatial distance threshold, and spatiotemporal object quantity threshold;

[0018] Regional feature information is obtained from the regional segmentation module, and ground flash data features are extracted from the spatiotemporal database. An adaptive algorithm and a feedback-based dynamic adjustment algorithm are used to adjust the threshold, with adjustment weights w. t w s w Min-Pts The formula is as follows: w t +w s +w min-Pts min-Pts = C;

[0019] Where w t w represents the threshold weight of time distance for ground flash data points s w represents the threshold weight of spatial distance factors on ground flash data points Min-Pts The threshold weight representing the number of spatiotemporal objects in the ground flash data points, min-Pts represents the starting point of the clustering analysis, and C represents the target value of the spatiotemporal clustering analysis;

[0020] The core object determination unit is used to select an object point from all ground flash data and determine the ground flash path;

[0021] The selected object point is P = {id, x, y, t}, where id is the unique identifier of the object point, x and y are the longitude and latitude information respectively, and t is the time information. It is then determined whether the object point already belongs to an existing cluster. If it does, the next object point is selected. Otherwise, it is determined whether the object point is a spatiotemporal core object. If it is not a spatiotemporal core object, the process returns to the object point selection step to select the next object point. If it is a spatiotemporal core object, a neighboring point search is performed.

[0022] The neighboring point search unit is used to search for all spatiotemporal neighboring points of the spatiotemporal core object point. For a point determined to be a spatiotemporal core object, based on time, longitude and latitude information, in the time dimension, if the time difference between another point and the core object point is not greater than the set time distance threshold; in the spatial dimension, if the spatial distance between the two points is not greater than the set spatial distance threshold, then the point is identified as a spatiotemporal neighboring point.

[0023] The cluster improvement unit is used to determine whether ground flash data is a spatiotemporal core object. Based on the time distance threshold, spatial distance threshold, and spatiotemporal object quantity threshold, it determines whether the number of neighbors within a given range is not less than the spatiotemporal object quantity threshold. The given range is a time period in the time dimension where the time difference is not greater than the time distance threshold, and in the spatial dimension, it is a region where the spatial distance is not greater than the spatial distance threshold.

[0024] Preferably, in the data acquisition module, a distributed sensor network is used to acquire ground flash data. The acquired data is preprocessed in real time, including outlier detection and removal. The Raida criterion is used to judge outliers, that is, if the difference between a data point and the mean is greater than three times the standard deviation, it is identified as an outlier and removed.

[0025] Preferably, in mountainous areas, slopes greater than 60° are considered extremely steep areas, slopes between 45° and 60° are considered steep slope areas, and slopes between 30° and 45° are considered moderate slope areas; slopes less than 20° are considered gentle slope areas; slope aspect is used to analyze the impact of vegetation cover and soil moisture differences on the path of ground flashes, and slope aspect is divided into sunny slopes and shady slopes;

[0026] In grasslands, slope is divided into steep slopes with a slope greater than 10° and gentle slopes with a slope less than 10°. Aspect is also divided to analyze the impact of vegetation cover and soil moisture differences on the path of ground flashes. Aspect is divided into sunny slopes and shady slopes.

[0027] In the desert, dune slopes greater than 30° are considered steep slopes, while those less than 30° are considered gentle slopes. Slope aspect is used to classify dunes: the windward side is the sunny slope, and the leeward side is the shady slope. This is used to analyze the impact of differences in sand compaction and moisture content on the path of lightning strikes.

[0028] Preferably, in the region division module, for further division of mountainous areas, a topographic relief index is introduced, and the formula for calculating topographic relief is: Where H max H is the highest point in the region. min D is the lowest elevation in the region, D is the difference between the average elevation in the region and the average elevation in the neighboring area, and S is the area of ​​the region. Based on the degree of topographic relief, the mountainous area is subdivided into high-relief mountainous area, medium-relief mountainous area and low-relief mountainous area. The lightning activity path in mountainous areas with different degrees of relief may be affected by the terrain to different degrees.

