Power grid multi-disaster coupling strength calculation method and system

Abstract

The invention provides a power grid multi-disaster coupling strength calculation method and system, and relates to the technical field of power emergency, and the method comprises the steps: collecting the weather forecast data of a target power transmission channel and the real-time operation parameters of a power grid; analyzing weather forecast data, and configuring virtual monitoring points at the midpoint of an ice-prone section of a line, the tower top of a tower in a thunder and lightning frequent area and the geometric center of a tower foundation in a soil frost heaving sensitive area; generating a dynamic feature point set based on the virtual monitoring points; analyzing real-time operation parameters of a power grid, and collecting an equipment state feature point set at monitoring points at four corners of a tower foundation platform, a joint of an insulator steel foot and a hanging point and an outlet of a wire strain clamp; and extracting a data point sequence with time-space continuous evolution characteristics in the dynamic characteristic point set as a core trajectory for constructing the three-dimensional dynamic ellipsoid. The method can improve the dynamic nature and accuracy of power grid risk assessment under the disaster chain effect, and guarantees the safe operation of a power transmission channel.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power emergency, in particular to a power grid multi-disaster coupling strength calculation method and system. BACKGROUND

[0002] The power grid is the core infrastructure of energy security supply. The target power transmission channel often faces the superimposed influence of icing, lightning, soil frost heaving and other disasters. These disasters do not occur in isolation, but form a disaster chain that continuously acts on power grid equipment. The coupling effect will exacerbate the risk of equipment damage and even cause large-scale power outages.

[0003] To ensure the safe operation of the power grid, existing technologies have carried out related research on multi-disaster monitoring and coupling strength calculation. However, the existing technology has the following technical defects, that is, it does not establish a three-dimensional spatial coupling relationship between the dynamic evolution characteristics of multi-disasters and the stable state characteristics of power grid equipment, and it does not integrate the vulnerability factors in the whole life cycle of the equipment. For example, the evolution process of multi-disasters has the characteristics of space-time continuity, and the health status of the power grid equipment is represented as a stable parameter cluster. The existing method only performs surface fusion of disaster data and equipment data, without deep coupling of disaster evolution trajectory and equipment state framework in the spatial geometric level. At the same time, the vulnerability changes in the whole life cycle of the equipment are not included in the coupling strength calculation, which makes the model unable to reflect the state difference of the equipment after long-term operation.

[0004] The limitations result in a lack of dynamicity and accuracy in the coupling strength calculation results, which cannot truly reflect the actual risk level of the power grid under the action of the disaster chain, and thus the emergency control decision instructions lack pertinence and are difficult to effectively resist the safety hazards caused by multi-disaster coupling. SUMMARY

[0005] The technical problem to be solved by the present application is to provide a power grid multi-disaster coupling strength calculation method, which can improve the dynamicity and accuracy of power grid risk assessment under the action of the disaster chain and provide reliable technical support for the safe operation of the power transmission channel.

[0006] To solve the above technical problems, the technical solution of the present application is as follows: In a first aspect, a power grid multi-disaster coupling strength calculation method is provided, the method comprising: Collecting meteorological forecast data and real-time operation parameters of the target power transmission channel; Analyzing the meteorological forecast data, configuring virtual monitoring points at the midpoint of the icing section of the line, the top of the tower in the lightning-prone area, and the geometric center of the tower foundation in the soil frost heaving sensitive area, and generating a dynamic feature point set based on the virtual monitoring points; Analyzing the real-time operation parameters of the power grid, collecting a device state feature point set at the four corners of the tower foundation pile cap, the connection between the insulator steel foot and the hanging point, and the outlet of the conductor tension clamp; Extract data point sequences with spatiotemporal continuous evolution characteristics from the dynamic feature point set as the core trajectory for constructing a three-dimensional dynamic ellipsoid; extract data point clusters reflecting the stable and healthy state of the equipment from the equipment status feature point set as the benchmark framework for constructing a three-dimensional polyhedron; construct a three-dimensional dynamic ellipsoid based on the core trajectory, and construct a three-dimensional polyhedron based on the benchmark framework; Calculate the spatial coupling intersection of a three-dimensional dynamic ellipsoid and a three-dimensional polyhedron to generate coupling parameters characterizing the correlation between disaster characteristics and equipment status; based on the coupling parameters, construct a two-dimensional correlation model that integrates the dynamic coupling characteristics of multiple disasters and the vulnerability of equipment throughout its entire life cycle. The two-dimensional correlation model is invoked to calculate the dynamic coupling strength under the action of disaster chain by iteratively fusing meteorological forecast data and real-time power grid operating parameters in real time, and to generate dynamic coupling strength calculation results. Based on the dynamic coupling strength calculation results, an emergency control decision instruction set for the target power transmission channel is generated.

[0007] Secondly, a power grid multi-hazard coupling strength calculation system, which executes the method described above, including: The data acquisition module is used to collect meteorological forecast data and real-time power grid operating parameters of the target power transmission channel; The parsing configuration module is used to parse meteorological forecast data and configure virtual monitoring points at the midpoint of the line section prone to icing, the top of the tower in the lightning-prone area, and the geometric center of the tower foundation in the soil frost heave sensitive area; and generate a dynamic feature point set based on the virtual monitoring points. To analyze real-time power grid operating parameters, a set of equipment status characteristic points is collected at the four corner monitoring points of the tower foundation, the connection between the insulator steel foot and the hanging point, and the outlet of the conductor tension clamp. The point set extraction module is used to extract data point sequences with spatiotemporal continuous evolution characteristics from the dynamic feature point set, which serve as the core trajectory for constructing a three-dimensional dynamic ellipsoid; it also extracts data point clusters from the equipment status feature point set that reflect the stable and healthy state of the equipment, which serve as the reference framework for constructing a three-dimensional polyhedron; the module constructs a three-dimensional dynamic ellipsoid based on the core trajectory, and then constructs a three-dimensional polyhedron based on the reference framework. The model building module is used to calculate the spatial coupling intersection of a three-dimensional dynamic ellipsoid and a three-dimensional polyhedron, and generate coupling parameters that characterize the correlation between disaster characteristics and equipment status. Based on the coupling parameters, a two-dimensional correlation model that integrates the dynamic coupling characteristics of multiple disasters and the vulnerability of equipment throughout its entire life cycle is constructed. The model invocation module is used to invoke the two-dimensional correlation model, which calculates the dynamic coupling strength under the action of disaster chain by real-time iterative fusion of meteorological forecast data and real-time power grid operating parameters, and generates dynamic coupling strength calculation results. The decision generation module is used to generate an emergency control decision instruction set for the target power transmission channel based on the dynamic coupling strength calculation results.

[0008] Thirdly, a computing device including a memory and a processor; The memory stores one or more computer programs, the one or more computer programs including instructions; when the instructions are executed by the processor, the computing device performs the method as described in the first aspect.

[0009] Fourthly, a computer-readable storage medium for storing a computer program for performing the method as described in the first aspect.

[0010] The above-described solution of the present invention has at least the following beneficial effects: By employing techniques such as acquiring meteorological forecast data and real-time power grid operating parameters, constructing dynamic feature point sets and equipment status feature point sets, constructing three-dimensional dynamic ellipsoids and three-dimensional polyhedra and calculating their spatial coupling intersection, building a two-dimensional correlation model that integrates the dynamic coupling characteristics of multiple disasters and the vulnerability of equipment throughout its entire life cycle, and using techniques for real-time iterative calculation of dynamic coupling strength and parameterized adjustment of emergency plans, this approach overcomes the problems of existing technologies that fail to establish a deep three-dimensional spatial coupling between multiple disasters and equipment status, and fail to integrate the vulnerability of equipment throughout its entire life cycle. It also overcomes the problems of insufficient dynamism and accuracy in coupling strength calculation and insufficient targeting of emergency control decisions. This improves the dynamism and accuracy of power grid risk assessment under disaster chains, enhances the adaptability of emergency control commands, effectively reduces safety hazards caused by multi-disaster coupling, and ensures the safe and stable operation of the target transmission channel. Attached Figure Description

[0011] Figure 1 A schematic diagram of the calculation method for the coupling strength of multiple disasters in a power grid; Figure 2 This is a schematic diagram of a power grid multi-hazard coupling strength calculation system. Figure 3 This is a schematic diagram of a computing device. Detailed Implementation

[0012] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0013] Embodiments of the present invention propose a method for calculating the coupling strength of multiple disasters in a power grid, such as... Figure 1 The diagram shown illustrates the process of this method, which includes: Step 1: Collect meteorological forecast data and real-time power grid operating parameters for the target transmission channel; Step 2: Analyze meteorological forecast data and configure virtual monitoring points at the midpoint of the section of the line prone to icing, the top of the tower in the lightning-prone area, and the geometric center of the tower foundation in the soil frost heave sensitive area; generate a dynamic feature point set based on the virtual monitoring points; Step 3: Analyze the real-time operating parameters of the power grid and collect a set of equipment status characteristic points at the four corner monitoring points of the tower foundation, the connection between the insulator steel foot and the hanging point, and the outlet of the conductor tension clamp; Step 4: Extract data point sequences with spatiotemporal continuous evolution characteristics from the dynamic feature point set as the core trajectory for constructing the three-dimensional dynamic ellipsoid; extract data point clusters reflecting the stable and healthy state of the equipment from the equipment status feature point set as the reference framework for constructing the three-dimensional polyhedron; construct the three-dimensional dynamic ellipsoid based on the core trajectory, and construct the three-dimensional polyhedron based on the reference framework. Step 5: Calculate the spatial coupling intersection of the three-dimensional dynamic ellipsoid and the three-dimensional polyhedron to generate coupling parameters that characterize the correlation between disaster characteristics and equipment status; based on the coupling parameters, construct a two-dimensional correlation model that integrates the dynamic coupling characteristics of multiple disasters and the vulnerability of equipment throughout its entire life cycle. Step 6: Call the two-dimensional correlation model and calculate the dynamic coupling strength under the action of disaster chain by real-time iterative fusion of meteorological forecast data and real-time power grid operating parameters, and generate the dynamic coupling strength calculation results. Step 7: Based on the dynamic coupling strength calculation results, generate an emergency control decision instruction set for the target power transmission channel.

