A knowledge graph-based severe weather power grid risk identification method and system

Abstract

The application discloses a kind of based on knowledge graph's severe weather power grid risk identification method and system, method includes: through the acquisition target area in multi-source weather forecast data and power grid topology structure data, carry out data cleaning and space-time calibration, utilize knowledge graph technology fusion meteorological data and power grid state data, generate physical risk correlation subgraph and cascading risk correlation subgraph, respectively using probability simulation method and graph theory cascading failure analysis method analysis power grid facility's physical damage risk and failure cascading diffusion risk, and determine the comprehensive risk grade of power grid, and according to the comprehensive risk grade of power grid generates power grid risk thermodynamic diagram.Power grid regional risk grade is realized visual expression, provides decision-making basis for power grid dispatching operation and pre-disaster warning.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of power grid risk identification, and particularly relates to a severe weather power grid risk identification method and system based on a knowledge graph. BACKGROUND

[0002] Currently, when facing extreme weather conditions, the power grid mainly relies on traditional weather monitoring and manual experience for risk analysis, lacks analysis of the complex correlation between multi-source weather elements and power grid facility operation states, and is difficult to accurately identify potential physical damage risks and cascading failure diffusion trends in a timely manner, resulting in delayed early warning response and failing to meet the actual needs of power grid risk identification and regional early warning under severe weather. SUMMARY

[0003] The application provides a severe weather power grid risk identification method and system based on a knowledge graph and a readable storage medium, which are used to solve the technical problem of delayed early warning response and failing to meet the actual needs of power grid risk identification and regional early warning under severe weather.

[0004] In a first aspect, the application provides a severe weather power grid risk identification method based on a knowledge graph, comprising:

[0005] Data cleaning and space-time calibration are performed on multi-source weather forecast data and power grid topology structure data in a target area to obtain weather data and power grid state data;

[0006] The weather data and power grid state data are fused based on a preset knowledge graph, weather features, associated node relationships of power grid facilities, and risk transmission paths are extracted, physical risk association subgraphs and cascading risk association subgraphs are generated;

[0007] Based on the physical risk association subgraphs, the probability simulation is adopted to analyze the tower collapse probability and the conductor dancing amplitude, and the physical damage risk of the power grid facilities under severe weather conditions is obtained;

[0008] Based on the cascading risk association subgraphs, the graph theory cascading failure analysis is combined to deduce the power grid cascading failure propagation range, and the failure cascading diffusion risk of the power grid facilities under severe weather conditions is obtained;

[0009] Based on the physical damage risk and the failure cascading diffusion risk of the power grid facilities under severe weather conditions, the comprehensive risk level of the power grid is determined;

[0010] A power grid risk heat map is generated according to the comprehensive risk level of the power grid.

[0011] In a second aspect, the application provides a severe weather power grid risk identification system based on a knowledge graph, comprising:

[0012] a data processing module configured to perform data cleaning and space-time calibration on multi-source weather forecast data and power grid topology data in a target area to obtain weather data and power grid state data;

[0013] a first generation module configured to fuse the weather data and the power grid state data based on a preset knowledge graph, extract weather features, associated node relationships of power grid facilities, and risk transmission paths, and generate a physical risk association subgraph and a cascading risk association subgraph;

[0014] a first analysis module configured to analyze tower collapse probability and conductor dancing amplitude based on the physical risk association subgraph by using probability simulation to obtain physical damage risk of power grid facilities under adverse weather conditions;

[0015] a second analysis module configured to deduce power grid cascading failure propagation range based on the cascading risk association subgraph by combining graph theory cascading failure analysis to obtain failure cascading diffusion risk of power grid facilities under adverse weather conditions;

[0016] a determination module configured to determine a comprehensive risk level of the power grid based on the physical damage risk and the failure cascading diffusion risk of the power grid facilities under adverse weather conditions;

[0017] a second generation module configured to generate a power grid risk heat map according to the comprehensive risk level of the power grid.

[0018] In a third aspect, an electronic device is provided, which includes at least one processor and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform steps of the method for identifying power grid risk under adverse weather based on a knowledge graph according to any one of the embodiments.

[0019] In a fourth aspect, the present application further provides a computer readable storage medium having a computer program stored thereon, and the program instructions are executed by a processor to enable the processor to perform steps of the method for identifying power grid risk under adverse weather based on a knowledge graph according to any one of the embodiments.

[0020] The knowledge graph-based severe weather power grid risk identification method and system of the application, by collecting multi-source weather forecast data and power grid topology structure data and performing cleaning and space-time calibration, ensures the timeliness and spatial consistency of the input data, improves the accuracy of risk analysis; by introducing a knowledge graph to fuse weather characteristics and power grid facilities, a physical risk correlation subgraph and a cascading risk correlation subgraph are constructed, realizing the structured expression of the complex correlation between weather and power grid elements; based on the physical risk correlation subgraph, probability simulation is carried out, which can quantitatively analyze the tower collapse probability and conductor dancing amplitude, so as to identify the physical damage risk of power grid facilities under severe weather; through the graph theory cascading failure analysis method, the fault cascading diffusion risk is analyzed, and the prediction ability of the fault cascading diffusion trend is improved; the physical damage risk and the fault cascading diffusion risk are comprehensively analyzed, the comprehensive risk level of the power grid is analyzed, and the power grid risk heat map is generated, realizing the visual expression of the power grid regional risk level, and providing decision basis for power grid dispatching operation and pre-disaster warning. BRIEF DESCRIPTION OF DRAWINGS

[0021] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiment description. Obviously, the drawings in the following description are some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without creative labor.

