Power grid abnormity identification method and related equipment

Abstract

The invention relates to the technical field of power systems, and discloses a power grid abnormity identification method and related equipment, and the method comprises the steps: obtaining the operation data of a target power grid; the operation data is archived, and the archived operation data is analyzed; constructing a power grid state diagram structure based on the power grid equipment state and the data analysis result of the target power grid; and performing anomaly identification on the target power grid according to the power grid state diagram structure. According to the invention, the problems of poor efficiency and low accuracy of a current power grid abnormity identification scheme can be solved.

Description

Technical Field

[0001] This application relates to the field of power technology technology, specifically to a method and related equipment for identifying power grid anomalies. Background Technology

[0002] As smart grids evolve towards high interconnectivity and dynamic complexity, real-time monitoring and anomaly identification of grid operation status have become core challenges. Under highly interconnected architectures, multiple grid devices, including distributed energy and energy storage devices, are accessed at high frequencies, forming a complex cross-regional network system. The electrical coupling and topological association between grid devices exhibit strong spatiotemporal nonlinear characteristics. In dynamic operating environments, load fluctuations, changes in renewable energy output, and the commissioning and decommissioning of grid devices cause the grid status to continuously evolve. The explosive growth and intertwined characteristics of multi-source heterogeneous data place higher demands on the accuracy of anomaly feature capture and the timeliness of analysis. Minor anomalies in grid devices in a certain region may rapidly spread to the entire system through network coupling. Constructing an intelligent monitoring system that covers the entire grid topology, integrates multi-source data, and adapts to dynamic scenarios, and achieves second-level anomaly perception, accurate location, and trend prediction, is key to ensuring the safe and reliable operation of the grid and improving system resilience.

[0003] Traditional flat data storage architectures struggle to establish correlations between topological events and historical data, leading to difficulties in reproducing historical fault modes and reliance on manual investigation for tracing anomalies. This makes it difficult to quickly pinpoint the root causes of sensor errors and parameter configuration mistakes. Furthermore, traditional anomaly identification methods rely solely on single-node data threshold judgments, neglecting the graph structure characteristics of static topological connections, dynamic electrical correlations, and historical fault synergy among power grid equipment. This makes it difficult to identify complex faults caused by topological anomalies or collaborative degradation of power grid equipment groups, resulting in frequent false alarms and missed alarms. Therefore, current power grid anomaly identification schemes suffer from poor timeliness and low accuracy. Summary of the Invention

[0004] This application provides a power grid anomaly identification method and related equipment, which can solve the problems of poor timeliness and low accuracy of current power grid anomaly identification schemes.

[0005] In a first aspect, embodiments of this application provide a method for identifying power grid anomalies, including: Obtain operational data from the target power grid; The operational data is archived, and the archived operational data is analyzed. Based on the power grid equipment status and data analysis results of the target power grid, a power grid status diagram structure is constructed; Anomalies in the target power grid are identified based on the power grid state diagram structure.

[0006] Optionally, in some embodiments of this application, the step of identifying anomalies in the target power grid based on the power grid state diagram structure includes: Extract the real-time operating characteristics of each node from the power grid state diagram structure; Based on the aforementioned real-time operating characteristics, candidate nodes are determined from multiple nodes; The candidate nodes are identified as abnormal based on a pre-defined database of abnormal cases.

[0007] Optionally, in some embodiments of this application, the step of identifying anomalies in the candidate nodes based on a preset anomaly case library includes: Obtain the feature deviation pattern of the candidate node; Based on the feature deviation pattern and the pre-fault characteristics of abnormal cases in the preset abnormal case library, the candidate nodes are identified as abnormal.

[0008] Optionally, in some embodiments of this application, the step of identifying anomalies in the candidate nodes based on the feature deviation pattern and the pre-failure features of anomaly cases in a preset anomaly case library includes: Determine the deviation parameters corresponding to the feature deviation pattern; The determined deviation parameters are compared with the pre-fault characteristics of abnormal cases in the preset abnormal case library in multiple dimensions to identify the abnormal state of the candidate node.

[0009] Optionally, in some embodiments of this application, the step of constructing a power grid state diagram structure based on the power grid equipment status and data analysis results of the target power grid includes: Obtain the basic graph structure; Map the state of the power grid equipment in the target power grid to the node attributes of the basic graph structure; The data analysis results are mapped to the node dynamic attributes and edge dynamic attributes of the basic graph structure; Based on the mapped graph structure, the power grid state graph structure of the target power grid is constructed.