[0014] In this embodiment of the invention, meteorological forecast data and real-time operating parameters of the power grid are accurately collected for the target transmission channel. The feature point set is deployed to closely match the core areas affected by multiple disasters such as icing, lightning, and soil frost heave, as well as key equipment parts such as towers, insulators, and conductors, making it highly targeted. Through spatial coupling analysis of three-dimensional dynamic ellipsoids and three-dimensional polyhedra, the deep correlation between the dynamic evolution of multiple disasters and the stable state of equipment is intuitively presented, making the coupling logic clearer. The two-dimensional correlation model fully incorporates the vulnerability of equipment throughout its entire life cycle, adapting to the state differences of equipment after long-term operation, making the assessment more comprehensive. Relying on a real-time iterative fusion mechanism, data is dynamically updated and calculation results are continuously optimized, effectively improving the dynamism and accuracy of coupling strength assessment. The generated emergency control decision instructions are dynamically adjusted in combination with the actual disaster coupling characteristics, meeting on-site needs and making them more targeted and operable. The method effectively strengthens the power grid's ability to predict and prevent risks of multi-disaster coupling, reduces the probability of safety hazards, and maximizes the safe and stable operation of the target transmission channel.

[0015] In a preferred embodiment of the present invention, step 1 above may include: Step 1.1: Obtain multi-dimensional grid forecast data covering the target power transmission channel area from the meteorological data server. Specifically, this includes: constructing a multi-source meteorological data acquisition link through the meteorological data sharing platform deployed by the meteorological center, the regional meteorological observation station network, and temporary meteorological monitoring stations along the power transmission channel; accessing multi-dimensional grid forecast data covering the entire target power transmission channel area. The data dimensions clearly include daily minimum temperature, relative humidity, precipitation type and duration related to icing; lightning activity probability and lightning current amplitude forecast values ​​related to lightning; and soil stratification temperature, soil volumetric water content and freezing depth forecast data related to soil frost heave. The data time resolution is set to 15 minutes to 1 hour to ensure the capture of short-term disaster evolution characteristics. The spatial grid accuracy is set to 1 km × 1 km to 5 km × 5 km to accurately cover the potential disaster impact area within a 5 km radius along the power transmission channel.

[0016] Step 1.2 involves performing spatiotemporal alignment and interpolation on the multi-dimensional grid forecast data to generate meteorological forecast data that precisely matches the geographical coordinates of the transmission channel. Specifically, this includes: uniformly calibrating the timestamps of multi-source meteorological data using the standard time of the power grid dispatching terminal as a reference; comparing the deviation between each data acquisition time and the reference time, and correcting the time offset for data with a deviation exceeding 30 seconds to achieve time synchronization alignment between different data sources; spatially extracting the centerline latitude and longitude coordinates, precise tower location coordinates, and elevation data of the target transmission channel; and filtering out discrete data within the transmission channel's coverage area. The grid forecast data is used to determine the target locations for data interpolation, including the center point of each kilometer segment of the transmission channel and the location of each tower. The straight-line distance from each target location to all surrounding grid data points is calculated, with closer grid data points assigned higher weights and farther away grid data points assigned lower weights. Based on the values ​​of each grid data point and their corresponding weights, a weighted average is used to calculate the meteorological data value for each target location. Finally, meteorological forecast data is generated with each kilometer segment of the transmission channel as the unit and each tower as an independent node, ensuring that the data is accurately matched with the geographical coordinates of the transmission channel.

[0017] Step 1.3: Obtain real-time telemetry and teleindication data from each monitoring point within the target transmission channel from the power grid dispatch data acquisition and monitoring terminal. Specifically, this includes: establishing a full-link data acquisition channel; covering all monitoring points of key equipment within the target transmission channel, such as tower foundations, insulator strings, conductors, tension clamps, and grounding devices; telemetry data specifically includes the settlement and tilt angle of tower foundations, leakage current, surface contamination, and temperature of insulators, operating current, voltage, temperature, sag, vibration amplitude, and mechanical stress of conductors, and continuous parameters such as contact resistance and temperature of tension clamps; teleindication data specifically includes status indicators such as normal equipment operation, abnormal alarms, and fault tripping, the open / close position signals of circuit breakers and disconnectors, and fault alarm signals such as insulator flashover and excessive conductor icing. The telemetry data acquisition frequency can be set from 1 second to 5 seconds, and the teleindication data acquisition frequency can be set from 100 milliseconds to 500 milliseconds. These settings can be adjusted according to changes in circumstances and are not fixed, but must be consistent with the real-time requirements of equipment operation monitoring.

[0018] Step 1.4 involves quality verification and missing value repair of real-time telemetry and teleindication data to form standardized real-time power grid operating parameters. Specifically, this includes: employing a two-layer quality verification mechanism to process real-time telemetry data. The first layer, based on equipment factory technical parameters, industry operating standards, and historical data statistical patterns, sets reasonable value ranges for each parameter, directly eliminating extreme abnormal data exceeding these ranges. The second layer verifies the consistency between data by analyzing the inherent logical relationships between parameters such as current and power, voltage and current, and temperature and stress, marking data with logical contradictions as suspicious data and conducting secondary verification. This process is applied to data transmission. For missing values ​​caused by interruptions or temporary sensor malfunctions, if the continuous missing time is short, linear interpolation of adjacent time data is used to supplement them; if the continuous missing time is long, the missing data is supplemented by combining historical operating data of the equipment during the same period, operating data of similar equipment under the same environmental conditions, and current weather forecast trends, through data trend analysis and reasonable estimation to ensure data integrity; finally, all data after verification and repair are classified and organized, and the data encoding, field format and unit are unified to form structured real-time power grid operating parameter data containing unique equipment identifiers, collection timestamps, parameter names, standard parameter values ​​and data quality levels.

[0019] In this embodiment of the invention, multi-dimensional grid forecast data of the target power transmission channel area is accurately obtained from the meteorological data server. After spatiotemporal alignment and interpolation processing, it is accurately matched with the geographical coordinates of the power transmission channel. At the same time, real-time telemetry and teleindication data of each monitoring point are obtained from the power grid dispatch data acquisition and monitoring terminal. Standardized real-time power grid operation parameters are formed through quality verification and missing value repair. This not only ensures the comprehensiveness and reliability of the data source, but also improves the accuracy, completeness and standardization of the data. It lays a solid data foundation for the accurate construction of feature point sets, the effective construction of three-dimensional models and the accurate calculation of coupling strength, and enhances the rationality and operability of the entire calculation method.

[0020] In a preferred embodiment of the present invention, step 2 above may include: Step 2.1: Based on the spatial and temporal range of the disaster defined by meteorological forecast data, configure virtual icing monitoring points at the midpoints of the icing-prone sections of the transmission line. Extract data corresponding to these virtual monitoring points to generate a set of icing growth monitoring points. This includes: analyzing icing-related parameters in the meteorological forecast data to determine the start time, duration, and spatial coverage of the icing disaster; combining the line design drawings of the target transmission channel, historical icing fault records, and overlaying topographic data along the line to delineate the icing-prone area as a closed polygonal spatial range; first, obtain the starting coordinates (including latitude, longitude, and elevation) and ending coordinates of each icing-prone line segment; by calculating the arithmetic mean of the values ​​at the starting and ending points in the three dimensions of longitude, latitude, and elevation, accurately obtain the three-dimensional coordinates of the midpoint of the line segment to ensure that the midpoint is precisely at the geometric center of the line segment; to verify whether the midpoint is within the icing disaster polygonal area, from the... A virtual ray is emitted horizontally to the right from the midpoint, and the intersection of the ray with the polygon boundary is characterized one by one. If the ray intersects with a boundary segment, it is counted once. When the ray passes through a vertex of the polygon, it is necessary to determine whether the two adjacent boundary segments of the vertex are located on both sides of the ray. It is counted only when they are on both sides to avoid double counting. If the total number of intersections is odd, the midpoint is determined to be in the disaster area. If it is even, the monitoring point position is finely adjusted to the inside of the polygon. A virtual icing monitoring point is configured at the verified midpoint position, and the data collection frequency is set to be consistent with the weather forecast data update frequency. Data such as temperature, relative humidity, precipitation intensity, precipitation type, wind speed, and supercooled water droplet content corresponding to the monitoring point are continuously extracted. The data is organized in the order of collection timestamp, and each record is labeled with the monitoring point coordinates and the number of the icing-prone line segment to which it belongs, forming a structured set of icing growth monitoring points.

[0021] Step 2.2 involves configuring virtual lightning monitoring points on the tops of power poles in lightning-prone areas. Data corresponding to these virtual monitoring points is extracted to generate a lightning activity monitoring point set. This includes: extracting lightning warning information from meteorological forecast data to determine the potential occurrence time, warning level, and spatial distribution range of lightning disasters; retrieving lightning activity statistics from the past five years along the transmission line and, combined with topographic analysis, delineating lightning-prone areas such as highlands, ridges, and open areas as irregular polygonal regions; then locating the unique identifier and precise coordinates of all power poles within the high-risk areas, using the center point of the tower top as a reference to determine the spatial location of the virtual lightning monitoring points; and, to determine whether a monitoring point is located in the core disaster impact area, first collecting the precise coordinates of known high-lightning-occurrence points within the area and calculating the coordinates of the monitoring point relative to each... The coordinate differences of high-incidence points in latitude, longitude, and elevation are calculated. The difference in each dimension is squared, and the squares of the three dimensions are summed. The square root of the sum is then taken to obtain the straight-line distance between the monitoring point and each high-incidence point. The closer the distance, the higher the probability that the monitoring point is affected by lightning. At the same time, the method consistent with step 2.1 is used to verify whether the monitoring point is within the polygon of the lightning-prone area. That is, virtual rays are emitted from the monitoring point and the number of intersections is counted to ensure accurate spatial attribution. Data acquisition parameters of the monitoring points are configured, and key parameters such as atmospheric electric field intensity, lightning discharge frequency, estimated lightning current amplitude, and lightning azimuth are extracted in real time. The collected data are integrated according to the time series, and each data record is associated with the tower number, monitoring point coordinates, and acquisition time to form a complete set of lightning activity monitoring points.

[0022] Step 2.3 involves configuring virtual frost heave monitoring points at the geometric center of the tower foundations in the soil frost heave sensitive area. Data corresponding to these virtual monitoring points is extracted to generate a frost heave monitoring point set. Specifically, this includes: analyzing forecast values ​​for soil temperature and moisture content from meteorological forecasts to clarify the time window and spatial coverage of soil frost heave disasters; collecting soil type distribution maps, groundwater depth data, and historical frost heave disaster records along the transmission line, delineating silty soil and clay soil distribution areas and areas with shallow groundwater depths as soil frost heave sensitive polygonal areas; and, based on the tower foundation construction archives, obtaining the structural dimensions, depth, and three-dimensional coordinates of all vertices of the outline of each tower foundation within the sensitive area. The total number of vertices is first counted, and then the sum of the values ​​for all vertices in the longitude, latitude, and elevation directions is calculated separately. The sum in each direction is then divided by the total number of vertices. The geometric center coordinates of the tower foundation outline are obtained. This calculation method ensures that the center position fits the core stress area of ​​the foundation and accurately reflects the overall frost heave effect of the foundation. The virtual ray method is used to verify whether the geometric center is within the frost heave sensitive polygon. Virtual rays are emitted from the center in any direction. By counting the number of effective intersections between the ray and the polygon boundary (to avoid the problem of double counting of vertices), it is confirmed whether the monitoring point is within the sensitive area, and the monitoring point is prevented from exceeding the disaster impact range. Frost heave virtual monitoring points are configured at the verified geometric center positions, and a layered acquisition mode is set to cover the key soil layers from the ground surface to the bottom of the foundation. Data such as temperature, volumetric water content, freezing state, and estimated frost heave amount of each soil layer are continuously extracted. The data are organized according to the acquisition time and soil layer depth, and the tower foundation number and coordinates corresponding to the monitoring points are labeled to form a frost heave monitoring point set.