[0029] Preferably, in the core object judgment unit, time series analysis is added to the judgment of the ground flash path. An autoregressive moving average model is used to model and predict the time series data. If the deviation between the predicted value and the actual value is within a certain range, it is considered that the point may belong to the same ground flash path. Let the time series of the selected object point be {t1, t2, ..., t...} n The model formula is:

[0030] in For autoregressive coefficients, θ j The moving average coefficient, ∈ t is a white noise sequence, where P is the average distance from the ground flash data point to other ground flash data points in the same terrain area, and q is the average distance from the ground flash data point to the nearest ground flash data point in a different terrain area.

[0031] Preferably, in the neighboring point search unit, a KD-tree data structure is used to search for spatial neighboring points. For a given spatiotemporal core object point, neighboring points that meet the spatial distance threshold range are found in the KD-tree.

[0032] Preferably, in the cluster improvement unit, a fuzzy clustering algorithm is introduced. For newly added object points, their affiliation is determined based on their fuzzy membership degree with existing cluster centers. Let the object point P = {id, x, y, t}, and the existing cluster centers be C = {x, t}. c ,y c ,t c The formula for calculating fuzzy membership degree is:

[0033] Where u ij Let d be the model membership degree of object point i belonging to cluster j. ijLet d be the distance from object point i to cluster center j, C be the number of clusters, and m be the fuzzy index. The fuzzy membership degree is used to determine whether an object point is a core object and its corresponding cluster, where d... kj The distance is the Euclidean distance from the k-th lightning data object point to the j-th cluster center. The Euclidean distance is calculated based on the longitude and latitude data of the object point in the spatiotemporal database.

[0034] Preferably, in the region partitioning module, the effectiveness of the region partitioning results is evaluated using a clustering evaluation index coefficient. The data point P = {id, x, y, t}, a(P) is the average distance from point P to other points within the same region, and the silhouette coefficient is... Where s(P) is the profile coefficient of the ground flash data point P, a(P) is the average distance of the ground flash data point P to the ground flash data points in the same terrain area, and b(P) is the average distance of the ground flash data point P to the nearest ground flash data points in different terrain areas. The average value of the profile coefficients of all data points is calculated to obtain the evaluation value of the entire region division result.

[0035] Preferably, the formula for calculating the information entropy of electric field strength measurements is as follows: in Similarly, the information entropy of the current intensity and voltage measurements can be obtained, where H(E) represents the information entropy of the electric field intensity measurements, n is the number of electric field intensity measurements, and E i Let p(E) be the measured value of the i-th electric field intensity. i E represents the proportion of the i-th electric field strength measurement in the sum of all measurements. j Let j be the measured value of the j-th electric field intensity; calculate the weighting coefficient based on the information entropy, and the formula for calculating the weighting coefficient is: Where ωE i Represents the measured value of electric field strength E i The corresponding weighting coefficients, H(E) i E is the measured value of the electric field strength. i The information entropy is given by n, which is the number of electric field strength measurements, and k is the index used for summation. The denominator is calculated by iterating through all electric field strength measurements from 1 to n. The weighting coefficients for current strength and voltage are calculated similarly. The determined weighting coefficients are then weighted and averaged to obtain the combined value of electric field strength, current strength, and voltage.

[0036] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0037] This invention provides a system for measuring electrical parameters along a lightning strike path based on spatiotemporal clustering analysis. It utilizes sensors to measure electric field strength, current transformers to measure current intensity, and voltage sensors to measure voltage. Different types of sensors are adjusted to suit the characteristics of their respective electrical parameters, enabling more accurate capture of electric field, current, and voltage signals along the lightning strike path. Compared to traditional single or poorly versatile measuring devices, this significantly improves measurement accuracy. When measuring the instantaneous large current of a lightning strike, the current transformer, with its suitable range and rapid response, can accurately record changes in current intensity, providing reliable data for studying lightning current characteristics. Furthermore, the electrical parameter measurement module employs a data fusion algorithm, combining the information entropy of the measured electric field strength, current intensity, and voltage values ​​to calculate weighting coefficients. This effectively eliminates errors and uncertainties inherent in individual measurements, improving the reliability and stability of the measurement results. Attached Figure Description

