An accident research and dynamic disposal method and system

Abstract

The application provides an accident research and dynamic disposal method and system, the method comprises the following steps: acquiring real-time power grid topology data, dynamically gathering the basic model of each circuit system device, and dynamically updating the whole network topology atlas; identifying the abnormal working condition of the power grid; constructing an accident plan knowledge base; automatically matching the current scene according to the multi-dimensional operation data of the power grid, and generating an accident disposal plan; screening and feature matching of real-time alarm information based on a fault feature identification model, and aggregating to obtain a complete fault event; calling the whole network topology atlas, the operation data of each circuit system device and the historical disposal plan for cross analysis, and generating an auxiliary disposal strategy; and quantitatively counting and evolution analysis based on the historical record and real-time account of the complete fault event. The application realizes intelligent research and judgment, automatic generation of a plan and dynamic disposal decision fusion of a power grid accident, and improves the efficiency and intelligent level of power grid fault disposal.

Description

Technical Field

[0001] This invention belongs to the technical field of power system automation, specifically relating to a method and system for accident assessment and dynamic handling. Background Technology

[0002] The construction and development of power grids are becoming increasingly rapid and complex, with the grid scale continuously expanding and the grid structure becoming increasingly intricate. The safe and stable operation of power grids faces ever-growing challenges. Traditional power grid monitoring and management methods are no longer sufficient to meet the demands of efficient and intelligent management in modern power grids.

[0003] Currently, the development of power grid emergency response plans relies primarily on manual experience, which has the following shortcomings: First, while the D5000 system provides real-time monitoring and fault diagnosis of power grid over-limits, it lacks detailed analysis capabilities for over-limit information. Fault diagnosis can only provide relevant alarm information, lacking handling strategies and post-event analysis capabilities. Second, current emergency response plan development depends on the manual experience and subjective judgment of dispatchers, resulting in inconsistent plan quality, low development efficiency, and difficulty in coping with increasingly complex power grid operating conditions. Third, when a power grid fault occurs, dispatchers need to sift through massive amounts of alarm information to identify valid signals, determine the fault type and scope, and formulate handling plans within a short period of time. This places a heavy workload and significant decision-making pressure on them, making them prone to misjudgments and omissions.

[0004] Existing technologies lack the ability to systematically identify risks at weak points in power grid operation, and cannot automatically identify inherent risks and operational risks of the grid structure such as single power supply of main transformers, single section of busbars, and single power supply of lines; they lack the ability to automatically verify abnormal operating modes, and cannot promptly detect deviations between equipment operating modes and standard modes; they lack the ability to deeply mine and analyze trends in historical fault data, making it difficult to provide data support for power grid operation and maintenance planning and risk prevention and control. Summary of the Invention

[0005] To address the aforementioned technical problems, this invention provides a method and system for accident assessment and dynamic handling, thereby overcoming the shortcomings of the prior art.

[0006] In a first aspect, the present invention provides a method for accident assessment and dynamic handling, the method comprising: Real-time power grid topology data is acquired, and basic models of various circuit system devices are dynamically aggregated to construct and dynamically update the entire network topology map based on the basic models and the real-time power grid topology data. Based on the network topology map and the actual operating conditions of the power grid, the operating status of the power grid is perceived to identify abnormal operating conditions of the power grid. Construct an accident emergency response plan knowledge base, and manage and categorize the knowledge base in a unified manner; The system automatically matches the current scenario based on multi-dimensional power grid operation data, and generates an accident response plan based on the contingency plan knowledge base and the integrated network topology map. The system acquires real-time alarm information from the power grid and establishes a fault feature identification model for the equipment. Based on the fault feature identification model, the system filters and matches the real-time alarm information to identify fault information. The system then aggregates the fault information and the real-time alarm information to confirm the complete fault event. The entire network topology map, the operating data of each circuit system device, and historical contingency plans are retrieved and cross-analyzed to obtain the scope of the fault impact and its changing trend, and to generate auxiliary handling strategies. Quantitative statistics and evolution analysis are performed based on the complete historical records and real-time ledgers of the fault events to output statistical results.

[0007] Compared with existing technologies, the beneficial effects of this invention are as follows: It dynamically aggregates the basic model of primary equipment in the entire power grid, completing real-time reconstruction and dynamic updating of the entire power grid topology map. Combined with the actual operating conditions of the power grid, it accurately captures four core abnormal scenarios: fault tripping, equipment overload, abnormal operating modes, and daily operating risks, achieving real-time perception of the entire power grid's operating status. Through mechanisms such as unified fault analysis retrieval time windows, partitioned permission verification, equipment type routing, deduplication sorting, and batch storage, it achieves automated screening, analysis, and storage of fault alarm data, significantly improving fault handling efficiency. Through centralized storage, version management, and execution tracking of the contingency plan knowledge base, combined with maintenance operation mode extraction, islanding analysis, power outage range determination, and branch logic of handling steps, it automatically generates standardized accident contingency plans. Through equipment fault feature identification models, automatic interference signal filtering, and automatic fault event aggregation, it shortens fault identification and judgment time and improves the accuracy and timeliness of judgment in complex fault scenarios. By automatically generating structured fault handling strategies, it provides reliable decision-making basis for dispatching personnel. Through aggregation and trend evolution analysis, it provides data reference for power grid operation and maintenance planning and early risk control.

[0008] Furthermore, after the step of acquiring real-time power grid topology data, the method further includes: A power grid topology feature intelligent recognition model is constructed based on a graph neural network, and the power grid topology feature intelligent recognition model is trained. Based on the trained intelligent identification model of power grid topology features, target detection is performed on the real-time power grid topology data to identify weak nodes, risk points, and equipment operation-related logic.

[0009] Furthermore, the step of sensing the power grid's operating status based on the entire network topology map and the actual operating conditions of the power grid to identify abnormal operating conditions of the power grid includes: The system captures several abnormal scenarios of the power grid based on the actual operating conditions of the power grid, and performs real-time perception of the power grid's operating status based on the entire network topology map. The abnormal scenarios include capturing accident tripping, equipment overload, abnormal operating mode, and daily operating risks.

[0010] Furthermore, the step of capturing accident tripping includes: Set a time window for fault querying and filter the set of valid alarms; Perform partition permission verification and deduplication on the alarm set data, and analyze, aggregate and sort the alarm set; The expression for filtering the set of valid alarms is: ; In the formula, This indicates the selection of the set of valid alarms. , These represent the start time and end time of the data retrieval, respectively. , These represent the set of fault keywords and the set of interference elimination keywords, respectively. Indicates the responsibility partition code, Indicates the EMS system identifier. Indicates the alarm type. This indicates a change in remote signaling. Represents the sequence of events. This indicates the query function for alarms with the identifier 3000. This represents the alarm query function for a data source with an identifier of 5000 and an alarm type of remote signaling change. This represents an alarm query function for a data source with an identifier of 5000 and an alarm type of event sequence record.