[0023] Step 2.4 aggregates the monitoring point sets for icing growth, lightning activity, and frost heave to form an initial dynamic feature point set. This includes: unifying the data format of the three monitoring point sets; standardizing the coordinate system of all data records to WGS-84 and the timestamps to UTC standard time; aligning the icing growth, lightning activity, and frost heave monitoring point sets along the time dimension using the collection timestamp as the key field; setting a reasonable spatial search radius for each monitoring point based on the impact range of different disasters and the spatial distribution density of the monitoring points to ensure coverage of other related disaster monitoring points; and traversing all other disaster monitoring points centered on the current monitoring point. For monitoring points of different disaster types, the straight-line distance between them is calculated (i.e., the square root of the sum of the squares of their coordinate differences). The distance is then used to determine if it is less than a set search radius. If it is, the point is considered a spatially adjacent point, establishing a correlation between monitoring points of different disaster types. Simultaneously, weights are assigned based on the straight-line distance between adjacent points and the current monitoring point; the closer the distance, the larger the weight coefficient, providing a priority basis for subsequent data matching. All aligned monitoring data is deduplicated to remove duplicate records and invalid data. All processed data is then aggregated and categorized and indexed according to disaster type, monitoring point coordinates, and collection time, forming an initial dynamic feature point set covering multiple disaster types, multiple spatial locations, and multiple time nodes.

[0024] Step 2.5: From the initial dynamic feature point set, select the data point sequence whose data values ​​exhibit continuous changes in the time and spatial dimensions as the final dynamic feature point set. Specifically, this includes: arranging the data in the initial dynamic feature point set in ascending order by collection timestamp, and comparing data from the same monitoring point at adjacent time nodes; setting a reasonable threshold for data fluctuation, determined based on the statistical fluctuation range of similar historical data, and eliminating abnormal data points whose numerical mutations exceed the threshold; grouping the data points according to spatial location, with adjacent monitoring point data within the same disaster polygon area considered as a group; to verify the spatial continuity of the data within a group, first obtaining the three-dimensional coordinates of two adjacent monitoring points within the group to form a connecting line segment; then extracting each boundary of the disaster polygon one by one. For each line segment, the relative positional relationship between the endpoints of two line segments is analyzed. By determining the probability of the intersection between the extension of the line connecting adjacent monitoring points and the boundary line segment, cases where the two line segments are parallel or have no common point are excluded. If the two line segments have a common point, and this common point is located between the endpoints of the two line segments (not on the extension line), then it is determined that the line connecting adjacent monitoring points crosses the boundary of the disaster polygon, the spatial association of the data is broken, and the relevant discrete data points need to be removed. If they do not intersect, the spatial association is determined to be valid. The data points after being filtered by time and space dimensions are organized according to the structure of monitoring point coordinates, timestamps, and data values ​​to form a spatiotemporally continuous data point sequence. All valid data point sequences are summarized as the final dynamic feature point set to ensure that the data can completely reflect the continuous evolution process of multiple disasters.

[0025] In this embodiment of the invention, virtual monitoring points are precisely configured for key impact areas of multiple hazards such as icing, lightning, and soil frost heave. Data collection focuses on the core affected areas of the hazards, demonstrating strong targeting and specificity. By extracting data corresponding to each virtual monitoring point and aggregating it to form an initial dynamic feature point set, and then filtering data point sequences with continuously changing spatiotemporal dimensions, discrete and abnormal data are effectively eliminated, ensuring the continuity and reliability of the final dynamic feature point set. This provides a precise core trajectory for the subsequent construction of a three-dimensional dynamic ellipsoid, strongly supporting the accurate characterization of the dynamic evolution process of multiple hazards and providing a solid data foundation for improving the accuracy of coupled analysis of multiple hazards and equipment status.

[0026] In a preferred embodiment of the present invention, step 3 above may include: Step 3.1: From the real-time operating parameters of the power grid, analyze the foundation monitoring data corresponding to the monitoring points at the four corners of the tower foundation cap to generate a set of cap state points characterizing the foundation bearing state. Specifically, this includes: accurately selecting data entries bound to the monitoring points at the four corners of the tower foundation cap from standardized real-time operating parameters of the power grid. These parameters originate from real-time uploaded data from the power grid and real-time uploaded data from online foundation monitoring sensors. The monitoring data specifically covers core indicators such as vertical settlement, horizontal tilt angle, vertical pressure, horizontal thrust on the foundation, and lateral stress and pore water pressure in key soil layers surrounding the cap; and adjusting the data based on the unique number of the tower foundation. The construction drawings of the corresponding pier cap are retrieved, and the precise three-dimensional coordinates (latitude, longitude, and elevation) of the four corner monitoring points are obtained. A mapping relationship is established between the coordinates and the monitoring data to ensure that each data point can be accurately located to the specific corner of the pier cap. The data is sorted in ascending order according to the collection timestamp, and each record is labeled with the monitoring point number, such as the southeast corner or northwest corner of the pier cap, the tower number, and the coordinate information. Combining the design bearing capacity limit of the tower foundation with industry operation safety standards, extreme abnormal data that exceed the reasonable value range are eliminated, and finally a structured pier cap status point set is formed, which includes the tower number, pier cap monitoring point coordinates, monitoring point number, collection time, and various foundation bearing parameters.

[0027] Step 3.2 involves parsing the insulator monitoring data corresponding to the connection between the insulator steel foot and the hanging point from the real-time power grid operating parameters, generating a set of insulator state points characterizing insulation performance and mechanical fatigue. Specifically, this includes: locating the insulator monitoring data corresponding to the connection between the insulator steel foot and the hanging point in the established real-time power grid operating parameter library using equipment type and monitoring point location tags; data types specifically include the insulator's peak and average leakage current, surface contamination equivalent salt density, ash density, partial discharge signal amplitude and frequency, mechanical vibration amplitude and frequency at the connection point, contact stress between the steel foot and the hanging point, and real-time temperature at the connection point, etc. Each data item is directly related to the insulator's insulation performance stability and mechanical structure fatigue degree; and classifying and grouping the monitoring data according to the insulator's unique identification code, including the tower number, line segment number, and installation location sequence number, to ensure that all monitoring data for the same insulator are independently grouped and not mixed up. Each set of data is arranged in order of collection timestamp, and the precise location coordinates of the connection between the steel foot and the hanging point of the monitoring point are marked in detail. The corresponding tower number, line section number and insulator model information are also associated. By comparing the historical normal operation data range of the insulator, abnormal jump data caused by sensor instantaneous failure or electromagnetic interference are removed, and a structured insulator status point set containing the unique insulator identifier, precise location of the monitoring point, collection time, insulation performance parameters and mechanical fatigue parameters is generated.

[0028] Step 3.3 involves parsing the conductor monitoring data corresponding to the conductor tension clamp exit from the real-time power grid operating parameters to generate a conductor state point set characterizing the conductor stress and sag state. Specifically, this includes: extracting the conductor monitoring data corresponding to the conductor tension clamp exit from the real-time power grid operating parameters by filtering by line segment number and monitoring point type; core monitoring data includes parameters such as the conductor's dynamic stress value, static stress value, real-time sag height, conductor body temperature, aerobatic vibration amplitude and frequency, and secondary span oscillation amplitude. These data directly reflect the conductor's stress balance state and sag change trend, serving as a key basis for judging the safety of line transmission; based on the conductor's line... The section number, tension clamp model, and installation location number were used to retrieve the line design drawings and determine the precise spatial coordinates of the monitoring point at the tension clamp outlet. The coordinates were then accurately matched with the monitoring data. The data was sorted according to the acquisition timestamp to ensure that each data point corresponds to the conductor status at a specific time node. Each record was labeled with the conductor model, the line section number, the tension clamp number, and the monitoring point coordinates. For the few missing values ​​that occurred during data transmission, valid data from adjacent times was used to complete the data. At the same time, invalid data that clearly exceeded the rated stress and reasonable sag range of the conductor were removed. Finally, a structured conductor status point set representing the stress distribution and sag status of the conductor was formed.

[0029] Step 3.4: Aggregate the foundation status point sets, insulator status point sets, and conductor status point sets to form an initial equipment status feature point set. Specifically, this includes: first, unifying the data formats of the foundation, insulator, and conductor status point sets; converting all data coordinate systems to the WGS-84 coordinate system consistent with the transmission channel; and uniformly calibrating the timestamps to UTC standard time to ensure consistent spatiotemporal references across different point sets; using the acquisition timestamp as the core key field, precisely aligning the three types of status point sets in the time dimension with millisecond-level precision, ensuring that the foundation, insulator, and conductor monitoring data of the same tower or line segment at the same time node can be correlated one-to-one; and matching data according to spatial location and equipment affiliation to identify foundation status points under the same tower number. The status data, the status data of all insulators on the tower, and the status data of conductors at the tension clamp exits on both sides of the tower are categorized into the same data group. For line sections spanning towers, the data groups are further subdivided according to mileage station numbers to ensure the accuracy of spatial correlation. The correlated data is deduplicated, and the criteria for determination are the same equipment type, the same monitoring point, and the same timestamp. If duplicate records exist, the record with the highest data quality level is retained. At the same time, logically contradictory data, such as data where the conductor stress and sag change trends are completely contradictory, are removed. All processed data are summarized, and a three-level classification index is established according to equipment type, monitoring point coordinates, and acquisition time to form an initial equipment status feature point set covering the core status of key equipment such as tower foundations, insulators, and conductors.