[0022] Figure 1 A flow chart of a knowledge graph-based severe weather power grid risk identification method provided by an embodiment of the present application;

[0023] Figure 2 A structural block diagram of a knowledge graph-based severe weather power grid risk identification system provided by an embodiment of the present application;

[0024] Figure 3 A structural schematic diagram of an electronic device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0025] In order to make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the following will combine the drawings in the embodiments of the present application to clearly and completely describe the technical scheme in the embodiments of the present application. Obviously, the described embodiments are some embodiments of the present application, not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the protection scope of the present application.

[0026] Please refer to Figure 1 which shows a flow chart of a knowledge graph-based severe weather power grid risk identification method of the present application.

[0027] AsFigure 1 As shown, the knowledge graph-based severe weather power grid risk identification method specifically includes the following steps:

[0028] Step S101, data cleaning and space-time calibration are performed on the multi-source weather forecast data and power grid topology structure data in the target area to obtain weather data and power grid state data.

[0029] In this step, multi-source weather forecast data covering the target area are obtained.

[0030] Specifically, a plurality of meteorological monitoring stations are set up in the target area, and each meteorological monitoring station regularly collects wind speed data, precipitation data, temperature data, humidity data, and lightning activity data. Specifically, wind speed data is recorded by an anemometer in real time to record wind speed and direction; precipitation data is measured by an automatic rain gauge to determine the cumulative amount of precipitation; temperature data and humidity data are automatically measured by temperature and humidity sensors at fixed time intervals; lightning activity data is captured and recorded by lightning monitoring instruments installed at the meteorological monitoring stations.

[0031] Obtain the power grid topology structure data matching the target area.

[0032] Specifically, the topology structure data of all power grid facilities in the target area are obtained through a power grid geographic information system, including the specific spatial position of the transmission line, the type of the tower (such as a steel tower or a concrete tower), the specific number of the tower, the position of the transformer substation, and the connection relationship between the transformer substations. After obtaining the topology structure data of the power grid facilities, the data are saved in the form of a vector diagram.

[0033] The multi-source weather forecast data and the power grid topology structure data are respectively subjected to data cleaning processing.

[0034] Specifically, the multi-source weather forecast data are subjected to data cleaning algorithm to automatically detect abnormal values or missing values in the wind speed data, precipitation data, temperature data, humidity data, and lightning activity data. The detection of abnormal values adopts a statistical abnormal value detection method, such as the quartile method or the standard deviation threshold method. After detecting abnormal data, the data are corrected according to a data interpolation method, for example, linear interpolation or adjacent station weighted interpolation is used to fill and correct the wind speed, precipitation, temperature, humidity data, and lightning activity data. The abnormal data that cannot be effectively filled are deleted.

[0035] For the power grid topology data, a spatial data cleaning algorithm is used to detect redundant data and logical contradiction data in the transmission line position data, tower type data, tower number data, substation position data and connection relationship data. Specifically, redundant data is removed by a spatial data deduplication algorithm to remove duplicate points or duplicate tower numbers; logical contradiction data is corrected by a topological relationship checking algorithm, for example, the accuracy of the connection relationship between the substation and the transmission line is analyzed for spatial topological consistency and corrected.

[0036] The cleaned multi-source weather forecast data and the power grid topology data are subjected to spatio-temporal calibration to generate the weather data and the power grid state data.

[0037] Specifically, the cleaned weather forecast data and the power grid topology data are subjected to time calibration. Specifically, the different weather station data in the weather forecast data are unified to the same time scale, for example, every hour is taken as a unified timestamp, to ensure that the data of all weather stations have unified start and end times. The state acquisition time of each power grid facility in the power grid topology data is aligned with the timestamp of the weather forecast data to maintain consistency in the time dimension.

[0038] The implementation of spatial calibration is to use a spatial coordinate transformation method to unify the spatial coordinate positions of each weather monitoring station to the same spatial coordinate system as the power grid facility topology structure, for example, to use the geographic coordinate system or the Gauss projection coordinate system, to ensure the spatial coordinate consistency of the weather data and the power grid state data, thereby ensuring accurate matching and fusion.

[0039] In step S102, the weather data and the power grid state data are fused based on a preset knowledge graph to extract weather features, associated node relationships of power grid facilities and risk transmission paths to generate a physical risk association subgraph and a cascading risk association subgraph.

[0040] In this step, according to the preset knowledge graph, the weather data is taken as a weather node and the power grid state data is taken as a power grid facility node.

[0041] Specifically, the weather data of the target region is taken as a weather node. The wind speed data, precipitation data, temperature data, humidity data and lightning activity data collected in the weather data are respectively represented as weather nodes in the knowledge graph, and each weather node defines a determined data attribute, for example, the wind speed node takes the wind speed and wind direction as attributes; the precipitation node takes the cumulative precipitation in a unit time as an attribute; the temperature node and the humidity node take the measured temperature value and humidity value as attributes; the lightning activity node takes the lightning occurrence frequency as an attribute.