[0011] Furthermore, the step of automatically matching the current scenario based on multi-dimensional power grid operation data and generating an accident handling plan based on the contingency plan knowledge base and the integrated network topology map includes: It integrates multi-dimensional data such as power grid topology, real-time equipment operating parameters, regional load distribution patterns, and dispatch safety control criteria, and links them with the planned operation risk warning list, real-time weak points in power grid operation, and typical fault handling cases from previous years to automatically match the current power grid operation scenario. Based on the aforementioned contingency plan knowledge base and by integrating the entire network topology map to analyze accident-inducing factors and deduce accident evolution paths, an accident handling contingency plan adapted to the on-site working conditions is generated.

[0012] Furthermore, the step of filtering and feature matching the real-time alarm information based on the fault feature recognition model to identify fault information, and aggregating the fault information and the real-time alarm information to confirm the complete fault event includes: Based on the fault feature recognition model, the real-time alarm information is automatically identified and valid fault alarm data is filtered. Based on the valid fault alarm data and combined with the feature library, the fault type, fault location and fault cause are judged to obtain fragmented alarm information, and the fragmented alarm information is aggregated to obtain a complete fault event.

[0013] Furthermore, the step of performing quantitative statistical and evolution analysis based on the historical records and real-time ledgers of the complete fault events to output statistical results includes: Obtain full historical data on power grid faults and real-time fault ledgers; Based on fault type, fault event, equipment voltage level and type, the full historical data and the real-time fault ledger are quantitatively summarized, trend evolution is analyzed and patterns are mined, and special statistical results are output.

[0014] Secondly, the present invention also provides an accident assessment and dynamic response system, the system comprising: The acquisition and aggregation module is used to acquire real-time power grid topology data and dynamically aggregate the basic models of various circuit system devices, so as to construct and dynamically update the entire network topology map based on the basic models and the real-time power grid topology data. The sensing and identification module is used to sense the operating status of the power grid based on the entire network topology map and the actual operating conditions of the power grid, so as to identify abnormal operating conditions of the power grid. A management module is built to construct an accident contingency plan knowledge base and to uniformly manage and classify the contingency plan knowledge base. The matching and fusion module is used to automatically match the current scenario based on multi-dimensional power grid operation data, and generate an accident handling plan based on the contingency plan knowledge base and the fusion of the whole network topology map; A filtering module is established to obtain real-time alarm information from the power grid and to establish a fault feature identification model for the equipment. Based on the fault feature identification model, the real-time alarm information is filtered and feature-matched to identify fault information. The fault information and the real-time alarm information are then aggregated to confirm the complete fault event. The scheduling and analysis module is used to retrieve the entire network topology map, the operating data of each circuit system device, and historical handling plans for cross-analysis to obtain the scope of fault impact and its changing trend, and generate auxiliary handling strategies. The statistical analysis module is used to perform quantitative statistics and evolution analysis based on the historical records and real-time ledgers of the complete fault events, so as to output statistical results.

[0015] Thirdly, the present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the above-mentioned accident assessment and dynamic handling method.

[0016] Fourthly, the present invention also provides a storage medium storing a computer program thereon, which, when executed by a processor, implements the above-described method for accident assessment and dynamic handling. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart of the accident assessment and dynamic handling method in the first embodiment of the present invention; Figure 2 This is a structural block diagram of the accident assessment and dynamic response system in the second embodiment of the present invention; Figure 3 This is a schematic diagram of the structure of the electronic device in the third embodiment of the present invention.

[0019] Explanation of key component symbols: 10. Acquisition and Aggregation Module; 20. Perception and Recognition Module; 30. Construction and Management Module; 40. Matching and Fusion Module; 50. Establishment and Filtering Module; 60. Scheduling and Analysis Module; 70. Statistical Analysis Module; 80. Bus; 81. Processor; 82. Memory; 83. Communication interface.

[0020] The embodiments of the present invention will be further described below with reference to the accompanying drawings. Detailed Implementation

[0021] To facilitate understanding of the present invention, a more complete description will be given below with reference to the accompanying drawings. Several embodiments of the invention are illustrated in the drawings. However, the invention can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete.

[0022] It should be noted that when a component is said to be "fixed to" another component, it can be directly on the other component or there may be an intervening component. When a component is said to be "connected to" another component, it can be directly connected to the other component or there may be an intervening component. The terms "vertical," "horizontal," "left," "right," and similar expressions used in this document are for illustrative purposes only.

[0023] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0024] Example 1 Please see Figure 1 The figure shows the accident assessment and dynamic handling method in the first embodiment of the present invention, the method including steps S1 to S7: S1. Obtain real-time power grid topology data and dynamically aggregate the basic models of each circuit system device to construct and dynamically update the entire network topology map based on the basic models and the real-time power grid topology data. It should be noted that power grid topology data is the fundamental data describing the connection relationships between various electrical devices in the power grid (such as substations, main transformers, transmission lines, switches, busbars, etc.). By accessing the real-time power grid topology data source of the D5000 system, the basic model of primary equipment of the entire power grid is dynamically aggregated, completing the real-time reconstruction and dynamic updating of the entire power grid topology map. The topology map is presented in a graph data structure, where nodes represent power grid equipment and edges represent the electrical connections between equipment. As the power grid operation mode changes (such as switching operations, equipment commissioning and decommissioning, etc.), the topology map is updated synchronously, providing basic data support for subsequent anomaly identification, fault assessment, and contingency plan generation.

[0025] Specifically, step S1 includes steps S11 to S12: S11, Construct a power grid topology feature intelligent recognition model based on graph neural network, and train the power grid topology feature intelligent recognition model; It should be noted that Graph Neural Networks (GNNs) are deep learning models specifically designed for processing graph-structured data, effectively capturing the dependencies and global features between nodes in a power grid topology. Based on graph neural network architectures such as GraphSAGE, an intelligent recognition model for power grid topology features is constructed. First, a training dataset containing normal and abnormal power grid maps is built, divided into training, validation, and test sets according to a certain ratio. The power grid topology data is standardized and uniformly converted into a graph data structure, and manually labeled using frameworks such as PyTorch Geometric, categorizing it into "weak nodes," "risk locations," "operating modes," and "topological relationships." Using a graph neural network backbone pre-trained on a large-scale power system dataset, a transfer learning strategy is employed to train the model, and the model performance is monitored using the validation set.