[0030] Step 3.5: From the initial set of equipment status feature points, select a cluster of stable data points whose data fluctuation range is below a set threshold within a preset time period, as the final set of equipment status feature points. This specifically includes: setting a preset time period based on equipment operating characteristics and monitoring needs. This period must cover a complete normal operating fluctuation cycle of the equipment, such as 24 hours, to ensure comprehensive capture of the equipment's stable operating status; based on the equipment's manufacturer's technical parameter manual, industry operating standards, and statistical results of historical stable operating data from the past year, setting reasonable fluctuation thresholds for different parameter types, such as stress parameter thresholds based on ±5% of the rated stress, and temperature parameter thresholds based on ±3℃ of the normal operating temperature range; dividing the initial set of equipment status feature points into... Several continuous and non-overlapping data segments are used, each corresponding to a complete preset time period. For the same parameter data at the same monitoring point within each data segment, the difference between the maximum and minimum values ​​is calculated to obtain the actual fluctuation range of the parameter within that period. The fluctuation range of each parameter is compared with the corresponding set threshold. If the fluctuation range of all parameters is below the threshold, the data segment is determined to correspond to a stable operating state of the equipment. If any parameter fluctuation exceeds the standard, the data segment is removed. All data segments that meet the stability standard are integrated and organized according to the structure of equipment type, monitoring point, parameter type, stable time period, and data sequence to form a stable data point cluster that can reflect the normal health level of the equipment, which serves as the final set of equipment status feature points.

[0031] In this embodiment of the invention, the focus is on key components of critical equipment such as tower foundation caps, insulators, and conductors. Monitoring data corresponding to each component is precisely analyzed to generate point sets representing the equipment's load-bearing state, insulation performance, mechanical fatigue, stress, and sag status across multiple dimensions. This comprehensively covers key performance dimensions of equipment operation, providing highly targeted and comprehensive data. By aggregating various state point sets and filtering stable data point clusters with fluctuations below a threshold within a preset time period, abnormal fluctuation data can be effectively eliminated, ensuring the stability and reliability of the final equipment state characteristic point set. This provides a precise benchmark framework for constructing a three-dimensional polyhedron. It strongly supports the accurate characterization of the equipment's stable and healthy state and provides data support for the deep coupling analysis of multiple disasters and equipment status.

[0032] In a preferred embodiment of the present invention, step 4 above may include: Step 4.1 involves performing spatiotemporal trajectory fitting on the data point sequence in the final dynamic feature point set to form a core trajectory representing the disaster evolution path. This includes: acquiring all data point sequences in the final dynamic feature point set and arranging them in ascending order by collection timestamp to ensure the data points strictly follow the natural evolution time logic of disasters such as icing, lightning, and soil frost heave; extracting the latitude, longitude, and elevation three-dimensional spatial coordinates and corresponding timestamp information of each data point to construct a structured data matrix containing spatiotemporal dual-dimensional attributes, providing complete data support for trajectory fitting; and using the time interval between adjacent data points as the weighting criterion, with shorter time intervals resulting in higher weights. By calculating the spatial vectors between adjacent data points and performing a weighted average of all adjacent vectors, the overall smooth trend direction of the trajectory is obtained. Using a segment-by-segment connection method, adjacent data points that are continuous in time are first connected to form preliminary trajectory segments. Then, the inflection angles of the segments are adjusted according to the smooth trend direction to make the trajectory transition more natural and smooth. A trajectory deviation threshold is set, which is statistically determined based on the dispersion of historical disaster evolution trajectories. Isolated data points that deviate from the overall smooth trend beyond the threshold are screened and removed to avoid anomalies interfering with the authenticity of the trajectory. Finally, a continuous, smooth core trajectory that is highly consistent with the spatiotemporal evolution law of multiple disasters is formed.

[0033] Step 4.2 involves analyzing and identifying the extension direction and data density distribution of the core trajectory to determine the principal axis direction and semi-axis length of the three-dimensional dynamic ellipsoid. Specifically, this includes: extracting the three-dimensional spatial coordinates of all data points on the core trajectory; calculating the coordinate difference between the trajectory's starting and ending points to obtain an initial overall extension direction vector, which serves as a preliminary reference for the principal axis direction; selecting several key nodes on the core trajectory at equal time intervals, with the interval between each node set based on the average rate of disaster evolution; calculating the local direction vector of each key node and its adjacent nodes; comparing all local direction vectors with the initial overall direction vector for consistency; eliminating abnormal local vectors with excessive deviations; and performing a weighted average fusion of the remaining effective vectors to correct the result. The final core trajectory extension direction is the reference direction of the principal axis of the three-dimensional dynamic ellipsoid. A three-dimensional grid with equal spacing is divided according to the spatial scale along the power transmission channel. The grid size is determined based on the statistical results of the disaster's impact range. Data points on the core trajectory are assigned to their respective grids. The number of data points in each grid is counted to form a data density distribution map. The direction with the highest data density is determined as the major axis direction of the ellipsoid, the next highest as the central axis direction, and the lowest as the minor axis direction. The difference between the maximum and minimum coordinates of data points along each axis is calculated to obtain the effective coverage length for each direction. Half of this length is taken as the length of the corresponding semi-axis, ensuring that the semi-axis length accurately matches the differences in the disaster's impact range in different spatial directions.

[0034] Step 4.3, based on the principal axis direction and semi-axis length, construct a three-dimensional dynamic ellipsoid, specifically including: traversing all data points on the core trajectory, calculating the arithmetic mean of the three-dimensional coordinates of each data point, and using the spatial position corresponding to the arithmetic mean as the geometric center of the three-dimensional dynamic ellipsoid to ensure that the ellipsoid can accurately cover the core area of ​​disaster evolution; mapping and aligning the determined major axis, median axis, and minor axis directions with the coordinate axes of the WGS-84 three-dimensional coordinate system to establish a spatial orientation reference for the ellipsoid, ensuring that the spatial attitude of the ellipsoid is consistent with the disaster diffusion direction; and according to the specific lengths of the three semi-axis, extending from the geometric center to both ends along the corresponding axes to determine the ellipsoid's position along the major axis, median axis, and minor axis directions. The boundary extreme points are used to define the spatial range of the ellipsoid; parameters are set at uniform angular intervals, and the three-dimensional coordinates of each grid point on the ellipsoid surface are calculated sequentially. The position of each grid point is derived from the geometric center coordinates, principal axis direction, and corresponding semi-axis length to ensure that the grid points are evenly distributed on the ellipsoid surface; all surface grid points are connected sequentially according to spatial orientation to form a continuous and closed ellipsoid surface structure, ensuring the integrity and smoothness of the ellipsoid; combined with the dynamic evolution characteristics of the disaster, time attributes are bound to the ellipsoid, and the geometric center, axis length, and spatial orientation of the ellipsoid are dynamically updated according to the core trajectory position corresponding to different timestamps, ultimately forming a complete three-dimensional dynamic ellipsoid that can fit the disaster evolution process in real time.

[0035] Step 4.4 involves mapping the stable data point clusters from the final equipment status feature point set to three-dimensional space to determine the vertex set of the solid polyhedron. Specifically, this includes: using the WGS-84 three-dimensional coordinate system consistent with the transmission channel to map the stable data point clusters from the final equipment status feature point set to coordinates, with each stable data point's latitude, longitude, and elevation coordinates directly corresponding to a unique spatial location in the coordinate system; storing the equipment status parameters represented by the data points, such as foundation bearing pressure, insulator leakage current, and conductor stress, as attribute information of the corresponding spatial points, linked one-to-one with the coordinate information; and classifying the data according to the equipment type of the tower foundation, insulator, and conductor. The data points are categorized and grouped, and the specific equipment location corresponding to each group of data points is clearly defined. For each group of data points, the data points with the largest and smallest values ​​are selected in each equipment state parameter dimension. These extreme points can accurately reflect the operating state boundary of the equipment in that parameter dimension. The spatial straight-line distance between extreme points in the same parameter dimension is calculated. If the distance is less than the reasonable threshold corresponding to the equipment structural size, it is regarded as a redundant point with too close spatial location. The points that can more comprehensively represent the equipment state boundary are retained. The extreme points selected in all equipment types and all parameter dimensions are summarized to form a vertex set that can fully cover the spatial boundary of the equipment state and has no redundancy.

[0036] Step 4.5: Based on the spatial topological relationships defined in the vertex set, a reference frame for constructing the solid polyhedron is generated by sequentially connecting each vertex. Based on this reference frame, the solid polyhedron is constructed, specifically including: defining the connection relationships and spatial layout between the tower foundation, insulators, and conductors according to the physical structure design drawings of the power transmission equipment; defining the spatial topological relationships of each vertex in the vertex set based on this; defining the vertex range of the same equipment part and the adjacent association rules between vertices of different equipment parts to ensure that the topological relationships conform to the actual spatial structure of the equipment; based on the defined spatial topological relationships, starting from the vertex of the tower foundation, connecting all vertices of the same foundation in a clockwise or counterclockwise order to form the corresponding local polygonal face of the foundation. Next, the vertices corresponding to the insulators and conductors are connected sequentially to form local polygonal faces of each equipment part. Adjacent vertices of different equipment parts are connected sequentially according to topological relationships, such as the apex vertex and the insulator installation point vertex, and the insulator vertex and the conductor hanging point vertex, so that each local polygonal face is seamlessly connected, constructing a closed three-dimensional reference frame. The integrity of the reference frame is checked to see if there are any unconnected isolated vertices or broken edges. If problems are found, the topological relationships and vertex connection order are corrected according to the physical structure of the equipment. Based on the reference frame, the mesh inside each polygonal face is divided with uniform density, and the blank areas inside the face are filled to form a continuous and complete three-dimensional surface. The three-dimensional surfaces of all equipment parts are integrated to construct a three-dimensional polyhedron that can accurately represent the stable and healthy state of the equipment.

[0037] In this embodiment of the invention, by fitting the spatiotemporal trajectory of the data point sequence of the final dynamic feature point set, the dynamic evolution path of multiple hazards is accurately captured and a core trajectory is formed. The principal axis direction and semi-axis length of the three-dimensional dynamic ellipsoid are determined by combining the trajectory extension direction and data density distribution, enabling the constructed three-dimensional dynamic ellipsoid to realistically and concretely represent the spatiotemporal evolution characteristics and impact range of multiple hazards. Simultaneously, the stable data point cluster of the final equipment state feature point set is mapped to three-dimensional space, clarifying the vertex set of the three-dimensional polyhedron and connecting the vertices through spatial topological relationships to generate a baseline framework. The constructed three-dimensional polyhedron can accurately characterize the spatial state and performance boundaries of stable equipment health. The construction of these two types of geometries provides intuitive and accurate spatial model support for the deep coupling analysis of multi-hazard dynamic evolution and equipment stability, effectively building a bridge between hazard characteristics and equipment state, and improving the rationality, accuracy, and visualization of coupling strength calculation.