[0042] The power grid state data is taken as a power grid facility node, and attribute definitions of the power grid facility node are determined. The transmission line location node is represented by a specific latitude and longitude coordinate position; the tower type node is represented by a material type of a tower, such as a tower type or a concrete tower type; the tower number node is represented by a unique number in the power grid facility topology data; and the substation location node is represented by a latitude and longitude position and a range to which it belongs. The above nodes are respectively represented as corresponding data nodes in the knowledge graph.

[0043] An association relationship between the meteorological nodes and the power grid facility nodes is established by an association analysis method.

[0044] Specifically, based on the meteorological nodes and the power grid facility nodes, an association analysis algorithm is selected to establish the association relationship between the nodes. A statistical-based association rule analysis algorithm is used to realize establishment of the association relationship between the nodes. The algorithm is configured to set two optimization parameters of a minimum support degree and a minimum confidence degree, to establish frequent association relationships between the wind speed node, the precipitation node, the lightning activity node, and the tower type node, the tower number node, and the transmission line location node by calculating frequencies at which the meteorological nodes and the power grid facility nodes appear together. A formula for calculating the association strength is represented by a number of times at which the nodes appear together divided by a total number of times at which the meteorological nodes appear. The association analysis result is used to determine a risk transmission relationship between the nodes.

[0045] A risk transmission path between the meteorological data and the power grid state data is identified.

[0046] Specifically, based on the established association relationship between the meteorological nodes and the power grid facility nodes, a specific path of risk transmission is identified based on a graph path analysis method. For example, if the wind speed node has a high association strength with a specific tower type node, a risk transmission path of the increased wind speed through the wind speed node to the tower type node is identified. A breadth-first search or a depth-first search path identification method is used to traverse the association relationship network between the nodes to determine the risk transmission path.

[0047] A physical risk association subgraph and a cascading risk association subgraph are constructed based on the association relationship and the risk transmission path.

[0048] Specifically, the physical risk association subgraph and the cascading risk association subgraph are constructed by a knowledge graph construction technology. The nodes of the physical risk association subgraph are composed of the meteorological nodes and the power grid facility nodes directly affected by the meteorological conditions, and the edges between the nodes are represented by the association relationship obtained by the association analysis. The cascading risk association subgraph extends the association relationship between the power grid facility nodes in the physical risk association subgraph to represent the risk transmission path of the fault cascade between the nodes according to the power grid topology connection relationship.

[0049] Step S103, based on the physical risk association subgraph, the probability simulation is used to analyze the tower collapse probability and the conductor dancing amplitude to obtain the physical damage risk of the power grid facility under severe weather conditions.

[0050] In this step, the physical risk association subgraph is taken as the analysis object.

[0051] Specifically, the physical risk association subgraph is composed of meteorological nodes and power grid facility nodes in the target area, and each node is connected through an association relationship, which represents the interaction between the meteorological nodes and the power grid facility nodes. Through the knowledge graph database retrieval function, the wind speed, precipitation, temperature, humidity, and lightning activity meteorological nodes in the target area and the power grid facility nodes directly affected by the meteorological nodes, such as transmission line location nodes, tower type nodes, tower number nodes, and substation location nodes, are extracted from the constructed physical risk association subgraph as initial input objects for the analysis process.

[0052] The tower collapse probability and the conductor dancing amplitude are selected as analysis indicators.

[0053] Specifically, according to the characteristics of power grid facilities that are easily affected by meteorological factors such as wind speed and precipitation in actual operation, the tower collapse probability and the conductor dancing amplitude are determined as analysis indicators. The tower collapse probability represents the possibility of tower collapse under given wind speed and precipitation conditions, and the conductor dancing amplitude represents the amplitude of transmission conductor vibration or oscillation under the action of specific meteorological conditions. The selection of tower collapse probability and conductor dancing amplitude is based on the actual needs of safe operation of power grid and analysis of power grid accident cases.

[0054] According to the association relationship between the meteorological nodes and the power grid facility nodes in the physical risk association subgraph, the probability model of the tower collapse probability and the conductor dancing amplitude is established.

[0055] Specifically, the construction method of the tower collapse probability model is as follows: the occurrence of tower collapse is regarded as a probability event, and a probability regression analysis method is used to establish the model. The model input is the wind speed node data and the precipitation node data, as well as the tower type node data directly associated with the nodes, and the model output is the occurrence probability of tower collapse under given meteorological conditions. The occurrence probability is obtained by calculating the ratio between the number of similar meteorological conditions and the corresponding tower collapse events and the total number of historical data, and the parameters of the probability model are determined by fitting the historical data.

[0056] The construction of the conductor dancing amplitude model is as follows: according to the wind speed data in the meteorological node and the grid facility node attributes such as the conductor specification, the length of the power transmission line, and the tower spacing, a function relationship for calculating the conductor dancing amplitude is established based on the theory of aerodynamics. The conductor dancing amplitude calculation adopts an empirical formula, and the wind speed, the conductor material type, the conductor cross-sectional area, and the distance between the towers are used as independent variables to calculate the theoretical vibration amplitude of the conductor. All the parameters in the model are determined based on historical data or field measurement data.

[0057] According to the tower collapse probability and the probability model of the conductor dancing amplitude, the tower collapse probability and the conductor dancing amplitude are calculated by using the probability simulation method.