[0026] S12, Based on the trained intelligent identification model of power grid topology features, target detection is performed on the real-time power grid topology data to identify weak nodes, risk points and equipment operation-related logic of the power grid. It should be noted that during actual monitoring tasks, the real-time power grid topology data pushed by the D5000 system is input into the trained intelligent power grid topology feature recognition model. The model automatically performs target detection on the topology data and outputs accurate identification results of weak nodes, risk points, and equipment operation correlation logic in the power grid. For example, the model can identify transformers operating under heavy load for extended periods, critical buses in single-source power supply mode, and areas with abnormal topology connections. Tests have shown that the accuracy rate for weak node identification can reach over 92%, and the accuracy rate for risk point identification can reach over 95%, achieving automation and high efficiency in power grid situational awareness.

[0027] S2, based on the entire network topology map and the actual operating conditions of the power grid, perceive the operating status of the power grid in order to identify abnormal operating conditions of the power grid; It should be noted that, based on the completion of the entire network topology map, further integration with the actual operating conditions of the power grid (including real-time telemetry data, remote signaling status, power flow distribution, etc.) and relying on the relationships between devices in the topology map enables real-time, comprehensive perception of the power grid's operating status. Its core objective is to automatically identify abnormal operating conditions deviating from normal operation from massive amounts of power grid operating data, including equipment failures, abnormal operating modes, and potential risks, providing accurate situational input for subsequent fault assessment and handling decisions.

[0028] Specifically, step S2 includes step S21: S21, Combine the actual operating conditions of the power grid to capture several abnormal scenarios of the power grid, and perform real-time perception of the power grid operation status in the whole domain according to the full network topology map. Among them, the several abnormal scenarios include capturing accident tripping, equipment overload, abnormal operation mode and daily operation risk. It should be noted that power grid anomaly scenarios are diverse and complex. The system focuses on capturing four core anomaly scenarios: accidental tripping (equipment tripping due to fault protection), equipment overload (equipment operating load exceeding rated limits), abnormal operating conditions (equipment's actual operating state differs from standard operating conditions), and daily operational risks (weak links and potential risks in the power grid identified based on the current operating mode). Leveraging the correlations within the topology map, the system can quickly locate the equipment scope, affected area, and severity of anomalies, achieving real-time, comprehensive, and accurate perception of the power grid's operational status.

[0029] Specifically, step S21 includes steps S211 to S212: S211, Set the fault query time window and filter the valid alarm set; It should be noted that the D5000 power grid system generates a massive amount of alarm information, including many invalid, repetitive, or interference signals. First, a unified fault query time window is set, and missing time parameters are automatically filled in. Specifically, the fault query start time is set. Fault query end time ,in The current system time is used. Based on this, the valid alarm set is filtered using an alarm data query function, and invalid alarms outside the time window are removed.

[0030] S212, perform partition permission verification and deduplication on the alarm set data, and analyze, aggregate and sort the alarm set; It should be noted that after filtering out valid alarms, multi-level refined processing is required. First, alarm data undergoes partitioned access verification. The power grid is divided into different responsibility zones by region; alarms across zones may be false alarms or fall outside the jurisdiction of the zone and need to be removed. Second, deduplication is performed using the equipment interval ID as a unique identifier to avoid duplicate alarms from the same equipment and the same fault, leading to duplicate analysis. Then, alarms are automatically distributed to the corresponding specialized fault analysis service for processing based on equipment type (line, transformer, bus, capacitor, automatic transfer switch, etc.). Finally, the processed fault results are sorted according to priority, fault time, and millisecond-level time sequence, and after completing the equipment ledger information, they are batch-entered into the database to provide standardized data for subsequent fault statistics and trend analysis.

[0031] The expression for filtering the set of valid alarms is: ; In the formula, This indicates the selection of the set of valid alarms. , These represent the start time and end time of the data retrieval, respectively. , These represent the set of fault keywords and the set of interference elimination keywords, respectively. Indicates the responsibility area code. Indicates the EMS system identifier. Indicates the alarm type. This indicates a change in remote signaling. Represents the sequence of events. This indicates the query function for alarms with the identifier 3000. This represents the alarm query function for a data source with an identifier of 5000 and an alarm type of remote signaling change. This represents an alarm query function for a data source with an identifier of 5000 and an alarm type of event sequence record.

[0032] It's worth noting that after filtering out valid alarms, further refined processing is required. First, the alarm data undergoes partition permission verification—the power grid is divided into different responsibility zones by region. Alarms crossing zones may be false alarms or belong to areas outside the jurisdiction of the assigned zone, and therefore need to be removed. The partition permission verification and deduplication verification expressions are as follows: ; ; In the formula, Indicates a single alarm. For area coding, The partition is for the current task to run. For factory / station numbering. The partition to which the alarm belongs; This is a function for validating the validity of partitions. For a single alarm, For effective alarm collection; For device interval ID, This is a cache set of analyzed fault intervals; Then, based on the equipment type (line, transformer, busbar, capacitor, automatic transfer switch, grounding transformer, load transfer, bus tie / bus branch, etc.), the alarm is automatically distributed to the corresponding specialized fault analysis service for processing. The equipment type routing expression is: ; In the formula, This indicates the analysis result of a single alarm. Provides specialized analysis services for the corresponding equipment. Indicates the type of equipment failure. This indicates the fault analysis service method. Indicates method parameters; The aggregate expression for the fault results is: ; In the formula, This represents the final list of deduplicated faulty devices. This indicates the total number of fault results. Indicates the index of the fault result. Indicates the first The results of the alarm analysis are as follows: Represents a faulty device value object; The expression for normalizing the time field is: ; In the formula, This field represents the millisecond time of the fault. Indicates the alarm type of the data acquisition and monitoring system. Indicates the original warning type; The pre-entry validation expression for filtering manual operations and invalid data from repeated historical faults is: ; This represents the function for determining whether a faulty entity is added to the database. This indicates the number of historical fault records. This indicates the number of manual remote control operation records; The expression for sorting valid faults according to priority, time, and millisecond-level timing is as follows: ; This represents a comparison function for sorting faulty entities. , These represent two different fault entities to be sorted. , These represent two different failure times. , These represent two different fault times in milliseconds.