[0038] Figure 1 This is a schematic diagram of the measurement system of the present invention;

[0039] Figure 2 This is a schematic diagram of the spatiotemporal clustering analysis module of the present invention;

[0040] Figure 3 This is a block diagram of the core object judgment logic of the present invention. Detailed Implementation

[0041] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0042] To further understand the content of this invention, a detailed description of the invention will be provided in conjunction with the accompanying drawings.

[0043] Combination Figures 1-3 The present invention provides a system for measuring electrical parameters of ground flash activity paths based on spatiotemporal clustering analysis, comprising:

[0044] The data acquisition module is used to collect lightning data, which includes the time, longitude, and latitude information of the lightning event.

[0045] In the data acquisition module, a distributed sensor network is used to collect ground flash data. The collected data is preprocessed in real time, including outlier detection and removal. The Raida criterion is used to judge outliers, that is, if the difference between a data point and the mean is greater than three times the standard deviation, it is considered an outlier and removed.

[0046] This approach addresses the issues of incomplete and inaccurate lightning data collection. Furthermore, outlier detection and removal ensure data accuracy, reducing interference from abnormal data in subsequent analysis and providing a high-quality data foundation for later regional division, spatiotemporal clustering analysis, and electrical parameter measurements. Accurate and comprehensive data facilitates more precise analysis of lightning activity paths, improves understanding of lightning activity patterns, and thus provides a more reliable basis for developing lightning protection and disaster early warning measures.

[0047] The electrical parameter measurement module is used to measure the electrical parameters along the path of the lightning strike, including electric field strength, current intensity, and voltage.

[0048] The sensors monitor the area along the path of the lightning strike in real time, acquiring accurate electrical parameter data. The sensors measure the electric field strength, the current intensity is measured using a current transformer, and the voltage is measured using a voltage sensor. In the electrical parameter measurement module, a data fusion algorithm is used to measure the electric field strength, current intensity, and voltage values. The information entropy calculation formula for the measured electric field strength value is as follows: in Similarly, the information entropy of the current intensity and voltage measurements can be obtained, where H(E) represents the information entropy of the electric field intensity measurements, n is the number of electric field intensity measurements, and E i Let p(E) be the measured value of the i-th electric field intensity. i E represents the proportion of the i-th electric field strength measurement in the sum of all measurements. j Let j be the measured value of the j-th electric field intensity; calculate the weighting coefficient based on the information entropy, and the formula for calculating the weighting coefficient is: Where ωE i Represents the measured value of electric field strength E i The corresponding weighting coefficients, H(E) i E is the measured value of the electric field strength. i The information entropy is given by n, which is the number of electric field strength measurements, and k is the index used for summation. The denominator is calculated by iterating through all electric field strength measurements from 1 to n. The weighting coefficients for current strength and voltage are calculated similarly. The determined weighting coefficients are then weighted and averaged to obtain the combined value of electric field strength, current strength, and voltage.

[0049] The electrical parameter measurement system of the present invention also includes: a region division module, used for region division. The region division determines the region boundary by analyzing the altitude and topographic features, and divides the region into three regions: mountainous area, grassland and desert.

[0050] The region division module further divides the three regions based on slope and aspect;

[0051] The slope areas are divided into extremely steep areas, steep slope areas, medium slope areas, and gentle slope areas;

[0052] By combining geographic information system (GIS) technology and utilizing its high-resolution topographic data, vegetation cover data, and soil type information, regional boundaries can be further precisely determined and regional divisions refined. This solves the problem of coarse regional divisions in existing technologies, which fail to fully consider the impact of different topographic conditions on lightning activity. Detailed regional divisions enable a better study of the patterns and electrical parameter variations of lightning activity within different topographic regions.