[0010] Optionally, in some embodiments of this application, constructing the power grid state diagram structure of the target power grid based on the mapped graph structure includes: Integrate the node and edge attributes of the mapped graph structure; The integrated graph structure is validated. When the integrated graph structure passes the verification, the integrated graph structure is determined as the power grid state graph structure of the target power grid.

[0011] Optionally, in some embodiments of this application, the step of archiving the runtime data and analyzing the archived runtime data includes: The running data is archived according to the preset storage structure; Extract data characteristics from the archived runtime data; The archived operational data is analyzed based on the aforementioned data characteristics.

[0012] Secondly, embodiments of this application provide a power grid anomaly identification device, comprising: The acquisition module is used to acquire the operating data of the target power grid; The analysis module is used to archive the operational data and analyze the archived operational data. The construction module is used to construct a power grid state diagram structure based on the power grid equipment status and data analysis results of the target power grid; The identification module is used to identify anomalies in the target power grid based on the power grid state diagram structure.

[0013] Thirdly, embodiments of this application provide 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 steps of the power grid anomaly identification method as described in the first aspect.

[0014] Fourthly, embodiments of this application provide a storage medium storing a computer program that can be loaded by a processor and executed as the power grid anomaly identification method described in the first aspect.

[0015] This application provides a method, apparatus, electronic device, and storage medium for power grid anomaly identification. After acquiring operational data of a target power grid, the operational data is archived and analyzed. Then, based on the power grid equipment status and data analysis results, a power grid state graph structure is constructed. Finally, anomalies in the target power grid are identified according to the power grid state graph structure. In the power grid anomaly identification scheme provided in this application, topological associations are established during data archiving, overcoming the pain points of traditional storage lacking topological associations and difficulty in fault reproduction. The constructed power grid state graph integrates equipment status with global topology and electrical association features, compensating for the shortcomings of traditional methods that ignore graph structures. Anomalies are identified multi-dimensionally based on the graph structure, avoiding false alarms and missed alarms due to single-node threshold judgments, ultimately improving the timeliness and accuracy of anomaly identification, thereby solving the problems of poor timeliness and low accuracy in current power grid anomaly identification schemes. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1This is an application environment diagram of the power grid anomaly identification method provided in the embodiments of this application; Figure 2 This is a flowchart illustrating the power grid anomaly identification method provided in an embodiment of this application; Figure 3 This is a schematic diagram of the power grid anomaly identification system provided in an embodiment of this application; Figure 4 This is a schematic diagram of the power grid anomaly identification device provided in the embodiments of this application; Figure 5 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0018] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numerals in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of systems and methods consistent with those detailed in the appended claims or with some aspects of this application.

[0019] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover descriptions such as non-exclusive inclusion, so that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element. Furthermore, components, features, and elements with the same names in different embodiments of this application may have the same meaning or different meanings, the specific meaning of which must be determined by its interpretation in that specific embodiment or further in conjunction with the context of that specific embodiment.

[0020] It should be understood that the specific embodiments described herein are merely illustrative of this application and are not intended to limit this application.

[0021] In the following description, the use of suffixes such as "module," "part," or "unit" to denote elements is solely for the purpose of illustrative purposes and has no specific meaning in itself. Therefore, "module," "part," or "unit" may be used interchangeably.

[0022] To address the aforementioned technical problems and overcome the shortcomings of existing technologies, this application provides a power grid anomaly identification method and related equipment, which can accurately determine and quickly protect fault areas in active distribution networks, thereby improving the reliability and adaptability of differential protection for active distribution networks.

[0023] Figure 1 This is a diagram illustrating the application environment of a power grid anomaly identification method in one embodiment. (Refer to...) Figure 1 This power grid anomaly identification method is applied to a power grid anomaly identification system. The system includes a terminal 110 and a server 120. The terminal 110 and server 120 are connected via a network. The terminal 110 can be a desktop terminal or a mobile terminal; the mobile terminal can be at least one of a mobile phone, tablet, or laptop. The server 120 can be a standalone server or a server cluster consisting of multiple servers.

[0024] Server 120 is configured to execute the above-mentioned power grid anomaly identification method, including: acquiring the operating data of the target power grid; archiving the operating data and analyzing the archived operating data; constructing a power grid state diagram structure based on the power grid equipment status and data analysis results of the target power grid; and identifying anomalies in the target power grid according to the power grid state diagram structure.