[0058] Specifically, the Monte Carlo probability simulation method is used to input the parameters of the probability model, and the tower collapse and conductor dancing under meteorological conditions are simulated multiple times. In each simulation calculation, the wind speed and precipitation data are randomly extracted as initial conditions to input the tower collapse probability model, and a large number of simulations are repeatedly performed to obtain a stable tower collapse probability value. In the simulation calculation process of the conductor dancing amplitude, the Monte Carlo method is also used to randomly extract wind speed data, and the conductor dancing amplitude is calculated in combination with the conductor parameters. The final calculation result is ensured to be stable and reliable by increasing the simulation times sufficiently.

[0059] Based on the tower collapse probability and the conductor dancing amplitude, the physical damage risk of the grid facility under severe weather conditions is determined.

[0060] Specifically, the tower collapse probability and the conductor dancing amplitude obtained by simulation are defined as damage risk factors, respectively. The physical damage risk of the grid facility is calculated by a weighted comprehensive method. For example, the weight of the tower collapse risk factor is determined according to the influence degree of the tower collapse on the operation of the power grid; and the weight of the conductor dancing risk factor is determined according to the short-circuit fault risk of the power grid that may be caused by the conductor dancing. The calculation formula of the physical damage risk is the product of the tower collapse probability and the tower collapse risk weight plus the product of the conductor dancing amplitude and the conductor dancing risk weight, so as to determine the physical damage risk value of the grid facility.

[0061] In step S104, based on the cascading risk correlation subgraph, the cascading failure propagation range of the power grid is deduced by combining the graph theory cascading failure analysis, and the failure cascading diffusion risk of the grid facility under severe weather conditions is obtained.

[0062] In this step, the cascading risk correlation subgraph is taken as the analysis object, and the cascading failure propagation range of the grid facility is taken as the analysis index.

[0063] Specifically, the cascading risk correlation subgraph is composed of meteorological nodes, power grid facility nodes and risk transmission paths between nodes. The cascading risk correlation subgraph represents the failure propagation characteristics of power grid facilities affected by meteorological factors. Power grid facility failures have propagation characteristics, and the propagation range determines the scale of the overall power grid affected, so the cascading failure propagation range of the power grid facility is selected as an analysis index.

[0064] According to the risk transmission path between the meteorological nodes and the power grid facility nodes in the cascading risk correlation subgraph, a graph theory cascading failure analysis method is used to identify the risk transmission path between the meteorological data and the power grid state data.

[0065] Specifically, the network cascading failure method in graph theory is configured to analyze the process, and the risk transmission path between nodes is determined through the cascading risk correlation subgraph. The configuration of the network cascading failure analysis method includes determining the initial failure node and the failure propagation rule. The initial failure node is the power grid facility node associated with a specific meteorological node, such as a specific tower number node associated with a wind speed or lightning activity node. The failure propagation rule is that when a certain power grid facility node fails, the nodes topologically connected to it have a certain probability of also failing, which is represented by the node association failure probability determined by historical data. By traversing all nodes in the entire graph network, all possible risk transmission paths are determined to ensure the comprehensiveness of the failure propagation path.

[0066] Based on the risk transmission path, a cascading failure propagation model of the power grid facility is established.

[0067] Specifically, based on the rules of node failure propagation, a propagation probability calculation method is used to establish the cascading failure propagation model. In the cascading failure propagation model, each node is assigned a failure state and a propagation probability. The node failure state is normal operation or failure. The propagation probability between nodes is the possibility that a power grid facility node fails after a neighboring and topologically connected power grid facility node fails. The calculation method of the propagation probability is to divide the number of simultaneous node failures in the historical record by the total number of initial node failures alone. The cascading failure propagation model defines the probability relationship and propagation link of failure propagation to accurately simulate the power grid facility failure propagation process in real situations.

[0068] According to the cascading failure propagation model, the cascading failure propagation range of the power grid facility under adverse weather conditions is calculated.

[0069] Specifically, the network fault propagation simulation method is used to calculate the range of cascading failure propagation, specifically the network propagation simulation method, and the simulation process is as follows: a specific initial fault node is randomly or specifically selected, and the fault propagation process is simulated step by step according to the propagation rules of the fault propagation model. In each step of fault propagation, it is calculated whether the adjacent nodes of the current fault node also enter the fault state. After multiple simulations, the proportion of the total number of all power grid facility nodes affected by the cascading failure to the total number of all power grid facility nodes is calculated as a quantitative indicator of the range of cascading failure propagation. Through a large number of simulation calculations, a stable result of the range of cascading failure propagation is obtained.

[0070] Based on the range of cascading failure propagation of the power grid facility, the risk of fault cascade diffusion of the power grid facility under severe weather conditions is determined.

[0071] Specifically, the range of cascading failure propagation calculated is used as a risk quantitative indicator to analyze the size of the fault cascade diffusion risk of the power grid facility. The analysis method of the fault cascade diffusion risk is to calculate the risk weight multiplied by the diffusion range ratio, and the risk weight is a numerical value explicitly assigned according to the influence degree of cascading failure on the safe operation of the power grid. The calculation formula of risk level quantification is: the fault cascade diffusion risk of the power grid facility is equal to the range of cascading failure propagation multiplied by the risk weight.

[0072] Step S105, based on the physical damage risk and the fault cascade diffusion risk of the power grid facility under severe weather conditions, determine the comprehensive risk level of the power grid.