[0053] In mountainous areas, slopes greater than 60° are considered extremely steep, slopes between 45° and 60° are considered steep slopes, and slopes between 30° and 45° are considered moderate slopes; slopes less than 20° are considered gentle slopes. Slope aspect is used to analyze the impact of vegetation cover and soil moisture differences on the path of ground flashes. Slope aspect is divided into sunny slopes and shady slopes.

[0054] In grasslands, slope is divided into steep slopes with a slope greater than 10° and gentle slopes with a slope less than 10°. Aspect is also divided to analyze the impact of vegetation cover and soil moisture differences on the path of ground flashes. Aspect is divided into sunny slopes and shady slopes.

[0055] In the desert, dune slopes greater than 30° are considered steep slopes, while dune slopes less than 30° are considered gentle slopes. Slope aspect is used to classify dunes: the windward side is the sunny slope, and the leeward side is the shady slope. This is used to analyze the impact of differences in sand compaction and moisture content on the path of ground lightning.

[0056] In the regional division module, for the further division of mountainous areas, a topographic relief index is introduced. The formula for calculating topographic relief is: Where H max H is the highest point in the region. min D is the lowest elevation in the region, D is the difference between the average elevation in the region and the average elevation in the neighboring area, and S is the area of ​​the region. Based on the degree of topographic relief, the mountainous area is subdivided into high-relief mountainous area, medium-relief mountainous area and low-relief mountainous area. The lightning activity path in mountainous areas with different degrees of relief may be affected by the terrain to different degrees.

[0057] The lightning return stroke data of the three regions were analyzed separately. The analysis parameters included the number of lightning return strokes, return stroke interval, duration, return stroke distance, return stroke intensity, and number of channel groundings. Through comparative analysis, the typical characteristics of the lightning return stroke data of each region were determined. Statistical analysis methods were used to separately statistically analyze and compare the data of each region, and to calculate the average value, standard deviation, and other statistical indicators of each parameter in order to determine the typical characteristics.

[0058] Ground Flashback Count: Directly count the total number of ground flashbacks in this area, denoted as N;

[0059] Return stroke interval: Calculate the time interval between two consecutive ground flash return strokes, and then average the intervals for all return strokes. and standard deviation σ I ;

[0060] Duration: Calculate the average duration of each ground flash attack. and standard deviation σ T ;

[0061] Return stroke spacing: The spatial distance between two adjacent ground flash return strokes is measured and the average value is calculated. and standard deviation σ D ;

[0062] Return stroke intensity: The intensity value of the ground flash return stroke is obtained by measuring equipment, and its average value is calculated. and standard deviation σ S ;

[0063] Channel grounding count: Count the number of channel grounding hits during each ground flashback and calculate the average value. and standard deviation σ G .

[0064] The statistical indicators of various parameters in the three regions were compared and a table was created to visually show the differences in various parameters in the three regions, and to analyze the relationship between the terrain features of different regions and the typical characteristics of ground flash return data.

[0065] Formula for calculating the average value:

[0066] For parameter X (such as stroke interval, duration, etc.), its average value Where n is the number of data points, X i Let i be the value of the i-th data point;

[0067] Formula for calculating standard deviation:

[0068] For parameter X, its standard deviation By using statistical analysis steps and algorithmic formulas, the typical characteristics of lightning return stroke data in each region can be determined, providing a basis for further research on the patterns of lightning activity in different terrain regions.

[0069] The spatiotemporal database establishment module is used to construct a three-dimensional spatiotemporal database from the collected lightning data; the dimensions of the three-dimensional spatiotemporal database are determined to be time, longitude and latitude, each lightning data record is converted into an object point, and all object points are entered into the database in chronological order;

[0070] The spatiotemporal clustering analysis module is used to perform spatiotemporal clustering analysis on data in the spatiotemporal database. It includes a threshold setting unit, a core object judgment unit, a neighbor point search unit, and a cluster improvement unit.