[0033] The expression for automatically completing equipment ledger information based on equipment type is: ; For equipment subtypes, Voltage level coding, This is a function for querying device files; Extract remote signaling data from the time window before and after the fault, and set the following: ; In the formula, , These represent the start time and end time of remote information data retrieval, respectively. Indicates the time when the fault occurred; The expression for batch inbound data entry is: ; This indicates the end of the current batch of data entry index. Quantity of a single batch of goods received. This represents the total number of associated remote communication messages.

[0034] The expression for automatically piecing together fault analysis results and generating a message is: ; This indicates the concatenated fault analysis message. For formatting the fault time in Chinese characters, This is the description of the fault. This is the conclusion of the fault assessment; The expression for controlling message push based on debug mode and analysis type is: ; This represents the message push control function. Indicates a system debugging flag. Indicates the analysis type. This indicates that debug mode is enabled; The expression for counting the number of valid inbound faults and returning the result is: ; This indicates the total number of valid inbound failures returned. This represents the total number of faults successfully added to the inventory.

[0035] S3, Construct an accident contingency plan knowledge base, and manage and categorize the contingency plan knowledge base in a unified manner; It should be noted that emergency response plans are guiding documents for dispatchers to perform handling operations when power grid faults occur. A standardized, hierarchical, and traceable power grid fault response plan knowledge base system should be established to comprehensively cover various power grid accidents, equipment failures, and extreme operating conditions. The knowledge base possesses four core capabilities: centralized storage and archiving of plans, multi-version iteration management, full-process tracking of plan execution, and quantitative assessment of fault risks. Plans are categorized and organized according to fault type (e.g., line fault, main transformer fault, bus fault), voltage level (e.g., 220kV, 110kV), equipment category, and scope of accident impact. It supports full lifecycle management operations such as adding new plans, revising and invalidating historical plans, and archiving and sealing old plans, completely retaining the entire process record of plans from preparation, review, issuance to execution. It is worth noting that the expression for determining the state of the contingency plan generation is: ; In the formula, Indicates the status of the contingency plan generation; The standard title generation expression for the contingency plan is: ; In the formula, For the maintenance of the substation, For voltage level, For equipment maintenance. For equipment type, Indicates the standard title of the contingency plan; The expression for extracting the text set of maintenance and operation modes is: ; In the formula, A collection of text representing maintenance and operation modes. This indicates the function for extracting maintenance and operation modes. Indicates the cross-sectional time; The expression for analyzing islanded topologies across the entire network is: ; In the formula, This indicates the results of the island analysis. Indicates the equipment to be expected to malfunction. This function represents the analysis function for islanding in the entire network topology. The expression for determining the power loss range is: ; ; In the formula, , These represent the results of the power outage area determination and the list of power stations that have lost power, respectively. This represents the function for determining the power loss range. This indicates that the substation is under maintenance. Indicates a collection of busbar equipment. Indicates other related plants and stations, This indicates that the condition is true; The expression for determining special scenarios for T-connections is: ; In the formula, express Special wiring scenario markings; The expression for determining the third power source is: ; In the formula, This indicates the result of the third power supply determination. This indicates the existence of a third power source. This indicates that there is no third power source. The statistical expression for load loss is: ; In the formula, Indicates the total power loss load. Indicates the active power of the line. Indicates a collection of devices that have lost power. Indicates the cross-sectional time; The branch logic expression for the handling steps is: ; In the formula, Indicates the final handling steps, express Handling procedures for wiring scenarios, This indicates the handling steps for a third power supply scenario. This indicates the steps for handling typical scenarios; The expression for determining the status of the automatic switching action is: ; In the formula, Indicates the status of the automatic transfer action. Indicates the status of the automatic transfer switch (ATS) being in or out of service; The expression for determining the separation of small hydropower stations and important users is: ; In the formula, Represents the set of decoupling devices. This represents a collection of small hydroelectric equipment. This represents a set of important user equipment. This indicates the result of the power outage area determination.

[0036] S4. Automatically match the current scenario based on multi-dimensional power grid operation data, and generate an accident handling plan based on the contingency plan knowledge base and the integrated network topology map; It should be noted that traditional contingency plan development relies heavily on human experience, resulting in low efficiency and inconsistent quality. This new system abandons the traditional manual development model and achieves intelligent plan generation. It integrates multi-dimensional core data, including power grid topology, real-time equipment operating parameters, regional load distribution patterns, and dispatch safety control principles. It also links three reference bases: a list of planned operational risks, real-time weak points in power grid operation, and typical fault handling cases from previous years. The system automatically matches the current power grid operating scenario, intelligently analyzes accident-inducing factors and their spread, rapidly simulates various accident evolution paths, and adaptively generates customized accident handling plans tailored to the specific on-site conditions.

[0037] Specifically, step S4 includes steps S41 to S42: S41 integrates multi-dimensional data such as power grid topology, real-time equipment operating parameters, regional load distribution patterns, and dispatch safety control criteria, and links with the planned operation risk warning list, real-time weak points in power grid operation, and typical fault handling cases from previous years to automatically match the current power grid operation scenario. It should be noted that the data fusion layer for intelligent contingency plan generation involves several steps. First, the system obtains real-time power grid topology and equipment operating parameters (including measurements of voltage, current, active power, and reactive power) from the D5000 system. Simultaneously, it loads multi-dimensional data such as regional load distribution patterns (e.g., load characteristics at different times) and dispatch safety control criteria (e.g., equipment limits, N-1 criteria). Based on this, it further integrates three types of reference data: a planned operation risk warning list (reflecting potential risks from future planned operations), real-time weak points in power grid operation (reflecting current weaknesses in power grid operation), and historical typical fault handling cases (reflecting successful handling experience of similar faults in the past). Through comprehensive matching of multi-dimensional data, the system automatically identifies the type and characteristics of the current power grid operating scenario.

[0038] S42, Based on the contingency plan knowledge base and by integrating the network topology map to analyze accident-inducing factors and deduce accident evolution paths, an accident handling contingency plan adapted to the on-site working conditions is generated. It should be noted that after completing scenario matching, the system enters the core stage of contingency plan simulation and generation. Based on standardized contingency plan templates and historical cases stored in the contingency plan knowledge base, and combined with the relationships between devices in the network topology map, the system analyzes the possible triggering factors (such as equipment overload, protection malfunction, external damage, etc.) and the scope of spread (such as power outage areas, affected equipment, etc.) of the accident. Subsequently, the system simulates multiple possible accident evolution paths, assesses the impact and handling difficulty of each path, and finally selects the optimal handling path and generates an accident handling plan adapted to the current operating conditions. Throughout the process, the system automatically verifies the compliance of the contingency plan clauses, the completeness of the handling procedures, and the comprehensiveness of the response measures, corrects unreasonable handling links, and optimizes core handling steps such as load transfer, equipment isolation, and method adjustment.