[0038] In a preferred embodiment of the present invention, step 5 above may include: Step 5.1: Based on the principal axis direction and semi-axis length of the three-dimensional dynamic ellipsoid, define the spatial parameter equations of the three-dimensional dynamic ellipsoid. Specifically, this includes: establishing a local three-dimensional coordinate system with the geometric center of the three-dimensional dynamic ellipsoid as the origin, the local three-dimensional coordinate system maintaining spatial orientation consistent with the WGS-84 coordinate system; clarifying the geometric parameters of the three-dimensional dynamic ellipsoid, the geometric center coordinates are (x0, y0, z0), x0, y0, and z0 are the three coordinate components of the geometric center of the three-dimensional dynamic ellipsoid in the three-dimensional spatial coordinate system; the major axis length is 2a, the median axis length is 2b, and the minor axis length is 2c; the direction cosines corresponding to the major axis, median axis, and minor axis are l1, m1, n1, l2, m2, n2, and l3, m3, n3, respectively, the direction cosines being used to characterize the spatial orientation of the principal axis in the global coordinate system; defining the spatial parameter equations of the three-dimensional dynamic ellipsoid, the expressions for the x, y, and z direction equations are as follows: ; ; ; Where θ is the azimuth angle, ranging from 0 to 2π; φ is the polar angle, ranging from 0 to π; a, b, and c are the lengths of the major, middle, and minor axes, respectively; l1, m1, and n1 satisfy: Similarly, l2, m2, n2 and l3, m3, n3 also satisfy the normalization condition, so the integrated formula is: , , ; It can ensure the effectiveness of the main axis direction; through the parametric equations, the spatial coordinates of any point on the three-dimensional dynamic ellipsoid can be accurately described, and the spatial form and range of disaster evolution can be fully characterized.

[0039] Step 5.2: Based on the reference frame and vertex set of the solid polyhedron, determine the equations of each plane constituting the solid polyhedron. Specifically, this includes extracting the three-dimensional coordinates of all vertices in the vertex set of the solid polyhedron, where each vertex coordinate is represented by (x, y). i y i z i Let ) represent the vertex index (i = 1, 2, 3, up to N); based on the baseline framework of the solid polyhedron, the vertices are grouped according to spatial topological relationships, with each group of three non-collinear vertices corresponding to a plane, ensuring that each plane is the outer surface of the solid polyhedron; using the three vertices of a certain plane as an example... , , For example, to calculate the normal vector of the plane, we first need to calculate vector 1 and vector 2, and the formula expression is as follows: ; ; in, For vector 1, Let's take vector 2 as an example; then we obtain the normal vector through the cross product of the vectors. , ; ; ; A, B, and C are the three components of the normal vector of a plane on the outer surface of a solid polyhedron in the three-dimensional coordinate system. Substituting the coordinates of any vertex P1 (x1, y1, z1) into the general plane equation Ax + By + Cz + D = 0, we obtain D = -(Ax1 + By1 + Cz1). All planes are calculated in this way, resulting in the equation Ax + By + Cz + D = 0 for each plane, where A, B, and C are not simultaneously 0. Through all plane equations, the spatial boundary of the solid polyhedron is fully defined, ensuring that each surface of the solid polyhedron has a corresponding mathematical expression.

[0040] Step 5.3 involves solving the spatial geometric intersection by simultaneously solving the parametric equations of the dynamic ellipsoid and the plane equations of the polyhedron. Specifically, this includes: substituting the parametric equations of the dynamic ellipsoid into the plane equations of the polyhedron to obtain a system of equations for θ and φ; for each plane equation Ax + By + Cz + D = 0, substituting the parametric expressions for x, y, and z to form trigonometric function equations containing only θ and φ; solving this system of equations to obtain the combinations of θ and φ values ​​that satisfy the plane equations, where these values ​​correspond to the coordinates of the intersection points of the ellipsoid and the plane; and determining whether each intersection point lies within the plane. Inside the polyhedron, the sign of the result after substituting the point into all other plane equations is verified. If all the result signs are the same, the point is determined to be inside the polyhedron. The above substitution, solution, and verification process is repeated for all plane equations to collect all points that simultaneously satisfy the ellipsoid parametric equation and the polyhedron interior conditions. The collected point sets are deduplicated and sorted, and connected according to spatial position to form a continuous closed region. The continuous closed region is the spatial geometric intersection of the three-dimensional dynamic ellipsoid and the three-dimensional polyhedron, so as to achieve precise spatial coupling between the disaster impact range and the equipment state boundary.

[0041] Step 5.4 involves performing topological and metric analysis on the spatial geometric intersection to extract the intersection volume, intersection center coordinates, and intersection shape complexity index, which serve as coupling parameters characterizing the correlation between disaster characteristics and equipment status. Specifically, this includes: performing topological analysis on the spatial geometric intersection to determine its connectivity, identifying whether the intersection is a single continuous region or multiple discrete regions, and recognizing the boundary type of the intersection, clarifying that the boundary is a closed curve composed of an ellipsoid and a polyhedral plane; during metric analysis, the intersection is divided into several small tetrahedrons, with each small tetrahedron's vertices being points on the intersection boundary, and calculating the volume V of each small tetrahedron. i The volume expression is: V i =|(x2-x1)(y3-y1)(z4-z1)+(y2-y1)(z3-z1)(x4-x1)+(z2-z1)(x3-x1)(y4-y1)- (z2-z1)(y3-y1)(x4-x1)-(y2-y1)(x3-x1)(z4-z1)-(x2-x1)(z3-z1)(y4-y1)| / 6; in, Let be the coordinates of the four vertices of the small tetrahedron; sum the volumes of all the small tetrahedrons to obtain the intersection volume. ; Calculate the coordinates of the intersection center (x c y c , z c The expression is: ; in It is the coordinate of the j-th small tetrahedron, that is, the arithmetic mean of the coordinates of its four vertices; V j V is the volume of the small tetrahedron, and V is the total volume; calculate the intersection shape complexity index S, the expression is: S surface The surface area of ​​the intersection is obtained by calculating the sum of the areas of all the boundary surfaces of the intersection; The S value is a reference value for the surface area of ​​the intersection with a sphere of equal volume. A larger S value indicates a more complex shape of the intersection. The intersection volume V and the intersection center coordinates (x, y) are used as reference values. c y c , z c The shape complexity index S serves as the core coupling parameter characterizing the correlation between disaster characteristics and equipment status.

[0042] Step 5.5: Map the coupling parameters to model weight coefficients and the correlation matrix; based on the model weight coefficients and the correlation matrix, construct a two-dimensional correlation model that integrates the dynamic coupling characteristics of multiple disasters and the vulnerability of equipment throughout its entire life cycle. Specifically, this includes: first, clarifying the core structure of the two-dimensional correlation model, where the first dimension is the dynamic coupling characteristics of multiple disasters, and the input is the extracted intersection volume V and the coordinates of the intersection center (x, y). c y c , z c The system employs several coupled parameters, including the intersection shape complexity index S, to cover the dynamic impact characteristics of three types of disasters—icing, lightning, and soil frost heave—at different time stages. The second dimension is the equipment lifecycle vulnerability dimension, with inputs being lifecycle vulnerability indicators for key equipment components, including: foundation bearing capacity attenuation rate (calculated based on the ratio of cumulative foundation settlement to design limit); insulator insulation performance degradation coefficient (derived from the leakage current change rate and cumulative surface contamination); conductor stress fatigue accumulation value (obtained by integrating historical stress fluctuation amplitude and duration); and tower structure corrosion degree (calculated based on corrosion depth detection data and material corrosion resistance years). All indicator data originate from power grid equipment operation and maintenance records, periodic inspection reports, and statistical analysis of historical monitoring data.

[0043] Before constructing the model input layer, the input data is preprocessed: the multi-hazard dynamic coupling parameters are normalized using min-max normalization, mapping each parameter's historical maximum and minimum values ​​to the 0-1 range to avoid interference from parameters of different magnitudes during model training; the equipment lifecycle vulnerability index is standardized by subtracting the mean and dividing by the standard deviation to ensure the data follows a normal distribution, thus improving model fitting. During input layer construction, the normalized coupling parameters are categorized and encoded according to hazard type, with each hazard type corresponding to a 3D vector V, coordinate correlation value, and S, forming a total 9-dimensional coupling parameter vector across the three hazard types; the standardized vulnerability index is encoded as a 4-dimensional vulnerability index vector based on equipment location. The number of neurons in the input layer is determined to be 13 (9-dimensional plus 4-dimensional), perfectly matching the total dimensions of the input data.

[0044] A two-layer fully connected hidden layer is constructed: the first layer has 1.5 times the number of neurons as the input layer, i.e., 19 neurons, to extract shallow correlation features of two-dimensional data; the second layer has 0.8 times the number of neurons as the first layer, i.e., 15 neurons, to deeply mine complex nonlinear correlations in the data. The hidden layer transforms the input features through a nonlinear transformation function, effectively capturing the non-proportional correlation between the dynamic evolution of disasters and the decay of equipment vulnerability. For example, when the disaster coupling parameters change slightly, the equipment vulnerability may fluctuate significantly due to the cumulative effect. When initially assigning weights, the model weight coefficients w... V w d w SThe connection weights from the input layer to the first layer correspond to the intersection volume, center distance, and shape complexity association weights, respectively; the association matrix M, with 3 rows and 4 columns, corresponds to three types of disasters and four vulnerability indicators, and serves as the connection weight from the first layer to the second layer. Elements M... ij Through M ij =α w ·V+β·w d +γ·w S The calculations are performed, with α, β, and γ serving as weighting coefficients satisfying α + β + γ = 1. Initial values ​​are calibrated based on historical data from the past five years: α = 0.4, β = 0.3, γ = 0.3 for icing disasters; α = 0.3, β = 0.4, γ = 0.3 for lightning disasters; and α = 0.3, β = 0.3, γ = 0.4 for soil frost heave disasters. The model output layer is defined, with a single coupling strength value ranging from 0 to 1. 0 represents no correlation, 1 represents extremely strong correlation with equipment on the verge of failure, and 0.4 to 0.6 is the critical correlation interval requiring an early warning. The output layer is mapped to the target interval through a linear transformation. The hidden layer outputs are first weighted and summed, then scaled according to a preset threshold to ensure that the coupling strength value intuitively and accurately reflects the degree of correlation.