[0071] The threshold setting unit is used to set the time distance threshold, spatial distance threshold, and spatiotemporal object quantity threshold;

[0072] Regional feature information is obtained from the regional segmentation module, and ground flash data features are extracted from the spatiotemporal database. An adaptive algorithm and a feedback-based dynamic adjustment algorithm are used to adjust the threshold, with adjustment weights w. t w s w Min-Pts The formula is as follows: w t +w s +w min-Pts min-Pts = C;

[0073] Where w t w represents the threshold weight of time distance for ground flash data points s w represents the threshold weight of spatial distance factors on ground flash data points Min-Pts The threshold weight representing the number of spatiotemporal objects in the ground flash data points, min-Pts represents the starting point of the clustering analysis, and C represents the target value of the spatiotemporal clustering analysis;

[0074] The core object determination unit is used to select an object point from all ground flash data and determine the ground flash path;

[0075] In the core object judgment unit, time series analysis is added to the judgment of the lightning path. An autoregressive moving average model is used to model and predict the time series data. If the deviation between the predicted value and the actual value is within a certain range, it is considered that the point may belong to the same lightning path. Let the time series of the selected object point be {t1, t2, ..., t...} n The model formula is:

[0076] in For autoregressive coefficients, θ j The moving average coefficient, ∈ t is a white noise sequence, where P is the average distance from the ground flash data point to other ground flash data points in the same terrain area, and q is the average distance from the ground flash data point to the nearest ground flash data point in a different terrain area.

[0077] The selected object point is P = {id, x, y, t}, where id is the unique identifier of the object point, x and y are the longitude and latitude information respectively, and t is the time information. It is then determined whether the object point already belongs to an existing cluster. If it does, the next object point is selected. Otherwise, it is determined whether the object point is a spatiotemporal core object. If it is not a spatiotemporal core object, the process returns to the object point selection step to select the next object point. If it is a spatiotemporal core object, a neighboring point search is performed.

[0078] The neighboring point search unit is used to search for all spatiotemporal neighboring points of the spatiotemporal core object point. For a point determined to be a spatiotemporal core object, based on time, longitude and latitude information, in the time dimension, if the time difference between another point and the core object point is not greater than the set time distance threshold; in the spatial dimension, if the spatial distance between the two points is not greater than the set spatial distance threshold, then the point is identified as a spatiotemporal neighboring point.

[0079] In the neighboring point search unit, a KD-tree data structure is used to search for spatial neighboring points. For a given spatiotemporal core object point, neighboring points that meet the spatial distance threshold range are found in the KD-tree.

[0080] The cluster refinement unit is used to determine whether ground flash data is a spatiotemporal core object. Based on time distance thresholds, spatial distance thresholds, and spatiotemporal object quantity thresholds, it determines whether the number of neighbors within a given range is not less than the spatiotemporal object quantity threshold. The given range, in the time dimension, is a time period where the time difference is no greater than the time distance threshold; in the spatial dimension, it is a region where the spatial distance is no greater than the spatial distance threshold. The cluster refinement unit introduces a fuzzy clustering algorithm. For newly added object points, their affiliation is determined based on their fuzzy membership degree with existing cluster centers. Let object point P = {id, x, y, t}, and existing cluster centers be C = {x, y, t}. c ,y c ,t c The formula for calculating fuzzy membership degree is:

[0081] Where u ij Let d be the model membership degree of object point i belonging to cluster j. ij Let d be the distance from object point i to cluster center j, C be the number of clusters, and m be the fuzzy index. The fuzzy membership degree is used to determine whether an object point is a core object and its corresponding cluster, where d... kj The Euclidean distance is the distance from the k-th lightning data object point to the j-th cluster center. The Euclidean distance is calculated based on the longitude and latitude data of the object points in the spatiotemporal database.

[0082] In the region partitioning module, the effectiveness of the region partitioning results is evaluated using a clustering evaluation index coefficient. Data point P = {id, x, y, t}, a(P) is the average distance from point P to other points within the same region, and the silhouette coefficient is... Where s(P) is the profile coefficient of the ground flash data point P, a(P) is the average distance of the ground flash data point P to the ground flash data points in the same terrain area, and b(P) is the average distance of the ground flash data point P to the nearest ground flash data points in different terrain areas. The average value of the profile coefficients of all data points is calculated to obtain the evaluation value of the entire region division result.