[0039] It is worth noting that, in this embodiment, the pre-validation expression for the fault deduction process is: ; In the formula, This indicates the result of the preliminary validity check. Indicates the operating status of the plant. Indicates the type of plant / station; The first-level scene branch decision expression is: ; In the formula, This indicates the result of the first-level scene branch determination. This indicates that the main transformer is completely stopped. This indicates that the busbar is completely shut down. Indicates logical false; The expression for the second-level single-device fault subdivision judgment is: ; In the formula, This indicates the result of the secondary single-device fault subdivision judgment. Indicates the type of faulty device. Indicates the voltage level of the faulty equipment; The formula for calculating transferable load in a complete shutdown scenario is: ; In the formula, Indicates transferable load. Indicates the total load of the plant. This indicates the fixed load loss during a fault scenario; The formula for calculating load transfer utilization rate is: ; In the formula, Indicates the load transfer utilization rate. Indicates the equipment load limit; The expression for determining the reuse of historical data in a single-device failure scenario is: ; In the formula, This indicates the result of the historical data reuse determination. This represents a historical inference dataset. This indicates the creator field. Indicates the current user ID. The faulty device ID field indicates the faulty device ID. Indicates the ID of the currently faulty device; The expression for real-time load refresh of a cross section is: ; In the formula, This indicates that the cross-section load is being updated in real time. Indicates historical projection load, This indicates the amount of load fluctuation correction for a given period. Indicates the current simulation time. Indicates the time of the historical deduction section; The standby primary transformer filtering expression is: ; In the formula, This indicates a standby main transformer. Indicates the set of alternative main transformer equipment. Indicates the faulty plant / station number. Indicates the voltage level of the faulty main transformer. Indicates the operating status of the equipment. Indicates the candidate main transformer device ID. Indicates the faulty device ID; The priority matching expression for backup lines is: ; In the formula, Indicates the priority of the backup line. Indicates a set of alternative routes. Indicates the depot / station on the opposite side of the line. This indicates the automatic activation linkage indicator. Indicates the line's operating status. Indicates logical truth; The formula for calculating the load rate of the faulty main transformer is: ; In the formula, Indicates the load rate of the faulty main transformer. Indicates the real-time load of the faulty equipment. Indicates the rated capacity of the main transformer; The formula for calculating the shunt current of the backup line is: ; In the formula, This indicates the shunt current of the backup line. Indicates the fault current of the faulty line. Indicates the impedance of the faulty line. Indicates the impedance of the spare line; The expression for verifying load distribution security constraints is: ; In the formula, Indicates the first Transferring load along a transfer path Indicates the maximum load limit of standby equipment; The formula for calculating the total transferred load is: ; In the formula, Indicates the total transferred load. Indicates the number of valid transfer paths. Indicates the first Transferring loads via a transfer route; The formula for calculating the success rate of the transfer is: ; In the formula, Indicates the success rate of the transfer. This indicates the total power loss load of the faulty equipment.

[0040] S5, obtain real-time alarm information from the power grid and establish a fault feature identification model for the equipment, so as to filter and match the real-time alarm information based on the fault feature identification model to identify fault information, and aggregate the fault information and the real-time alarm information to confirm the complete fault event. It should be noted that when a power grid fault occurs, the D5000 system will send a large number of alarm signals in a short period of time, including remote signaling changes, protection actions, over-limit alarms, and equipment changes. These signals are numerous and intertwined, making it difficult for humans to quickly identify valid information. This step establishes an equipment fault feature recognition model to achieve automated screening of alarm information and intelligent aggregation of fault events. First, a deep analysis of the fault characteristics of the main equipment in the substation (including transformers, lines, busbars, capacitors, fault disconnection, and automatic transfer switches) is conducted, identifying the protection activation logic, switch action behavior, and alarm message characteristics corresponding to different equipment faults, forming a standardized fault judgment rule base. Based on this, an equipment fault feature recognition model is constructed for real-time processing of incoming alarm streams.

[0041] Specifically, step S5 includes steps S51 to S52: S51, Based on the fault feature recognition model, the real-time alarm information is automatically identified and valid fault alarm data is filtered. It should be noted that real-time alarm information from the power grid contains a large number of interference signals, including false alarms caused by equipment signal jitter, repeated pushes of the same event, and isolated signals without electrical correlation. This step relies on equipment action thresholds, timing logic relationships, and topological correlations to automatically identify and filter the raw alarm stream. For example, signals of repeated position changes by the same equipment within a very short period of time are identified as signal jitter and discarded; alarm combinations that do not conform to protection action logic are identified as invalid signals and filtered. Through multi-level signal filtering, the system extracts effective fault alarm data that truly reflects power grid faults from massive amounts of raw alarms, significantly reducing the data processing burden for subsequent analysis.

[0042] S52, based on the effective fault alarm data and combined with the feature library, the fault type, fault location and fault cause are judged to obtain fragmented alarm information, and the fragmented alarm information is aggregated to obtain a complete fault event. It should be noted that the filtered valid fault alarm data is still fragmented; a single alarm may only reflect one aspect of the fault (such as a switch change), while a complete fault often involves the actions of multiple devices and protection devices. This step compares the filtered valid signals item by item with the built-in equipment fault feature database, matching protection action types, equipment tripping points, and power flow change amplitudes to pinpoint the specific equipment and scope involved in the fault. Based on this, following the principle of "same fault sequence, same fault site, same fault equipment," the scattered fragmented alarm information is integrated into a complete fault event, and the fault category is classified (single equipment fault, cascading fault, power outage fault, etc.). The fault level is determined by combining the number of power-loss equipment and the load scale, and a preliminary judgment conclusion is output.

[0043] S6. Retrieve the entire network topology map, the operating data of each circuit system device, and the historical handling plans for cross-analysis to obtain the scope of the fault impact and the changing trend, and generate auxiliary handling strategies. It should be noted that after a fault event is confirmed, dispatchers need to formulate a response plan within a very short time, resulting in a heavy workload and significant decision-making pressure. This step provides intelligent decision-making assistance to dispatchers. The system retrieves the entire network topology map (to understand the connection relationship between the faulty device and other devices), real-time operating data of each device (to understand the power flow changes before and after the fault), and contingency plans for similar historical faults (to understand effective response experience), performing multi-dimensional cross-analysis. Through analysis, the system determines the actual impact range of the fault (which devices are out of power, which users are affected), the trend of power flow changes (load transfer direction, overload risk devices, etc.), and comprehensively assesses the severity level and spread risk of the incident. Based on this, relying on the built-in dispatching and response rule library, the system intelligently breaks down the response steps such as fault isolation, load transfer, mode restoration, and safety control, automatically generating a clear and practical auxiliary response strategy.