[0073] In this step, according to the tower collapse probability and conductor dancing amplitude in the physical damage risk of the power grid facility and the range of cascading failure propagation in the fault cascade diffusion risk of the power grid facility, a power grid comprehensive risk index system is constructed.

[0074] Specifically, the power grid comprehensive risk analysis index system includes three first-level indexes: tower collapse probability index, conductor dancing amplitude index and cascading failure propagation range index, and each first-level index is divided into multiple second-level indexes. For example, the tower collapse probability index is refined into three second-level indexes: light collapse probability, medium collapse probability and heavy collapse probability; the conductor dancing amplitude index is refined into three second-level indexes: light dancing amplitude, medium dancing amplitude and severe dancing amplitude; the cascading failure propagation range index is refined into three second-level indexes: small range propagation, medium range propagation and large range propagation. The construction of the index system is based on the actual operation data of the power grid and the statistical results of historical fault data.

[0075] Based on the power grid comprehensive risk analysis index system, a power grid comprehensive risk level analysis model is established, and the comprehensive risk level of the power grid is determined according to the power grid comprehensive risk level analysis model.

[0076] Specifically, the power grid comprehensive risk level analysis model adopts a comprehensive analysis algorithm combining the analytic hierarchy process and the fuzzy comprehensive evaluation method. The analytic hierarchy process is used to determine the weights of the first-level indexes and the second-level indexes in the power grid comprehensive risk analysis index system. The process of weight determination is as follows: a group of experts in the field of power grid operation compares all indexes two by two and gives a judgment matrix of relative importance. The relative weights of the indexes in the judgment matrix are calculated by the eigenvalue solution method. The calculation method is to solve the eigenvector corresponding to the maximum eigenvalue of the judgment matrix, and the normalized weight values of the indexes are obtained.

[0077] Based on the determined weights, the power grid comprehensive risk level analysis model is constructed by combining the fuzzy comprehensive evaluation method. When the model is constructed, the membership functions are set, for example, five risk levels of extremely low risk, low risk, medium risk, high risk and extremely high risk are set for the tower collapse probability. The membership degree is calculated by the difference between the value of each risk index and the boundary of the given risk level in the membership function.

[0078] Taking the power grid facilities in the target area as the analysis object, the index data in the comprehensive risk analysis index system are input into the power grid comprehensive risk level analysis model. According to the determined index weights and the membership degree matrix, the fuzzy comprehensive operation is performed. The fuzzy comprehensive operation method is to use the weighted average method to calculate the comprehensive risk level analysis result. The calculation method is as follows: the fuzzy operation is performed on each first-level index, and then the corresponding weights are weighted to obtain the comprehensive risk score of the first-level index; the comprehensive risk scores of all first-level indexes are weighted and summed again to determine the overall power grid comprehensive risk score. According to the comprehensive risk score, the risk level is divided, for example, the higher the comprehensive risk score value, the higher the risk level, and finally the comprehensive risk level analysis result of the power grid in the target area under adverse weather conditions is obtained.

[0079] In step S106, a power grid risk heat map is generated according to the comprehensive risk level of the power grid.

[0080] In this step, based on the comprehensive risk level of the power grid in the target area, the target area is divided into multiple spatial units.

[0081] Specifically, based on the geographical range of the target area, the spatial grid division method commonly used in geographic information systems is adopted to divide the target area regularly according to the latitude and longitude. The size of the spatial unit is determined by the grid side length to ensure that each spatial unit can fully reflect the spatial characteristics of the risk distribution of the power grid facilities. The spatial unit division method is as follows: the boundary latitude and longitude range of the target area is determined; according to the actual area size and spatial accuracy requirement of the target area, the grid side length is calculated, for example, the grid side length is determined by dividing the total area of the target area by the desired number of spatial units and then taking the square root, and then the spatial range and boundary coordinates of all spatial units of the target area are obtained.

[0082] Each spatial unit is assigned a risk level corresponding to the comprehensive risk level.

[0083] Specifically, based on the comprehensive risk level analysis result of the power grid, the comprehensive risk level of the power grid facilities within each spatial unit is taken as the risk level of the spatial unit. If multiple power grid facilities are included in the spatial unit, the average value of the comprehensive risk levels of the multiple power grid facilities is taken as the representative value of the risk level of the spatial unit. Specifically, the risk level values of all power grid facilities within the spatial unit are summed and divided by the total number of power grid facilities within the unit to obtain the risk level value of the spatial unit. Based on the risk level corresponding to the comprehensive risk level analysis result of the power grid, the calculated value of each spatial unit is classified in intervals, and the risk level category is assigned.

[0084] Based on the risk level of the spatial unit, a spatial interpolation method is used to generate a power grid risk heat map.

[0085] Specifically, the spatial interpolation method uses the Kriging interpolation algorithm to generate the power grid risk heat map. The Kriging interpolation method is as follows: taking the risk level of the spatial unit as the known data point, a Kriging interpolation model is constructed. The construction process of the Kriging interpolation model includes determination of the semi-variogram function, calculation of the fitting parameters, and solution of the interpolation point weight. The semi-variogram function is determined by calculating the square value of the difference between the risk level values of any two spatial units and their spatial distance relationship; the model parameters of the semi-variogram function are determined by the least squares method; and the interpolation point weight is obtained by solving the semi-variogram function matrix equation.