[0083] Working principle:

[0084] First, the data acquisition module uses a distributed sensor network to collect lightning data based on different terrain features. The collected data is preprocessed using the Laida criterion to remove outliers, providing an accurate data foundation for subsequent analysis. Next, the region division module, based on factors such as altitude, topographic features, slope, and aspect, and combined with Geographic Information System (GIS) technology, finely divides the region to identify different terrain areas, providing regional characteristic information for spatiotemporal clustering analysis. Then, the spatiotemporal database construction module builds a three-dimensional spatiotemporal database from the collected lightning data, converting the data into object point format and entering it chronologically. Simultaneously, it performs labeling and classification and employs data compression technology for efficient data storage and retrieval, providing data support for spatiotemporal clustering analysis. Next, the spatiotemporal clustering analysis module begins operation. The threshold setting unit obtains information from the region division module and the spatiotemporal database, and adjusts the threshold using an adaptive algorithm combined with machine learning technology. The core object identification unit selects object points from the data and determines whether they belong to existing clusters. If not, it determines whether they are spatiotemporal core objects, incorporating time series analysis during the process. If it is a spatiotemporal core object, the neighbor point search unit uses it as a benchmark and searches for neighboring points that meet the conditions in both time and space using a KD-tree data structure. The cluster improvement unit determines the affiliation of newly added object points based on the threshold and combined with a fuzzy clustering algorithm, thereby accurately identifying the lightning path and core objects. Finally, the electrical parameter measurement module uses sensors to monitor the area along the lightning activity path in real time, acquiring electrical parameter data. Different electrical parameters are measured using specific sensors, and a data fusion algorithm is introduced to improve measurement accuracy. Combined with wireless sensor network technology, real-time remote monitoring and data transmission are achieved, providing accurate and timely electrical parameter information for lightning protection and disaster early warning.

[0085] Through comprehensive data acquisition, the use of a distributed sensor network, and real-time preprocessing, the accuracy and reliability of the data are ensured. Fine-grained regional division is achieved by subdividing regions based on altitude and topographic features, considering the impact of different terrain conditions on lightning activity paths and electrical parameters. Accurate spatiotemporal clustering analysis is employed, with each unit collaboratively using an adaptive algorithm to adjust thresholds and determine lightning paths and core objects. Precise electrical parameter measurement utilizes real-time sensor monitoring and incorporates data fusion algorithms to improve accuracy. Furthermore, innovations are made in regional division and core object identification, resulting in significant advantages in accuracy, reliability, and practicality compared to existing technologies.