[0025] This application establishes topological associations during data archiving, overcoming the pain points of traditional storage lacking topological associations and being difficult to reproduce faults. It constructs a power grid state graph that integrates equipment status with global topology and electrical association features, making up for the shortcomings of traditional methods that ignore graph structures. Based on the graph structure, it identifies anomalies from multiple dimensions, avoiding false alarms and missed alarms due to single-node threshold judgments, and ultimately improving the timeliness and accuracy of anomaly identification. This solves the problems of poor timeliness and low accuracy in current power grid anomaly identification schemes.

[0026] Please see Figure 2 , Figure 2 This is a flowchart illustrating a power grid anomaly identification method according to an embodiment of this application. This embodiment primarily uses the application of this power grid anomaly identification method to a server as an example for illustration. Specifically, the power grid anomaly identification method provided in this embodiment may include the following steps: S101. Obtain the operating data of the target power grid.

[0027] The target power grid's operational data is multi-dimensional data reflecting the real-time operating status and related characteristics of power grid equipment, including basic operating parameters and multi-source correlated data. Among them, basic operating parameters are the conventional core operating indicators of power grid equipment, including the effective value of voltage and the effective value of current, which are used to determine whether the basic operation of the equipment is stable. Multi-source correlated data are extended data that help determine the health of equipment and potential anomalies, including vibration signals, real-time oil temperature curves, and load fluctuation data, which are used to capture subtle signs such as equipment aging and latent faults.

[0028] For example, sampling strategies can be dynamically adjusted according to the status of power grid equipment to collect basic operating parameters and multi-source correlation data. At the same time, by combining historical fault records of power grid equipment, preventive test data and real-time status monitoring signals, the status of each device in the target power grid can be determined in real time. This status includes healthy status, aging status and warning status. Among them, healthy status means that the equipment operating parameters are stable for a long time, there are no historical fault records, and the preventive test results meet the standards; aging status means that the equipment has been in operation for a long time, or some parameters have slight fluctuations; warning status means that the equipment has recorded parameters exceeding the threshold, or that related equipment has experienced a fault.

[0029] Next, the sampling frequency and the type of data collected are adjusted to achieve targeted data acquisition. For example, for devices in a healthy state, a low-frequency sampling strategy is implemented to collect only basic operating parameters and reduce data redundancy. For devices in an aging state or a warning state, a high-frequency sampling mode is triggered to collect basic operating parameters and multi-source correlated data simultaneously to ensure that abnormal signs are captured.

[0030] S102. Archive the running data and analyze the archived running data.

[0031] Operational data archiving refers to marking operational data directly related to topology changes such as equipment decommissioning or topology reconfiguration as archived data when the power grid experiences such events. This process generates a unique topology version number with a timestamp, stores the data according to a three-tier storage architecture, and binds it to key data processing information.

[0032] For example, specifically, the operating status of the target power grid is monitored in real time to identify two key events: equipment decommissioning or topology reconfiguration. If such events are detected, historical operating data of the power grid equipment associated with the change is filtered and uniformly marked as archived. Then, archived data with a storage period less than or equal to a first threshold is stored in a cache medium; archived data with a storage period greater than the first threshold and less than or equal to a second threshold is stored in a solid-state drive; archived data with a storage period greater than the second threshold is transferred to a Blu-ray storage medium; and sampling strategies, data cleaning algorithm parameters, input / output feature summaries, processing timestamps, and operation subjects are bound to the corresponding archived data and stored together in their respective storage areas.

[0033] Furthermore, based on the key information bound to the archived data, a directed acyclic graph (DAG) is used to visualize the data processing flow. Specifically, nodes in the DAG represent processing steps, and edges represent data flow and transformation logic. Each node stores key information corresponding to the archived data. If an abnormality in equipment operation is subsequently identified, the archived data corresponding to the abnormality can be located through the ledger, and its original measurement data source and parameter configuration of the processing steps can be traced to investigate whether the abnormality was caused by the data processing stage.

[0034] Next, the archived data under different topology version numbers are compared and analyzed, such as comparing the differences in operating parameters of similar equipment before and after topology refactoring, and assessing the impact of topology changes on power grid operating efficiency and stability.