[0086] Using the Kriging interpolation model, the risk level of the center point of each spatial unit in the target region is calculated by interpolation. The formula for interpolation calculation is: the risk level value of the interpolation point is the weighted sum of all known spatial unit risk level values multiplied by the corresponding weight, and the weight value is determined according to the semi-variogram function. After all the interpolation calculations for the target region are completed, the risk level values obtained by the interpolation point calculation are mapped to specific color representations, for example, using a gradual color representation from low risk level to high risk level, such as green, yellow, orange, red, and purple, to represent the risk level.

[0087] In summary, the method of the present application collects multi-source weather forecast data and power grid topology data, and performs cleaning and space-time calibration to ensure the timeliness and spatial consistency of the input data, thereby improving the accuracy of risk analysis. By introducing a knowledge graph to integrate weather characteristics and power grid facilities, a physical risk correlation subgraph and a cascading risk correlation subgraph are constructed to realize the structured expression of complex correlations between weather and power grid elements. Based on the physical risk correlation subgraph, probability simulation is carried out to quantitatively analyze the probability of tower collapse and the amplitude of conductor dancing, thereby identifying the physical damage risk of power grid facilities under adverse weather conditions. Through the graph theory cascading failure analysis method, the risk of fault cascade diffusion is analyzed to improve the prediction ability of fault cascade diffusion trend. By integrating the physical damage risk and the fault cascade diffusion risk, the comprehensive risk level of the power grid is analyzed, and a power grid risk heat map is generated to realize the visual expression of the risk level of the power grid region, thereby providing a decision basis for power grid dispatching and pre-disaster warning.

[0088] Please refer to Figure 2 which shows a structure block diagram of a severe weather power grid risk identification system based on a knowledge graph.

[0089] As Figure 2 shown, the severe weather power grid risk identification system 200 includes a data processing module 210, a first generation module 220, a first analysis module 230, a second analysis module 240, a determination module 250, and a second generation module 260.

[0090] The data processing module 210 is configured to perform data cleaning and space-time calibration on multi-source weather forecast data and power grid topology data in a target region to obtain weather data and power grid state data. The first generation module 220 is configured to fuse the weather data and power grid state data based on a preset knowledge graph to extract weather characteristics, correlation node relationships of power grid facilities, and risk transmission paths, and to generate a physical risk correlation subgraph and a cascading risk correlation subgraph. The first analysis module 230 is configured to analyze the probability of tower collapse and the amplitude of conductor dancing based on the physical risk correlation subgraph using probability simulation to obtain the physical damage risk of power grid facilities under adverse weather conditions. The second analysis module 240 is configured to deduce the propagation range of cascading power grid failures based on the cascading risk correlation subgraph combined with graph theory cascading failure analysis to obtain the fault cascade diffusion risk of power grid facilities under adverse weather conditions. The determination module 250 is configured to determine the comprehensive risk level of the power grid based on the physical damage risk and the fault cascade diffusion risk of the power grid facilities under adverse weather conditions. The second generation module 260 is configured to generate a power grid risk heat map according to the comprehensive risk level of the power grid.

[0091] It should be understood that Figure 2 the modules described in the specification and the reference Figure 1The individual steps in the method described above correspond. Thus, the operations and features described above for the method and the corresponding technical effects apply equally to the modules in Figure 2 in which the modules are not described again here.

[0092] In some embodiments, the present application also provides a computer readable storage medium having stored thereon a computer program, the program instructions being executed by a processor to cause the processor to perform the knowledge graph-based severe weather power grid risk identification method in any of the method embodiments described above.

[0093] As an implementation, the computer readable storage medium of the present application stores computer executable instructions, and the computer executable instructions are configured to:

[0094] cleaning and spatio-temporal calibration are performed on the multi-source weather forecast data and the power grid topology structure data in the target area to obtain weather data and power grid state data;

[0095] Based on the preset knowledge graph, the weather data and the power grid state data are fused to extract weather features, associated node relationships of power grid facilities, and risk transmission paths, to generate a physical risk association subgraph and a cascading risk association subgraph;

[0096] Based on the physical risk association subgraph, the probability simulation is used to analyze the tower collapse probability and the conductor dancing amplitude to obtain the physical damage risk of the power grid facility under severe weather conditions;

[0097] Based on the cascading risk association subgraph, the graph theory cascading failure analysis is combined to deduce the power grid cascading failure propagation range to obtain the failure cascading diffusion risk of the power grid facility under severe weather conditions;

[0098] Based on the physical damage risk and the failure cascading diffusion risk of the power grid facility under severe weather conditions, the comprehensive risk level of the power grid is determined;

[0099] According to the comprehensive risk level of the power grid, a power grid risk heat map is generated.