[0044] S7. Based on the historical records and real-time ledgers of the complete fault events, perform quantitative statistics and evolution analysis to output statistical results; It should be noted that the accumulation of power grid fault data is a crucial resource for operation and maintenance decision-making and risk prevention and control. This step focuses on building a multi-dimensional fault statistical analysis system based on the full historical data and real-time fault ledger of power grid faults, providing data references for power grid operation and maintenance planning and early risk prevention and control.

[0045] Specifically, step S7 includes steps S71 to S72: S71, obtain full historical data of power grid faults and real-time fault ledger; It should be noted that the system extracts all historical fault records from the database (including fault time, faulty equipment, fault type, lost load, handling process, etc.) as well as the current real-time fault log (including faults being processed and faults confirmed on the same day). This data constitutes the basic data source for multi-dimensional statistical analysis.

[0046] S72, based on fault type, fault event, equipment voltage level and type, performs quantitative summary, trend evolution analysis and pattern mining on the full historical data and the real-time fault ledger, and outputs special statistical results; It should be noted that this step analyzes fault data from five core dimensions: fault severity level, fault trigger type, fault occurrence time, equipment voltage level, and fault equipment category, quantifying and summarizing the data and assessing its trend evolution. The system supports user-defined statistical periods and filtering of business scopes, automatically generating specialized statistical results such as fault frequency trends, distribution of high-incidence fault locations, lists of vulnerable equipment, and peak fault values ​​for different time periods. Furthermore, the system aligns with the daily operational needs of dispatching, providing customized reports and data chart visualization capabilities. This intuitively presents the fault occurrence patterns and operational weaknesses of the regional power grid, providing scientific data support for power grid maintenance planning, equipment upgrade prioritization, and proactive risk control.

[0047] It is worth noting that the fault data is quantitatively summarized and trend analysis is performed from five core dimensions: fault severity level, fault trigger type, fault occurrence time, equipment voltage level, and fault equipment category. The formula for calculating the anomaly percentage in each dimension is as follows: ; In the formula, This indicates the percentage of anomalies in the corresponding dimension. This indicates the number of anomalies in the corresponding dimension. This indicates the total number of faulty devices.

[0048] In summary, the accident assessment and dynamic handling method in the above embodiments of the present invention, through deep integration with the real-time topology data source of the D5000 system, dynamically aggregates the basic model of primary equipment of the entire power grid, completes the real-time reconstruction and dynamic update of the entire power grid topology map, and accurately captures four core abnormal scenarios—accident tripping, equipment overload, abnormal operating conditions, and daily operating risks—in combination with the actual operating conditions of the power grid, thereby achieving real-time perception of the entire power grid operating status. Through mechanisms such as unified fault analysis and retrieval time window, partitioned permission verification, equipment type routing, deduplication sorting, and batch data entry, the method achieves high accuracy in fault alarm data. The automated screening, analysis, and storage of data significantly improves fault handling efficiency; through quantitative models for single-source risk assessment of main transformers, single-section risk assessment of busbars, single-source risk assessment of lines, and risk assessment of power outages for critical users, it automatically identifies inherent grid risks and temporary operational risks, outputting standardized risk levels and descriptions to provide accurate data support for grid risk prevention and control; through comparison of equipment operating status differences, verification of abnormal bus grounding status, and mapping of equipment hierarchical relationships, it automatically detects deviations between equipment operating modes and standard modes, providing timely warnings of potential abnormal situations; through centralized storage of the contingency plan knowledge base, Version management and execution tracking, combined with maintenance operation mode extraction, islanding analysis, power outage range determination, and handling step branch logic, automatically generate standardized accident contingency plans; through a historical data reuse mechanism based on both user and equipment dimensions, coupled with section refresh formulas, significantly reduce the computational power consumption of repeated simulations; compatible with multiple mainstream power grid fault scenarios, including full-station outages and single-equipment failures, and adaptable to different voltage levels; through differentiated quantitative matching formulas for main transformers and line backup equipment, combined with load rate calculation, shunt current calculation, and load allocation safety constraint verification, ensure the safety and feasibility of power transfer schemes; through equipment failure... Feature recognition models, automatic interference signal filtering, and automatic fault event aggregation shorten fault identification and judgment time and improve the accuracy and timeliness of judgment in complex fault scenarios. By cross-analyzing the impact range of accidents and power flow change trends from multiple dimensions, combined with the built-in dispatch and handling rule library, structured fault handling strategies are automatically generated within 15 seconds, providing reliable decision-making basis for dispatching personnel. By quantitatively summarizing and analyzing the trend evolution of fault data from five core dimensions—fault level, fault type, fault time, equipment voltage level, and equipment type—data references are provided for power grid operation and maintenance planning and early risk control.

[0049] Example 2 Please see Figure 2 The figure shows an accident assessment and dynamic handling system according to a second embodiment of the present invention. The system includes: The acquisition and aggregation module 10 is used to acquire real-time power grid topology data and dynamically aggregate the basic models of each circuit system device to construct and dynamically update the entire network topology map based on the basic models and the real-time power grid topology data. The sensing and identification module 20 is used to sense the operating status of the power grid based on the entire network topology map and the actual operating conditions of the power grid, so as to identify abnormal operating conditions of the power grid. A construction management module 30 is used to build an accident contingency plan knowledge base and to uniformly manage and classify the contingency plan knowledge base; The matching and fusion module 40 is used to automatically match the current scenario based on the multi-dimensional operation data of the power grid, and generate an accident handling plan based on the contingency plan knowledge base and the fusion of the whole network topology map; A filtering module 50 is established to obtain real-time alarm information of the power grid and establish a fault feature identification model of the equipment. Based on the fault feature identification model, the real-time alarm information is filtered and feature matched to identify fault information. The fault information and the real-time alarm information are aggregated to confirm the complete fault event. The scheduling and analysis module 60 is used to retrieve the entire network topology map, the operating data of each circuit system device, and historical handling plans for cross-analysis to obtain the scope of fault impact and its changing trend, and generate auxiliary handling strategies. The statistical analysis module 70 is used to perform quantitative statistics and evolution analysis based on the historical records and real-time ledgers of the complete fault events, so as to output statistical results.