[0086] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0087] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A system for measuring electrical parameters of ground flashover activity paths based on spatiotemporal clustering analysis, characterized in that, include, The data acquisition module is used to collect lightning data, which includes the time, longitude, and latitude information of the lightning event. The electrical parameter measurement module is used to measure the electrical parameters along the path of the lightning strike, including electric field strength, current intensity, and voltage. The sensor monitors the area in the path of the lightning strike in real time, obtains accurate electrical parameter data, and uses the sensor to measure the electric field strength, the current intensity to measure the current intensity through the current transformer, and the voltage to measure the voltage. In the electrical parameter measurement module, a data fusion algorithm is used to measure the electric field strength, current intensity, and voltage value; The electrical parameter measurement system also includes a region division module, which is used for region division. The region division is determined by analyzing the altitude and topographic features, and the region is divided into three regions: mountainous, grassland and desert. The region division module further divides the three regions based on slope and aspect; The slope areas are divided into extremely steep areas, steep slope areas, medium slope areas, and gentle slope areas; The spatiotemporal database establishment module is used to construct a three-dimensional spatiotemporal database from the collected lightning data; the dimensions of the three-dimensional spatiotemporal database are determined to be time, longitude and latitude, each lightning data record is converted into an object point, and all object points are entered into the database in chronological order; The spatiotemporal clustering analysis module is used to perform spatiotemporal clustering analysis on data in the spatiotemporal database. It includes a threshold setting unit, a core object judgment unit, a neighbor point search unit, and a cluster improvement unit. The threshold setting unit is used to set the time distance threshold, spatial distance threshold, and spatiotemporal object quantity threshold; Regional feature information is obtained from the regional segmentation module, and ground flash data features are extracted from the spatiotemporal database. An adaptive algorithm and a feedback-based dynamic adjustment algorithm are used to adjust the threshold, with adjustment weights w. t w s w Min-Pts The formula is as follows: w t +w s +w min-Pts min-Pts = C; Where w t w represents the threshold weight of time distance for ground flash data points s w represents the threshold weight of spatial distance factors on ground flash data points Min-Pts The threshold weight representing the number of spatiotemporal objects in the ground flash data points, min-Pts represents the starting point of the clustering analysis, and C represents the target value of the spatiotemporal clustering analysis; The core object determination unit is used to select an object point from all ground flash data and determine the ground flash path; The selected object point is P = {id, x, y, t}, where id is the unique identifier of the object point, x and y are the longitude and latitude information respectively, and t is the time information. It is then determined whether the object point already belongs to an existing cluster. If it does, the next object point is selected. Otherwise, it is determined whether the object point is a spatiotemporal core object. If it is not a spatiotemporal core object, the process returns to the object point selection step to select the next object point. If it is a spatiotemporal core object, a neighboring point search is performed. The neighboring point search unit is used to search for all spatiotemporal neighboring points of the spatiotemporal core object point. For a point determined to be a spatiotemporal core object, based on time, longitude and latitude information, in the time dimension, if the time difference between another point and the core object point is not greater than the set time distance threshold; in the spatial dimension, if the spatial distance between the two points is not greater than the set spatial distance threshold, then the point is identified as a spatiotemporal neighboring point. The cluster improvement unit is used to determine whether the ground flash data is a spatiotemporal core object. Based on the time distance threshold, spatial distance threshold, and spatiotemporal object quantity threshold, it determines whether the number of neighbors within a given range is not less than the spatiotemporal object quantity threshold. The given range is a time period in the time dimension where the time difference is not greater than the time distance threshold, and in the spatial dimension, it is a region where the spatial distance is not greater than the spatial distance threshold. In the neighbor search unit, a KD-tree data structure is used for spatial neighbor search. For a given spatiotemporal core object point, neighboring points that meet the spatial distance threshold are found in the KD-tree. In the cluster improvement unit, a fuzzy clustering algorithm is introduced. For newly added object points, their affiliation is determined based on their fuzzy membership degree with existing cluster centers. Let object point P = {id, x, y, t}, and existing cluster centers be C = {x, y, t}. c ,y c ,t c The formula for calculating fuzzy membership degree is: Where u ij Let d be the model membership degree of object point i belonging to cluster j. ij Let d be the distance from object point i to cluster center j, C be the number of clusters, and m be the fuzzy index. The fuzzy membership degree is used to determine whether an object point is a core object and its corresponding cluster, where d... kj The distance is the Euclidean distance from the k-th lightning data object point to the j-th cluster center. The Euclidean distance is calculated based on the longitude and latitude data of the object point in the spatiotemporal database.

2. The system for measuring electrical parameters of a ground flash activity path based on spatiotemporal clustering analysis according to claim 1, characterized in that: In the data acquisition module, a distributed sensor network is used to collect ground flash data. The collected data is preprocessed in real time, including outlier detection and removal. The Raida criterion is used to judge outliers, that is, if the difference between a data point and the mean is greater than three times the standard deviation, it is considered an outlier and removed.