[0045] Next, model training was conducted: First, a dataset was prepared, collecting historical data from 100 transmission lines and 5000 towers over the past five years, including quarterly disaster monitoring data such as ice thickness, lightning frequency, soil frost heave, equipment condition monitoring data such as foundation settlement, leakage current, and conductor stress, as well as the calculation results of coupling parameters V and (x). c y c , z c The data includes S and equipment fault records. After screening, invalid data and samples with more than 10% missing data were removed, retaining 8000 valid samples. The labeling rules are as follows: for faults such as insulator flashover, the severity is labeled as 0.8 to 1.0, minor as 0.8 to 0.9, and severe as 0.9 to 1.0. During normal operation, the health level is labeled as 0 to 0.2, healthy as 0 to 0.1, and good as 0.1 to 0.2. When the critical state parameter is close to the warning threshold, the warning level is labeled as 0.4 to 0.6, general as 0.4 to 0.5, and severe as 0.5 to 0.6. The data are divided into a training set of 5600 groups and a validation set of 2400 groups in a 7:3 ratio.

[0046] Training hyperparameters are set as follows: the initial learning rate is 0.001, employing an adaptive momentum optimization algorithm that automatically decreases by 5% when the bias increases and remains unchanged when the bias decreases, ensuring training convergence; the number of iterations is 100 epochs to balance feature learning and computational efficiency; and the batch size is 32 to balance training efficiency and parameter update stability. Iterative training process: the training set is input in batches according to the batch size. After preprocessing and encoding in the input layer, it is passed to the first hidden layer to extract shallow features. After nonlinear transformation, it is deeply fused by the second layer and finally output to obtain the predicted coupling strength. After each batch calculation, the bias is measured by dividing the sum of the absolute difference between the mean absolute error of the predicted sample value and the label value by the number of samples. Based on backpropagation of the bias, the connection weights of each layer are updated using gradient descent. The update magnitude is positively correlated with the bias and the learning rate, and α, β, and γ are adjusted synchronously. Every 20 iterations, the mean prediction bias is calculated using the validation set to evaluate generalization ability. Early stopping and overfitting control are implemented: if the mean deviation of the validation set decreases by less than 0.001 for 10 consecutive rounds, the model is considered to have converged and training is stopped; if it increases for 3 consecutive rounds, early stopping is initiated, and the parameters with the smallest deviation are retained, including the connection weights of each layer, the correlation matrix M, and α, β, and γ.

[0047] After training, performance was validated using a 2000-set independent test set, and accuracy, recall, and F1 score were calculated. Accuracy was defined as the percentage of samples with errors within ±0.05, recall was defined as the percentage of correctly predicted samples with errors within ±0.05, and the F1 score was the harmonic mean of the two. Preset standards were: accuracy ≥ 90%, recall ≥ 85%, and F1 ≥ 87%. If these standards were not met, the number of neurons in the hidden layers was adjusted (e.g., changing the first layer to 20, the second layer to 16), and the initial learning rate (e.g., 0.0008 or 0.0012), etc., and hyperparameters were retrained until the requirements were met, ultimately resulting in a two-dimensional association model.

[0048] In this embodiment of the invention, by defining the spatial parameter equations of a three-dimensional dynamic ellipsoid and the plane equations of a three-dimensional polyhedron, a precise geometric and mathematical foundation is provided for the spatial correlation analysis of the dynamic evolution characteristics of multiple disasters and the stable state of equipment. By solving the spatial geometric intersection through simultaneous equations, a quantitative coupling calculation of the disaster impact range and the equipment safety boundary is realized, ensuring the rationality and accuracy of the correlation analysis. By extracting coupling parameters such as the intersection volume, the intersection center coordinates, and the intersection shape complexity, the deep correlation between disaster characteristics and equipment status is effectively captured, providing multi-dimensional support for the quantification of coupling relationships. By mapping the coupling parameters to model weight coefficients and correlation matrices, a two-dimensional correlation model integrating the dynamic coupling characteristics of multiple disasters and the vulnerability of equipment throughout its entire life cycle is constructed, effectively breaking through the limitations of single-dimensional analysis and realizing the deep integration of the dynamic evolution of disasters and the long-term operational vulnerability of equipment. This provides reasonable and accurate model support for multi-disaster risk assessment, early warning, and operation and maintenance decisions for power transmission channels.

[0049] In a preferred embodiment of the present invention, step 6 above may include: Step 6.1 involves inputting the current meteorological forecast data and real-time power grid operating parameters into the two-dimensional correlation model. This includes: determining the current time node, which is consistent with the data collection cycle of the real-time power grid monitoring data, typically every 5 minutes, to ensure data timeliness; collecting the current meteorological forecast data, specifically including forecast values ​​for icing thickness, lightning probability and impact range, predicted soil frost heave, wind speed and direction, and other key meteorological indicators related to multiple disasters, sourced from refined grid forecasts issued by regional meteorological departments; simultaneously acquiring real-time power grid operating parameters, covering equipment status data such as vertical settlement and horizontal tilt angle of tower foundations, leakage current and surface contamination of insulators, dynamic stress values ​​of conductors, and real-time sag height; preprocessing the collected meteorological forecast data and real-time power grid operating parameters, removing invalid data due to transmission delays or sensor malfunctions, and standardizing the data format and coordinates to ensure the input data meets the input requirements of the two-dimensional correlation model; and finally, inputting the preprocessed complete data set into the trained two-dimensional correlation model.

[0050] Step 6.2: Based on the model weight coefficients and correlation matrix in the two-dimensional association model, perform fusion calculations on the input data to generate the initial coupling strength at the current time. Specifically, this includes: calling the model weight coefficients and correlation matrix stored in the two-dimensional association model, where the weight coefficients include the intersection volume correlation weight w. V Intersection center distance association weight w d Shape complexity associated weight w S The correlation matrix M, with 3 rows and 4 columns, corresponds to the relationship between three types of disasters—icing, lightning, and soil frost heave—and four equipment vulnerability indicators. Input meteorological forecast data and real-time power grid operating parameters are categorized by data type, mapping meteorological data to the multi-hazard dynamic coupling characteristic dimension and power grid parameters to the equipment lifecycle vulnerability dimension. Data within the same dimension is weighted based on weight coefficients; for example, icing-related meteorological data and power grid parameters are weighted by w... V =0.4、w d =0.3、w S The contribution value is calculated with a weight ratio of 0.3; the two-dimensional data are cross-fused through the correlation matrix M, and the weighted data of the disaster dimension is correlated with the corresponding indicators of the equipment dimension to explore the intrinsic relationship between the two; after nonlinear transformation of the hidden layer of the two-dimensional correlation model and linear mapping of the output layer, the initial coupling strength at the current moment is finally generated, where the initial coupling strength is between 0 and 1, which intuitively reflects the degree of correlation between the current disaster and the equipment status.

[0051] Step 6.3: Based on the initial coupling strength at the current moment, determine whether the meteorological forecast data and real-time power grid operating parameters at the next moment immediately adjacent to the current moment have been updated. Specifically, this includes: using the current moment as a benchmark, determining the next moment node immediately adjacent to the current moment, where the interval between the node and the current moment is also 5 minutes to maintain the continuity of the time series; setting the trigger condition for data update judgment based on the initial coupling strength at the current moment; if the initial coupling strength is greater than 0.6, i.e., above the critical correlation interval, increasing the judgment frequency and shortening the data query interval to once per minute; if the initial coupling strength is less than 0.4, querying at the normal 5-minute interval; querying the update status of the meteorological forecast data and real-time power grid operating parameters at the next moment through the real-time data interface of the data center, judging based on whether the data timestamp has been updated and whether the change in the core indicator value exceeds the preset threshold, where the meteorological data change threshold is 10% and the power grid parameter change threshold is 5%; recording the query results to clarify whether the data at the next moment meets the update conditions.

[0052] Step 6.4: If it is determined that the meteorological forecast data and the real-time operating parameters of the power grid have been updated, the updated meteorological forecast data and the real-time operating parameters of the power grid are input into the two-dimensional correlation model for iterative calculation. Specifically, this includes: If the query result shows that the meteorological forecast data and the real-time operating parameters of the power grid for the next time moment have been updated, that is, the data timestamp has been updated to the next time moment, and the core indicator values ​​have changed beyond the corresponding threshold, it indicates that the disaster evolution state or equipment operating state has changed significantly; immediately obtain the updated complete data set, including the latest meteorological forecast indicators and the real-time operating parameters of the power grid, repeat the preprocessing process of step 6.1, remove invalid data, unify the format and coordinate system to ensure data quality; re-input the preprocessed updated data into the two-dimensional correlation model, which automatically calls the built-in weight coefficients and correlation matrix, re-executes the fusion calculation process, generates the initial coupling strength for the next time moment, and completes one iterative calculation; based on the new initial coupling strength, the judgment process of step 6.3 is triggered again to continuously track the data update status and achieve dynamic iteration.

[0053] Step 6.5: If it is determined that the meteorological forecast data and the real-time operating parameters of the power grid have not been updated, the initial coupling strength at the current moment is output as the stable dynamic coupling strength calculation result. Specifically, if the query result shows that the meteorological forecast data and the real-time operating parameters of the power grid at the next moment have not been updated, that is, the data timestamp is still the current moment, and the core indicator values ​​have not changed beyond the preset threshold, the meteorological data changes have not reached 10%, and the power grid parameters have not changed by 5%, it indicates that the current disaster evolution state and equipment operation state are in a relatively stable stage; the stability of the initial coupling strength at the current moment is verified by retrieving the coupling strength calculation results of the two previous adjacent moments and calculating the absolute difference between the current initial coupling strength and these two results respectively. If both absolute differences are less than 0.01, the current initial coupling strength is determined to be stable and meets the output requirements. If either difference exceeds 0.01, wait for a 5-minute time interval and execute the judgment process in step 6.3 again to confirm whether the data at the next moment has not been updated. If the data is still not updated after the second judgment, recalculate the difference between the current initial coupling strength and the result at the latest adjacent moment until both consecutive differences are less than 0.01, then the stability verification is completed. The initial coupling strength that passes the stability verification is determined as the stable dynamic coupling strength calculation result for the current stage and is synchronously uploaded to the power grid operation and maintenance management platform. The stable dynamic coupling strength calculation result for the current stage will be directly used for multi-hazard risk classification assessment of transmission channels, determination of differentiated early warning thresholds, and priority ranking of operation and maintenance decisions.

[0054] In this embodiment of the invention, by inputting the current meteorological forecast data and the real-time operating parameters of the power grid into a two-dimensional correlation model, and relying on the built-in weight coefficients and correlation matrix of the model for targeted fusion calculation, the initial coupling strength is ensured, which can accurately capture the correlation state between the current multi-hazard evolution and the vulnerability of equipment throughout its entire life cycle. By judging whether the meteorological forecast data and the real-time operating parameters of the power grid in the next moment have been updated, a dynamic adaptation calculation mechanism is constructed. When the parameters are updated, the calculation is re-iterated in a timely manner to keep up with the state changes. When they are not updated, a stable result is directly output to avoid invalid calculations. This ensures both the real-time performance and dynamic adaptability of the coupling strength calculation, while also taking into account the computational efficiency and the stability of the results. The final output of stable dynamic coupling strength provides accurate quantitative support for real-time risk classification of multi-hazards in transmission channels, dynamic adjustment of operation and maintenance decisions, and determination of emergency response timing.