[0100] The computer-readable storage medium may include a program storage area and a data storage area, wherein the program storage area may store an operating system and application programs required for at least one function; the data storage area may store data created based on the use of the severe weather power grid risk identification system based on the knowledge graph, etc. In addition, the computer-readable storage medium may include a high-speed random access memory and may also include a memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state memory device. In some embodiments, the computer-readable storage medium may optionally include a memory remotely located relative to the processor, and these remote memories may be connected to the severe weather power grid risk identification system based on the knowledge graph via a network. Examples of the aforementioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0101] Figure 3 Schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. Figure 3 As shown, the device includes: a processor 310 and a memory 320. The electronic device may also include: an input device 330 and an output device 340. The processor 310, the memory 320, the input device 330 and the output device 340 may be connected via a bus or other means. Figure 3 The example of a bus connection is used. The memory 320 is the computer-readable storage medium mentioned above. The processor 310 executes various functional applications and data processing of the server by running the non-volatile software programs, instructions and modules stored in the memory 320, that is, implements the severe weather power grid risk identification method based on the knowledge graph of the above method embodiment. The input device 330 can receive input digital or character information, and generate key signal input related to user settings and function control of the severe weather power grid risk identification system based on the knowledge graph. The output device 340 may include a display device such as a display screen.

[0102] The electronic device can execute the method provided by the embodiment of the present invention, and has the functional modules and beneficial effects corresponding to the execution method. For technical details not fully described in this embodiment, please refer to the method provided by the embodiment of the present invention.

[0103] As an embodiment, the electronic device is applied to a severe weather power grid risk identification system based on a knowledge graph and is used for a client, and includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to:

[0104] Perform data cleaning and spatiotemporal calibration on multi-source weather forecast data and power grid topology data in the target area to obtain weather data and power grid status data;

[0105] The weather data and the power grid state data are fused based on a preset knowledge graph, weather features, associated node relationships of power grid facilities and risk transmission paths are extracted, a physical risk association subgraph and a cascading risk association subgraph are generated;

[0106] Based on the physical risk association subgraph, a probability simulation is adopted to analyze tower collapse probability and conductor dancing amplitude, and physical damage risk of power grid facilities under adverse weather conditions is obtained;

[0107] Based on the cascading risk association subgraph, combined with graph theory cascading failure analysis, the propagation range of power grid cascading failure is deduced, and the cascading diffusion risk of power grid facilities under adverse weather conditions is obtained.

[0108] Based on the physical damage risk and the cascading diffusion risk of the power grid facilities under adverse weather conditions, a comprehensive risk level of the power grid is determined.

[0109] A power grid risk heat map is generated according to the comprehensive risk level of the power grid.

[0110] Through the description of the above embodiments, those skilled in the art can clearly understand that the embodiments can be realized by means of software and necessary general hardware platforms, and of course, they can also be realized by hardware. Based on such understanding, the above technical solutions can be embodied in the form of a software product, which can be stored in a computer readable storage medium, such as a ROM / RAM, a magnetic disk, an optical disk, etc., and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods of the various embodiments or some parts of the embodiments.

[0111] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacements for some technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for identifying severe weather power grid risks based on knowledge graph, characterized in that: include: Perform data cleaning and spatiotemporal calibration on multi-source weather forecast data and power grid topology data in the target area to obtain weather data and power grid status data; The method of fusing the meteorological data and the power grid status data based on a preset knowledge graph, extracting weather characteristics, associated node relationships of power grid facilities, and risk transmission paths, and generating a physical risk associated subgraph and a chain risk associated subgraph includes: According to the preset knowledge graph, the meteorological data is used as a meteorological node and the power grid status data is used as a power grid facility node; The association relationship between meteorological nodes and power grid facility nodes is established through association analysis method; Identify risk transmission paths between meteorological data and power grid status data; Constructing a physical risk association subgraph and a chain risk association subgraph based on the association relationship and the risk transfer path; Based on the physical risk association subgraph, probabilistic simulation is used to analyze the probability of tower toppling and the amplitude of conductor galloping to obtain the physical damage risk of power grid facilities under severe weather conditions. The physical risk association subgraph, based on the physical risk association subgraph, probabilistic simulation is used to analyze the probability of tower toppling and the amplitude of conductor galloping to obtain the physical damage risk of power grid facilities under severe weather conditions, includes: The physical risk correlation subgraph is used as the analysis object; The probability of tower collapse and conductor galloping amplitude are selected as analysis indicators; Based on the correlation between meteorological nodes and power grid facility nodes in the physical risk correlation subgraph, a probability model of tower toppling probability and conductor galloping amplitude is established; According to the probability model of tower collapse probability and conductor galloping amplitude, the probability simulation method is used to calculate the tower collapse probability and conductor galloping amplitude; Determine the physical damage risk of power grid facilities under severe weather conditions based on the probability of tower toppling and the amplitude of conductor galloping; Based on the chain risk association subgraph, combined with the graph theory cascading failure analysis, the propagation range of the power grid cascading failure is deduced, and the fault cascading spread risk of the power grid facilities under severe weather conditions is obtained. The chain risk association subgraph, combined with the graph theory cascading failure analysis, the propagation range of the power grid cascading failure is deduced, and the fault cascading spread risk of the power grid facilities under severe weather conditions is obtained. The following are the results: The chain risk association subgraph is used as the analysis object, and the cascading failure propagation range of power grid facilities is used as the analysis indicator; Based on the risk transfer path between meteorological nodes and power grid facility nodes in the chain risk association subgraph, the graph theory cascading failure analysis method is used to identify the risk transfer path between meteorological data and power grid status data. Based on the risk transmission path, a cascading failure propagation model of power grid facilities is established; Calculate the cascading failure propagation range of power grid facilities under severe weather conditions based on the cascading failure propagation model; Determine the cascading failure risk of power grid facilities under severe weather conditions based on the propagation range of cascading failures of power grid facilities; determining a comprehensive risk level of the power grid based on the physical damage risk and the fault cascading risk of power grid facilities under severe weather conditions; A power grid risk heat map is generated based on the comprehensive risk level of the power grid.