[0050] In some optional embodiments, the acquisition and aggregation module 10 includes: A training unit is constructed to build an intelligent identification model of power grid topology features based on a graph neural network, and to train the intelligent identification model of power grid topology features. The identification and detection unit is used to perform target detection on the real-time power grid topology data based on the trained intelligent identification model of the power grid topology features, so as to identify weak nodes, risk points and equipment operation-related logic of the power grid.

[0051] In some alternative embodiments, the sensing and recognition module 20 includes: The sensing unit is used to capture several abnormal scenarios of the power grid in combination with the actual operating conditions of the power grid, and to perform real-time perception of the power grid operation status in the whole domain according to the full network topology map. The several abnormal scenarios include capturing accident tripping, equipment overload, abnormal operation mode, and daily operation risk.

[0052] In some alternative embodiments, the capturing sensing unit includes: Set up a filtering sub-unit to set the fault query time window and filter the set of valid alarms; The analysis and aggregation subunit is used to perform partitioning, permission verification, and deduplication on the alarm set data, as well as to analyze, aggregate, and sort the alarm set. The expression for filtering the set of valid alarms is: ; In the formula, This indicates the selection of the set of valid alarms. , These represent the start time and end time of the data retrieval, respectively. , These represent the set of fault keywords and the set of interference elimination keywords, respectively. Indicates the responsibility area code. Indicates the EMS system identifier. Indicates the alarm type. This indicates a change in remote signaling. Represents the sequence of events. This indicates the query function for alarms with the identifier 3000. This represents the alarm query function for a data source with an identifier of 5000 and an alarm type of remote signaling change. This represents an alarm query function for a data source with an identifier of 5000 and an alarm type of event sequence record.

[0053] In some alternative embodiments, the matching and fusion module 40 includes: The integrated linkage unit is used to integrate multi-dimensional data such as power grid topology, real-time equipment operating parameters, regional load distribution patterns, and dispatch safety control criteria, and link the planned operation risk warning list, real-time weak points in power grid operation, and typical fault handling cases from previous years to automatically match the current power grid operation scenario. The simulation generation unit is used to generate an accident handling plan that is adapted to the on-site working conditions by analyzing the accident-inducing factors and simulating the accident evolution path based on the contingency plan knowledge base and the integrated network topology map.

[0054] In some alternative embodiments, the filter establishment module 50 includes: The identification and filtering unit is used to automatically identify and filter valid fault alarm data based on the fault feature identification model; The judgment and aggregation unit is used to judge the fault type, fault location and fault cause based on the valid fault alarm data and in combination with the feature library, so as to obtain fragmented alarm information and aggregate the fragmented alarm information to obtain a complete fault event.

[0055] In some alternative embodiments, the statistical analysis module 70 includes: The acquisition unit is used to acquire the full historical data of power grid faults and the real-time fault ledger. The aggregation and mining unit is used to quantitatively aggregate, analyze trends and patterns in the full historical data and the real-time fault ledger based on fault type, fault event, equipment voltage level and type, and output specific statistical results.

[0056] The functions or operation steps implemented by the above modules and units are largely the same as those in the above method embodiments, and will not be repeated here.

[0057] The accident assessment and dynamic handling system provided in this embodiment of the invention has the same implementation principle and technical effects as the aforementioned method embodiment. For the sake of brevity, any parts not mentioned in the system embodiment can be referred to the corresponding content in the aforementioned method embodiment.

[0058] Example 3 The third embodiment of the present invention also proposes an electronic device, please refer to [link / reference]. Figure 3 The image shows an electronic device according to a third embodiment of the present invention.

[0059] The electronic device may include a processor 81 and a memory 82 storing computer program instructions.

[0060] Specifically, the processor 81 may include a central processing unit (CPU), an application specific integrated circuit (ASIC), or one or more integrated circuits that can be configured to implement this application.

[0061] The memory 82 may include a mass storage device for data or instructions. For example, and not limitingly, the memory 82 may include a hard disk drive (HDD), a floppy disk drive, a solid-state drive (SSD), flash memory, an optical disk drive, a magneto-optical disk drive, magnetic tape, or a Universal Serial Bus (USB) drive, or a combination of two or more of these. Where appropriate, the memory 82 may include removable or non-removable (or fixed) media. Where appropriate, the memory 82 may be internal or external to a data processing device. In a particular embodiment, the memory 82 is non-volatile memory. In a particular embodiment, the memory 82 includes read-only memory (ROM) and random access memory (RAM). Where appropriate, the ROM may be a mask-programmed ROM, a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), an electrically alterable read-only memory (EAROM), or flash memory, or a combination of two or more of these. Where appropriate, the RAM can be Static Random-Access Memory (SRAM) or Dynamic Random-Access Memory (DRAM). DRAM can be Fast Page Mode Dynamic Random-Access Memory (FPMDRAM), Extended Data Out Dynamic Random-Access Memory (EDODRAM), Synchronous Dynamic Random-Access Memory (SDRAM), etc.

[0062] The memory 82 can be used to store or cache various data files that need to be processed and / or communicated, as well as possible computer program instructions executed by the processor 81.

[0063] The processor 81 reads and executes the computer program instructions stored in the memory 82 to implement the accident assessment and dynamic handling method of the above embodiment 1.

[0064] In some embodiments, the electronic device may further include a communication interface 83 and a bus 80. For example, Figure 3 As shown, the processor 81, memory 82, and communication interface 83 are connected through bus 80 and complete communication with each other.

[0065] The communication interface 83 is used to enable communication between the various modules, devices, units, and / or equipment in this application. The communication interface 83 can also enable data communication with other components such as external devices, image / data acquisition devices, databases, external storage, and image / data processing workstations.

[0066] Bus 80 includes hardware, software, or both, that couples components of a device together. Bus 80 includes, but is not limited to, at least one of the following: data bus, address bus, control bus, expansion bus, and local bus. For example, and not as a limitation, bus 80 may include an Accelerated Graphics Port (AGP) or other graphics bus, an Extended Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a Hyper Transport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an InfiniBand interconnect, a Low Pin Count (LPC) bus, a memory bus, a Micro Channel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local Bus (VLB) bus, or other suitable buses, or a combination of two or more of these. Where appropriate, bus 80 may include one or more buses. Although this application describes and illustrates a specific bus, this application considers any suitable bus or interconnection.

[0067] The electronic device can acquire the accident assessment and dynamic handling system and execute the accident assessment and dynamic handling method of this embodiment.