[0055] In a preferred embodiment of the present invention, step 7 above may include: Step 7.1: Compare the stable dynamic coupling strength calculation result with multiple preset risk level thresholds to determine the current risk level. Specifically, this includes: preset coupling strength thresholds corresponding to multiple disaster risk levels, with a low-risk threshold of 0 to 0.4, a critical risk threshold of 0.4 to 0.6, and a high-risk threshold of 0.6 to 1.0. The threshold division is based on historical fault data and operation and maintenance experience calibration; obtain the output stable dynamic coupling strength calculation result, which is a specific value in the range of 0 to 1; compare the stable coupling strength with the preset thresholds one by one to determine its numerical range; if the stable coupling strength is between 0 and 0.4, the current risk level is determined to be low risk, indicating a weak correlation between the disaster and the equipment status; if it is between 0.4 and 0.6, it is determined to be critical risk, and an early warning mechanism needs to be activated; if it is between 0.6 and 1.0, it is determined to be high risk, indicating that the disaster may cause equipment failure; and simultaneously record the current risk level and the corresponding coupling strength value.

[0056] Step 7.2: Based on the current risk level, match the corresponding basic emergency control plan from the plan library. This specifically includes: constructing a structured emergency control plan library, which is divided into three sub-libraries according to risk level: low risk, critical risk, and high risk. Each sub-library contains specific basic plans for three types of disasters: icing, lightning, and soil frost heave. The basic emergency control plans clearly define the core control objectives, applicable scenarios, main execution processes, and basic parameters. For example, the icing-specific plan in the low-risk sub-library aims to strengthen monitoring, the lightning-specific plan in the critical risk sub-library focuses on early warning and patrol, and the soil frost heave-specific plan in the high-risk sub-library emphasizes emergency reinforcement. Based on the determined current risk level, retrieve the corresponding sub-library in the emergency control plan library. Combining the disaster type implied by the stable coupling strength, determine the source through the input data of the two-dimensional association model, and match the most suitable basic emergency control plan from the corresponding sub-library. For example, when the stable coupling strength is 0.7 and icing disaster is dominant, match the icing-specific basic plan from the high-risk sub-library.

[0057] Step 7.3 involves identifying and analyzing the specific disaster coupling characteristics reflected in the dynamic coupling strength calculation results to parametrically adjust the basic emergency control plan, forming a parametrically adjusted emergency control plan. Specifically, this includes: identifying and analyzing the specific disaster coupling characteristics reflected in the stable dynamic coupling strength, clarifying the dominant disaster type (i.e., icing, lightning, or soil frost heave), the concentrated vulnerable parts of equipment (i.e., foundations, insulators, or conductors), and the specific numerical range of the coupling strength; extracting core impact parameters for the dominant disaster type: for icing disasters, focusing on icing thickness and growth rate; for lightning disasters, focusing on lightning frequency and amplitude; and for soil frost heave disasters, focusing on frost heave amount and soil moisture. Based on the identification results, the direction of parameter adjustment is determined. For low-risk levels, the focus is on optimizing the monitoring frequency; for critical-risk levels, the focus is on adjusting the early warning threshold; and for high-risk levels, the focus is on strengthening the control intensity. Key parameters in the basic emergency control plan are quantitatively adjusted. For example, in the high-risk plan dominated by icing, the basic parameter for the start-up time of the ice-melting equipment is 30 minutes, which is adjusted to 45 minutes based on the coupling strength of 0.8. The basic value of the conductor stress control threshold is 150MPa, which is adjusted to 130MPa. The adjustment is based on the principle that the higher the coupling strength, the more optimized the intensity and duration of the control parameters. Different disaster types require different focuses in parameter adjustment to ensure that the plan is adapted to the current actual disaster and equipment status.

[0058] Step 7.4: Based on the parameterized emergency control plan, obtain an emergency control decision instruction set containing execution equipment, action commands, and triggering sequences. Specifically, this includes: parsing the parameterized emergency control plan and extracting the list of execution equipment involved, including ice-melting devices and inspection drones for ice accumulation disasters, lightning protection devices and fault location terminals for lightning disasters, and foundation reinforcement machinery and settlement monitoring equipment for soil frost heave disasters; according to the control process and logic of the plan, formulate specific action commands for each execution equipment, such as the command to activate the DC ice-melting mode for the ice-melting device, the command for the drone to conduct full-coverage inspection of the tower conductors, the command for the lightning protection device to activate the lightning protection gap, and the command for the foundation reinforcement machinery to compact the soil around the foundation; based on the equipment response priority and control sequence requirements, arrange the triggering order of the action commands, triggering the early warning notification command first under high-risk levels, then activating the emergency control equipment, and finally arranging personnel for on-site handling; triggering the inspection command first under critical risk levels, and then initiating subsequent control commands based on the inspection results; and integrating the execution equipment, action commands, and triggering sequences to form a structured emergency control decision instruction set.

[0059] In this embodiment of the invention, by comparing the stable dynamic coupling strength calculation results with multiple preset risk level thresholds, the current multi-hazard risk level of the transmission channel can be accurately determined, providing a clear priority guide for emergency control. Based on the risk level, the corresponding basic emergency control plan is matched from the plan library to ensure the standardization and timeliness of emergency response and avoid disorderly response. Combined with the identification and analysis of the specific disaster coupling characteristics reflected by the dynamic coupling strength, the basic emergency control plan is parametrically adjusted so that the plan can accurately adapt to the actual correlation between the current disaster evolution and equipment status, breaking through the adaptation limitations of general plans. Finally, an emergency control decision instruction set containing execution equipment, action instructions and triggering sequence is formed, realizing the accurate implementation and orderly execution of emergency control measures, effectively improving the emergency response efficiency of the power grid in dealing with multi-hazard risks, the pertinence of control measures and the rationality of operation and maintenance decisions.

[0060] This application also provides a power grid multi-hazard coupling strength calculation system, such as... Figure 2 The diagram shown is a schematic of the system, including: The data acquisition module is used to collect meteorological forecast data and real-time power grid operating parameters of the target power transmission channel; The parsing configuration module is used to parse meteorological forecast data and configure virtual monitoring points at the midpoint of the line section prone to icing, the top of the tower in the lightning-prone area, and the geometric center of the tower foundation in the soil frost heave sensitive area; and generate a dynamic feature point set based on the virtual monitoring points. To analyze real-time power grid operating parameters, a set of equipment status characteristic points is collected at the four corner monitoring points of the tower foundation, the connection between the insulator steel foot and the hanging point, and the outlet of the conductor tension clamp. The point set extraction module is used to extract data point sequences with spatiotemporal continuous evolution characteristics from the dynamic feature point set, which serve as the core trajectory for constructing a three-dimensional dynamic ellipsoid; it also extracts data point clusters from the equipment status feature point set that reflect the stable and healthy state of the equipment, which serve as the reference framework for constructing a three-dimensional polyhedron; the module constructs a three-dimensional dynamic ellipsoid based on the core trajectory, and then constructs a three-dimensional polyhedron based on the reference framework. The model building module is used to calculate the spatial coupling intersection of a three-dimensional dynamic ellipsoid and a three-dimensional polyhedron, and generate coupling parameters that characterize the correlation between disaster characteristics and equipment status. Based on the coupling parameters, a two-dimensional correlation model that integrates the dynamic coupling characteristics of multiple disasters and the vulnerability of equipment throughout its entire life cycle is constructed. The model invocation module is used to invoke the two-dimensional correlation model, which calculates the dynamic coupling strength under the action of disaster chain by real-time iterative fusion of meteorological forecast data and real-time power grid operating parameters, and generates dynamic coupling strength calculation results. The decision generation module is used to generate an emergency control decision instruction set for the target power transmission channel based on the dynamic coupling strength calculation results.

[0061] The computing system according to embodiments of the present invention can correspond to performing the methods described in the embodiments of the present invention, and the above and other operations and / or functions of each module of the computing system are respectively for implementing Figure 1 The corresponding process of the method in the illustrated embodiment will not be described in detail here for the sake of brevity.

[0062] This application also provides a computing device. This computing device can utilize a server.

[0063] like Figure 3 As shown in the figure, this is a schematic diagram of a computing device provided in an embodiment of this application. The computing device 700 includes a bus 701, a processor 702, a communication interface 703, and a memory 704. The processor 702, the memory 704, and the communication interface 703 communicate with each other via the bus 701.

[0064] The 701 bus can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of representation, Figure 3 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0065] The processor 702 can be any one or more of the following processors: central processing unit (CPU), graphics processing unit (GPU), microprocessor (MP), or digital signal processor (DSP).

[0066] Communication interface 703 is used for external communication. Memory 704 may include volatile memory, such as random access memory (RAM). Memory 704 may also include non-volatile memory, such as read-only memory (ROM), flash memory, hard disk drive (HDD), or solid-state drive (SSD). Memory 704 stores executable code, which processor 702 executes to perform the aforementioned method for calculating the coupling strength of multiple hazards in the power grid.

[0067] Specifically, in implementing the power grid multi-hazard coupling strength calculation system described in the above embodiments, and where each module or unit of the power grid multi-hazard coupling strength calculation system described in the above embodiments is implemented by software, the software or program code required to execute the functions of each module / unit in the power grid multi-hazard coupling strength calculation system described in the above embodiments can be partially or entirely stored in the memory 704. The processor 702 executes the program code corresponding to each unit stored in the memory 704 to execute the aforementioned power grid multi-hazard coupling strength calculation method.

[0068] This application also provides a computer-readable storage medium. The computer-readable storage medium can be any available medium capable of being stored by a computing device, or a data storage device such as a data center containing one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid-state drive). The computer-readable storage medium includes instructions that instruct the computing device to execute the above-described method for calculating the coupling strength of multiple disasters in a power grid.

[0069] This application also provides a computer program product comprising one or more computer instructions. When the computer instructions are loaded and executed on a computing device, all or part of the processes or functions described in this application are generated.

[0070] The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions may be transmitted from one website, computer, or data center to another website, computer, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) means.

[0071] When the computer program product is executed by a computer, the computer executes any of the aforementioned methods for calculating the coupling strength of multiple disasters in the power grid. The computer program product can be a software installation package; when any of the aforementioned methods for calculating the coupling strength of multiple disasters in the power grid is required, the computer program product can be downloaded and executed on the computer.