3. The system for measuring electrical parameters of ground flashover activity paths based on spatiotemporal clustering analysis according to claim 1, characterized in that: In mountainous areas, slopes greater than 60° are considered extremely steep, slopes between 45° and 60° are considered steep slopes, and slopes between 30° and 45° are considered moderate slopes; slopes less than 20° are considered gentle slopes. Slope aspect is used to analyze the impact of vegetation cover and soil moisture differences on the path of ground flashes. Slope aspect is divided into sunny slopes and shady slopes. In grasslands, slope is divided into steep slopes with a slope greater than 10° and gentle slopes with a slope less than 10°. Aspect is also divided to analyze the impact of vegetation cover and soil moisture differences on the path of ground flashes. Aspect is divided into sunny slopes and shady slopes. In the desert, dune slopes greater than 30° are considered steep slopes, while those less than 30° are considered gentle slopes. Slope aspect is used to classify dunes: the windward side is the sunny slope, and the leeward side is the shady slope. This is used to analyze the impact of differences in sand compaction and moisture content on the path of lightning strikes.

4. The system for measuring electrical parameters of ground flashover activity path based on spatiotemporal clustering analysis according to claim 3, characterized in that: In the regional division module, for the further division of mountainous areas, a topographic relief index is introduced. The formula for calculating topographic relief is: Where H max H is the highest point in the region. min D is the lowest elevation in the region, D is the difference between the average elevation in the region and the average elevation in the neighboring area, and S is the area of ​​the region. Based on the degree of topographic relief, the mountainous area is subdivided into high-relief mountainous area, medium-relief mountainous area and low-relief mountainous area. The lightning activity path in mountainous areas with different degrees of relief may be affected by the terrain to different degrees.

5. The system for measuring electrical parameters of ground flashover activity paths based on spatiotemporal clustering analysis according to claim 1, characterized in that: In the core object judgment unit, time series analysis is added to the judgment of the lightning path. An autoregressive moving average model is used to model and predict the time series data. If the deviation between the predicted value and the actual value is within a certain range, it is considered that the point may belong to the same lightning path. Let the time series of the selected object point be {t1, t2, ..., t...} n The model formula is: in For autoregressive coefficients, θ j The moving average coefficient, ∈ t is a white noise sequence, where P is the average distance from the ground flash data point to other ground flash data points in the same terrain area, and q is the average distance from the ground flash data point to the nearest ground flash data point in a different terrain area.

6. The system for measuring electrical parameters of a ground flash activity path based on spatiotemporal clustering analysis according to claim 4, characterized in that: In the region partitioning module, the effectiveness of the region partitioning results is evaluated using a clustering evaluation index coefficient. Data point P = {id, x, y, t}, a(P) is the average distance from point P to other points within the same region, and the silhouette coefficient is... Where s(P) is the profile coefficient of the ground flash data point P, a(P) is the average distance of the ground flash data point P to the ground flash data points in the same terrain area, and b(P) is the average distance of the ground flash data point P to the nearest ground flash data points in different terrain areas. The average value of the profile coefficients of all data points is calculated to obtain the evaluation value of the entire region division result.

7. The system for measuring electrical parameters of a ground flash activity path based on spatiotemporal clustering analysis according to claim 1, characterized in that: The formula for calculating the information entropy of electric field strength measurements is as follows: in Similarly, the information entropy of the current intensity and voltage measurements can be obtained, where H(E) represents the information entropy of the electric field intensity measurements, n is the number of electric field intensity measurements, and E i Let p(E) be the measured value of the i-th electric field intensity. i E represents the proportion of the i-th electric field strength measurement in the sum of all measurements. j Let j be the measured value of the j-th electric field intensity; calculate the weighting coefficient based on the information entropy, and the formula for calculating the weighting coefficient is: Where ωE i Represents the measured value of electric field strength E i The corresponding weighting coefficients, H(E) i E is the measured value of the electric field strength. i The information entropy is given by n, which is the number of electric field strength measurements, and k is the index used for summation. The denominator is calculated by iterating through all electric field strength measurements from 1 to n. The weighting coefficients for current strength and voltage are calculated similarly. The determined weighting coefficients are then weighted and averaged to obtain the combined value of electric field strength, current strength, and voltage.

Images