[0068] Furthermore, in conjunction with the accident assessment and dynamic handling method in Embodiment 1 above, this application can provide a storage medium for implementation. This storage medium stores computer program instructions; when these computer program instructions are executed by a processor, they implement the accident assessment and dynamic handling method of Embodiment 1 above.

[0069] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0070] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these all fall within the protection scope of the present invention. Therefore, the protection scope of this invention patent should be determined by the appended claims.

Claims

1. A method for accident assessment and dynamic handling, characterized in that, The method includes: Real-time power grid topology data is acquired, and basic models of various circuit system devices are dynamically aggregated to construct and dynamically update the entire network topology map based on the basic models and the real-time power grid topology data. Based on the network topology map and the actual operating conditions of the power grid, the operating status of the power grid is perceived to identify abnormal operating conditions of the power grid. Construct an accident emergency response plan knowledge base, and manage and categorize the knowledge base in a unified manner; The system automatically matches the current scenario based on multi-dimensional power grid operation data, and generates an accident response plan based on the contingency plan knowledge base and the integrated network topology map. The system acquires real-time alarm information from the power grid and establishes a fault feature identification model for the equipment. Based on the fault feature identification model, the system filters and matches the real-time alarm information to identify fault information. The system then aggregates the fault information and the real-time alarm information to confirm the complete fault event. The entire network topology map, the operating data of each circuit system device, and historical contingency plans are retrieved and cross-analyzed to obtain the scope of the fault impact and its changing trend, and to generate auxiliary handling strategies. Quantitative statistics and evolution analysis are performed based on the complete historical records and real-time ledgers of the fault events to output statistical results.

2. The accident assessment and dynamic handling method according to claim 1, characterized in that, After the step of acquiring real-time power grid topology data, the method further includes: A power grid topology feature intelligent recognition model is constructed based on a graph neural network, and the power grid topology feature intelligent recognition model is trained. Based on the trained intelligent identification model of power grid topology features, target detection is performed on the real-time power grid topology data to identify weak nodes, risk points, and equipment operation-related logic.

3. The accident assessment and dynamic handling method according to claim 1, characterized in that, The step of sensing the power grid's operating status based on the entire network topology map and the actual operating conditions of the power grid, in order to identify abnormal operating conditions of the power grid, includes: The system captures several abnormal scenarios of the power grid based on the actual operating conditions of the power grid, and performs real-time perception of the power grid's operating status based on the entire network topology map. The abnormal scenarios include capturing accident tripping, equipment overload, abnormal operating mode, and daily operating risks.

4. The accident assessment and dynamic handling method according to claim 3, characterized in that, The steps for detecting accidental tripping include: Set a time window for fault querying and filter the set of valid alarms; Perform partition permission verification and deduplication on the alarm set data, and analyze, aggregate and sort the alarm set; The expression for filtering the set of valid alarms is: ； In the formula, This indicates the selection of the set of valid alarms. , These represent the start time and end time of the data retrieval, respectively. , These represent the set of fault keywords and the set of interference elimination keywords, respectively. Indicates the responsibility area code. Indicates the EMS system identifier. Indicates the alarm type. This indicates a change in remote signaling. Represents the sequence of events. This indicates the query function for alarms with the identifier 3000. This represents the alarm query function for a data source with an identifier of 5000 and an alarm type of remote signaling change. This represents an alarm query function for a data source with an identifier of 5000 and an alarm type of event sequence record.

5. The accident assessment and dynamic handling method according to claim 1, characterized in that, The steps of automatically matching the current scenario based on multi-dimensional power grid operation data and generating an accident response plan based on the contingency plan knowledge base and the integrated network topology map include: It integrates multi-dimensional data such as power grid topology, real-time equipment operating parameters, regional load distribution patterns, and dispatch safety control criteria, and links them with the planned operation risk warning list, real-time weak points in power grid operation, and typical fault handling cases from previous years to automatically match the current power grid operation scenario. Based on the aforementioned contingency plan knowledge base and by integrating the entire network topology map to analyze accident-inducing factors and deduce accident evolution paths, an accident handling contingency plan adapted to the on-site working conditions is generated.

6. The accident assessment and dynamic handling method according to claim 1, characterized in that, The step of filtering and matching the real-time alarm information based on the fault feature recognition model to identify fault information, and aggregating the fault information and the real-time alarm information to confirm the complete fault event includes: Based on the fault feature recognition model, the real-time alarm information is automatically identified and valid fault alarm data is filtered. Based on the valid fault alarm data and combined with the feature library, the fault type, fault location and fault cause are judged to obtain fragmented alarm information, and the fragmented alarm information is aggregated to obtain a complete fault event.

7. The accident assessment and dynamic handling method according to claim 1, characterized in that, The steps for performing quantitative statistical analysis and evolution analysis based on the historical records and real-time ledgers of the complete fault events to output statistical results include: Obtain full historical data on power grid faults and real-time fault ledgers; Based on fault type, fault event, equipment voltage level and type, the full historical data and the real-time fault ledger are quantitatively summarized, trend evolution is analyzed and patterns are mined, and special statistical results are output.

8. An accident assessment and dynamic response system, characterized in that, The system includes: The acquisition and aggregation module is used to acquire real-time power grid topology data and dynamically aggregate the basic models of various circuit system devices, so as to construct and dynamically update the entire network topology map based on the basic models and the real-time power grid topology data. The sensing and identification module is used to sense the operating status of the power grid based on the entire network topology map and the actual operating conditions of the power grid, so as to identify abnormal operating conditions of the power grid. A management module is built to construct an accident contingency plan knowledge base and to uniformly manage and classify the contingency plan knowledge base. The matching and fusion module is used to automatically match the current scenario based on multi-dimensional power grid operation data, and generate an accident handling plan based on the contingency plan knowledge base and the fusion of the whole network topology map; A filtering module is established to obtain real-time alarm information from the power grid and to establish a fault feature identification model for the equipment. Based on the fault feature identification model, the real-time alarm information is filtered and feature-matched to identify fault information. The fault information and the real-time alarm information are then aggregated to confirm the complete fault event. The scheduling and analysis module is used to retrieve the entire network topology map, the operating data of each circuit system device, and historical handling plans for cross-analysis to obtain the scope of fault impact and its changing trend, and generate auxiliary handling strategies. The statistical analysis module is used to perform quantitative statistics and evolution analysis based on the historical records and real-time ledgers of the complete fault events, so as to output statistical results.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the accident assessment and dynamic handling method as described in any one of claims 1 to 7.

10. A storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the accident assessment and dynamic handling method as described in any one of claims 1 to 7.

Images