[0072] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for calculating the coupling strength of multiple hazards in a power grid, characterized in that, The method includes: Collect meteorological forecast data and real-time power grid operating parameters for the target power transmission channel; Analyze meteorological forecast data and configure virtual monitoring points at the midpoint of the section of the line prone to icing, the top of the tower in the lightning-prone area, and the geometric center of the tower foundation in the soil frost heave sensitive area; generate a dynamic feature point set based on the virtual monitoring points; To analyze real-time power grid operating parameters, a set of equipment status characteristic points is collected at the four corner monitoring points of the tower foundation, the connection between the insulator steel foot and the hanging point, and the outlet of the conductor tension clamp. Extract data point sequences with spatiotemporal continuous evolution characteristics from the dynamic feature point set as the core trajectory for constructing a three-dimensional dynamic ellipsoid; extract data point clusters reflecting the stable and healthy state of the equipment from the equipment status feature point set as the benchmark framework for constructing a three-dimensional polyhedron; construct a three-dimensional dynamic ellipsoid based on the core trajectory, and construct a three-dimensional polyhedron based on the benchmark framework; Calculate the spatial coupling intersection of a three-dimensional dynamic ellipsoid and a three-dimensional polyhedron to generate coupling parameters characterizing the correlation between disaster characteristics and equipment status; based on the coupling parameters, construct a two-dimensional correlation model that integrates the dynamic coupling characteristics of multiple disasters and the vulnerability of equipment throughout its entire life cycle. The two-dimensional correlation model is invoked to calculate the dynamic coupling strength under the action of disaster chain by iteratively fusing meteorological forecast data and real-time power grid operating parameters in real time, and to generate dynamic coupling strength calculation results. Based on the dynamic coupling strength calculation results, an emergency control decision instruction set for the target power transmission channel is generated.

2. The method for calculating the coupling strength of multiple disasters in a power grid according to claim 1, characterized in that, Analyze meteorological forecast data and configure virtual monitoring points at the midpoint of the section of the line prone to icing, the top of the tower in the lightning-prone area, and the geometric center of the tower foundation in the soil frost heave sensitive area; A dynamic feature point set is generated based on virtual monitoring points, including: Based on the spatial and temporal scope of the disaster defined by meteorological forecast data, virtual icing monitoring points are configured at the midpoint of the icing-prone sections of the line. Data corresponding to the virtual icing monitoring points are extracted to generate a set of icing growth monitoring points. Virtual lightning monitoring points are configured on the top of towers in areas prone to lightning strikes. Data corresponding to these virtual lightning monitoring points is extracted to generate a set of lightning activity monitoring points. Virtual frost heave monitoring points are configured at the geometric center of the tower foundation in the soil frost heave sensitive area. Data corresponding to the virtual frost heave monitoring points are extracted to generate a set of frost heave monitoring points. The monitoring points for ice accumulation growth, lightning activity, and frost heave are aggregated to form an initial dynamic feature point set. From the initial dynamic feature point set, a sequence of data points whose data values ​​exhibit continuous changes in the time and space dimensions is selected as the final dynamic feature point set.

3. The method for calculating the coupling strength of multiple disasters in a power grid according to claim 2, characterized in that, To analyze real-time power grid operating parameters, a set of equipment status characteristic points is collected at the four corner monitoring points of the tower foundation, the connection between the insulator steel foot and the hanging point, and the outlet of the conductor tension clamp. These points include: From the real-time operating parameters of the power grid, the foundation monitoring data corresponding to the monitoring points at the four corners of the tower foundation cap are analyzed to generate a set of cap state points that characterize the foundation bearing state; From the real-time operating parameters of the power grid, the monitoring data of the insulator corresponding to the connection between the insulator steel foot and the hanging point is analyzed to generate a set of insulator state points characterizing insulation performance and mechanical fatigue; From the real-time operating parameters of the power grid, analyze the conductor monitoring data corresponding to the outlet of the conductor tension clamp to generate a set of conductor state points characterizing the conductor stress and sag state; The aggregated pile cap state point set, insulator state point set, and conductor state point set are used to form the initial equipment state characteristic point set; From the initial set of device status feature points, a cluster of stable data points whose data fluctuation range is below a set threshold within a preset time period is selected as the final set of device status feature points.

4. The method for calculating the coupling strength of multiple disasters in a power grid according to claim 3, characterized in that, Extract data point sequences with spatiotemporal continuous evolution characteristics from the dynamic feature point set as the core trajectory for constructing a three-dimensional dynamic ellipsoid; extract data point clusters reflecting the stable and healthy state of equipment from the equipment status feature point set as the benchmark framework for constructing a three-dimensional polyhedron. A three-dimensional dynamic ellipsoid is constructed based on the core trajectory, and a three-dimensional polyhedron is constructed based on the benchmark framework, including: Spatiotemporal trajectory fitting is performed on the data point sequence in the final dynamic feature point set to form the core trajectory characterizing the disaster evolution path; By analyzing and identifying the extension direction and data density distribution of the core trajectory, the principal axis direction and semi-axis length of the three-dimensional dynamic ellipsoid can be determined. Based on the principal axis direction and the semi-axis length, a three-dimensional dynamic ellipsoid is constructed; By mapping the stable data point clusters in the final device state feature point set to three-dimensional space, the vertex set of the solid polyhedron can be determined. Based on the spatial topological relationships defined in the vertex set, the vertices are connected sequentially to generate a reference frame for constructing a solid polyhedron; based on the reference frame, the solid polyhedron is constructed.

5. The method for calculating the coupling strength of multiple disasters in a power grid according to claim 4, characterized in that, Calculate the spatial coupling intersection of a three-dimensional dynamic ellipsoid and a three-dimensional polyhedron to generate coupling parameters characterizing the correlation between disaster characteristics and equipment status; Based on coupling parameters, a two-dimensional correlation model is constructed that integrates the dynamic coupling characteristics of multiple disasters and the vulnerability of equipment throughout its entire lifecycle, including: Based on the principal axis direction and semi-axis length of the three-dimensional dynamic ellipsoid, the spatial parametric equation of the three-dimensional dynamic ellipsoid is defined; Based on the reference frame of the solid polyhedron and its vertex set, determine the equations of each plane that constitutes the solid polyhedron; The spatial geometric intersection is obtained by simultaneously solving the parametric equations of the three-dimensional dynamic ellipsoid and the plane equations of the three-dimensional polyhedron. By performing topological and metric analysis on the spatial geometric intersection, the intersection volume, intersection center coordinates, and intersection shape complexity index are extracted as coupling parameters characterizing the correlation between disaster characteristics and equipment status. The coupling parameters are mapped to model weight coefficients and correlation matrices; based on the model weight coefficients and correlation matrices, a two-dimensional correlation model is constructed that integrates the dynamic coupling characteristics of multiple disasters and the vulnerability of equipment throughout its entire life cycle.

6. The method for calculating the coupling strength of multiple disasters in a power grid according to claim 5, characterized in that, A two-dimensional correlation model is invoked to calculate the dynamic coupling strength under the influence of disaster chains by iteratively fusing meteorological forecast data with real-time power grid operating parameters. The calculation results include: The meteorological forecast data acquired at the current moment and the real-time operating parameters of the power grid are input into the two-dimensional correlation model; Based on the model weight coefficients and correlation matrix in the two-dimensional association model, the input data is fused and calculated to generate the initial coupling strength at the current moment. Based on the initial coupling strength at the current moment, determine whether the meteorological forecast data and real-time power grid operating parameters at the next moment immediately adjacent to the current moment have been updated; If it is determined that the meteorological forecast data and the real-time operating parameters of the power grid have been updated, the updated meteorological forecast data and the real-time operating parameters of the power grid will be input into the two-dimensional correlation model for iterative calculation again. If it is determined that the meteorological forecast data and the real-time operating parameters of the power grid have not been updated, the initial coupling strength at the current moment is output as the result of the stable dynamic coupling strength calculation.

7. The method for calculating the coupling strength of multiple disasters in a power grid according to claim 6, characterized in that, Based on the dynamic coupling strength calculation results, an emergency control decision instruction set for the target transmission channel is generated, including: The stable dynamic coupling strength calculation results are compared with multiple preset risk level thresholds to determine the current risk level; Based on the current risk level, match the corresponding basic emergency response plan from the contingency plan database; By identifying and analyzing the specific disaster coupling characteristics reflected in the dynamic coupling strength calculation results, the basic emergency control plan is parametrically adjusted to form a parametrically adjusted emergency control plan. Based on the parameterized emergency control plan, an emergency control decision instruction set containing execution equipment, action commands, and triggering sequences is obtained.

8. A power grid multi-hazard coupling strength calculation system, characterized in that, The system performs the method as described in any one of claims 1 to 7, comprising: The data acquisition module is used to collect meteorological forecast data and real-time power grid operating parameters of the target power transmission channel; The parsing configuration module is used to parse meteorological forecast data and configure virtual monitoring points at the midpoint of the line section prone to icing, the top of the tower in the lightning-prone area, and the geometric center of the tower foundation in the soil frost heave sensitive area; and generate a dynamic feature point set based on the virtual monitoring points. To analyze real-time power grid operating parameters, a set of equipment status characteristic points is collected at the four corner monitoring points of the tower foundation, the connection between the insulator steel foot and the hanging point, and the outlet of the conductor tension clamp. The point set extraction module is used to extract data point sequences with spatiotemporal continuous evolution characteristics from the dynamic feature point set, which serve as the core trajectory for constructing a three-dimensional dynamic ellipsoid; it also extracts data point clusters from the equipment status feature point set that reflect the stable and healthy state of the equipment, which serve as the reference framework for constructing a three-dimensional polyhedron; the module constructs a three-dimensional dynamic ellipsoid based on the core trajectory, and then constructs a three-dimensional polyhedron based on the reference framework. The model building module is used to calculate the spatial coupling intersection of a three-dimensional dynamic ellipsoid and a three-dimensional polyhedron, and generate coupling parameters that characterize the correlation between disaster characteristics and equipment status. Based on the coupling parameters, a two-dimensional correlation model that integrates the dynamic coupling characteristics of multiple disasters and the vulnerability of equipment throughout its entire life cycle is constructed. The model invocation module is used to invoke the two-dimensional correlation model, which calculates the dynamic coupling strength under the action of disaster chain by real-time iterative fusion of meteorological forecast data and real-time power grid operating parameters, and generates dynamic coupling strength calculation results. The decision generation module is used to generate an emergency control decision instruction set for the target power transmission channel based on the dynamic coupling strength calculation results.

9. A computing device, characterized in that, Including memory and processor; The memory stores one or more computer programs, the one or more computer programs including instructions; when the instructions are executed by the processor, the computing device performs the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium is used to store a computer program for performing the method as described in any one of claims 1 to 7.

Images