2. The method for identifying severe weather power grid risks based on knowledge graph according to claim 1, characterized in that: The data cleaning and spatiotemporal calibration of multi-source weather forecast data and power grid topology data in the target area to obtain weather data and power grid status data includes: Acquiring multi-source weather forecast data covering the target area; Acquiring power grid topology data matching the target area; performing data cleaning processing on the multi-source weather forecast data and the power grid topology data respectively; The cleaned multi-source weather forecast data and the power grid topology data are subjected to time and space calibration processing to generate the weather data and the power grid status data.

3. The method for identifying severe weather power grid risks based on knowledge graph according to claim 1, characterized in that: Determining the comprehensive risk level of the power grid based on the physical damage risk and the fault cascading risk of power grid facilities under severe weather conditions includes: A comprehensive grid risk indicator system is constructed based on the probability of tower toppling and the amplitude of conductor galloping in the physical damage risk of grid facilities, as well as the scope of chain failure propagation in the cascading risk of faults in grid facilities. Based on the comprehensive risk analysis index system of the power grid, a comprehensive risk level analysis model of the power grid is established, and the comprehensive risk level of the power grid is determined according to the comprehensive risk level analysis model of the power grid.

4. The method for identifying severe weather power grid risks based on knowledge graph according to claim 1, characterized in that: Generating a power grid risk heat map according to the comprehensive risk level of the power grid includes: Based on the comprehensive risk level of the power grid in the target area, the target area is divided into multiple spatial units; Assign a risk level corresponding to the overall risk level to each spatial unit; Based on the risk level of spatial units, a spatial interpolation method is used to generate a power grid risk heat map.

5. A severe weather power grid risk identification system based on knowledge graph, characterized by: include: A data processing module is configured to perform data cleaning and spatiotemporal calibration on multi-source weather forecast data and power grid topology data in a target area to obtain weather data and power grid status data; The first generation module is configured to fuse the meteorological data and the power grid status data based on a preset knowledge graph, extract weather characteristics, associated node relationships of power grid facilities, and risk transmission paths, and generate a physical risk association subgraph and a chain risk association subgraph, wherein the fusing of the meteorological data and the power grid status data based on the preset knowledge graph, extracting weather characteristics, associated node relationships of power grid facilities, and risk transmission paths, and generating a physical risk association subgraph and a chain risk association subgraph includes: According to the preset knowledge graph, the meteorological data is used as a meteorological node and the power grid status data is used as a power grid facility node; The association relationship between meteorological nodes and power grid facility nodes is established through association analysis method; Identify risk transmission paths between meteorological data and power grid status data; Constructing a physical risk association subgraph and a chain risk association subgraph based on the association relationship and the risk transfer path; The first analysis module is configured to analyze the probability of tower toppling and the amplitude of conductor galloping based on the physical risk association subgraph using probabilistic simulation to obtain the physical damage risk of power grid facilities under severe weather conditions. The physical risk association subgraph is configured to analyze the probability of tower toppling and the amplitude of conductor galloping based on the physical risk association subgraph using probabilistic simulation to obtain the physical damage risk of power grid facilities under severe weather conditions including: The physical risk correlation subgraph is used as the analysis object; The probability of tower collapse and conductor galloping amplitude are selected as analysis indicators; Based on the correlation between meteorological nodes and power grid facility nodes in the physical risk correlation subgraph, a probability model of tower toppling probability and conductor galloping amplitude is established; According to the probability model of tower collapse probability and conductor galloping amplitude, the probability simulation method is used to calculate the tower collapse probability and conductor galloping amplitude; Determine the physical damage risk of power grid facilities under severe weather conditions based on the probability of tower toppling and the amplitude of conductor galloping; The second analysis module is configured to deduce the propagation range of power grid cascading failures based on the chain risk association subgraph and in combination with graph theory cascading failure analysis to obtain the cascading failure propagation risk of power grid facilities under severe weather conditions, wherein the deduction of the propagation range of power grid cascading failures based on the chain risk association subgraph and in combination with graph theory cascading failure analysis to obtain the cascading failure propagation risk of power grid facilities under severe weather conditions includes: The chain risk association subgraph is used as the analysis object, and the cascading failure propagation range of power grid facilities is used as the analysis indicator; Based on the risk transfer path between meteorological nodes and power grid facility nodes in the chain risk association subgraph, the graph theory cascading failure analysis method is used to identify the risk transfer path between meteorological data and power grid status data. Based on the risk transmission path, a cascading failure propagation model of power grid facilities is established; Calculate the cascading failure propagation range of power grid facilities under severe weather conditions based on the cascading failure propagation model; Determine the cascading failure risk of power grid facilities under severe weather conditions based on the propagation range of cascading failures of power grid facilities; a determination module configured to determine a comprehensive risk level of the power grid based on the physical damage risk and the fault cascading risk of power grid facilities under severe weather conditions; The second generating module is configured to generate a power grid risk heat map according to the comprehensive risk level of the power grid.

6. An electronic device, characterized in that: include: At least one processor, and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 4.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 4 is implemented.

Images