[0035] Optionally, in some embodiments of this application, the step "archiving the runtime data and analyzing the archived runtime data" may specifically include: The running data is archived according to the preset storage structure; Extract data characteristics from the archived runtime data; The archived operational data is analyzed based on the aforementioned data characteristics.

[0036] For example, traceability of the entire data processing process can be achieved by constructing a digital ledger. A directed acyclic graph (DAG) is used to visualize the data processing flow, where nodes correspond to each processing step, and edges are used to represent the data flow and data transformation logic. Each node stores key information, including summary information of input and output characteristics, parameter configuration details of the processing step, timestamp of the processing operation, and the subject that performed the operation. This digital ledger supports reverse tracing of abnormal features. When a voltage abnormality is detected at any node, the parameter configuration details of the processing step corresponding to the abnormality can be directly located through the ledger, and the original measurement data source of the original data can be traced back to achieve full-link traceability from abnormal results to original data.

[0037] Furthermore, a historical normal feature range can be embedded in each node. If the data processed by a certain node exceeds this range, it can be traced back through the ledger: locate the parameter configuration of the corresponding processing step and the source of the original measurement data to investigate the root cause of the anomaly.

[0038] S103. Based on the power grid equipment status and data analysis results of the target power grid, construct a power grid status diagram structure.

[0039] The power grid state diagram consists of nodes V and edges E. The attributes of node V include basic equipment parameters, real-time operating characteristics, and health status labels; edge E includes static topology connections, static connection status identifiers, dynamic electrical associations, and historical association characteristics.

[0040] Specifically, the state of the power grid equipment in the target power grid is bound to the V attribute of the node. At the same time, the data analysis results are mapped to the dynamic attributes of the nodes and edges. Based on the attribute mapping results, an initial graph structure containing static structure and dynamic features is built. Then, according to the real-time topology changes of the power grid, the graph structure is dynamically adjusted, and a topology version number is assigned to the adjusted power grid state graph to be consistent with the version number of the archived data in the data analysis results. Thus, the power grid state graph structure is formed.

[0041] Optionally, in some embodiments of this application, the step of "constructing a power grid state diagram structure based on the power grid equipment status and data analysis results of the target power grid" may specifically include: Obtain the basic graph structure; Map the state of the power grid equipment in the target power grid to the node attributes of the basic graph structure; The data analysis results are mapped to the node dynamic attributes and edge dynamic attributes of the basic graph structure; Based on the mapped graph structure, the power grid state graph structure of the target power grid is constructed.

[0042] For example, from the equipment status determination results of the target power grid, the health status labels of each device are extracted, and the health status labels are directly mapped to the health status label field of the corresponding node in the basic diagram. Together with the pre-bound basic parameters of the device, they constitute the complete static attributes of the node. For nodes that have not pre-bound basic parameters (such as newly added device nodes), the parameters are retrieved from the power grid equipment ledger and the mapping is completed.

[0043] Simultaneously, the real-time operating characteristics of each device are extracted from the data analysis results, and the real-time operating characteristics are mapped to the node dynamic attribute fields of the corresponding nodes according to the timestamp. In addition, the dynamic electrical association data in the data analysis results is extracted and mapped to the dynamic electrical association fields of the corresponding edges. Furthermore, the historical association characteristics between devices can be extracted from the archived data of the data analysis results and mapped to the historical association characteristic fields of the corresponding edges. Furthermore, the system monitors whether events such as equipment decommissioning or topology reconfiguration occur in the power grid. If equipment decommissioning occurs, the historical attributes of the decommissioned equipment nodes are retained, and dynamic features are stopped from being updated. For newly added equipment, nodes and edges are added and activated to ensure that the graph structure is consistent with the actual topology of the power grid. Assign a final topology version number to the completed graph structure, keeping it consistent with the version number of the archived data in the data analysis results. Finally, output the data-driven graph structure, namely the power grid state graph structure.

[0044] Optionally, in some embodiments of this application, the step "constructing the power grid state diagram structure of the target power grid based on the mapped graph structure" may specifically include: Integrate the node and edge attributes of the mapped graph structure; The integrated graph structure is validated. When the integrated graph structure passes the verification, the integrated graph structure is determined as the power grid state graph structure of the target power grid.

[0045] S104. Anomaly identification is performed on the target power grid based on the power grid state diagram structure.

[0046] For example, specifically, key attributes of node V are extracted, such as the real-time operating characteristics and health status labels of each node, while key attributes of edge E are extracted, such as the static topological connections and dynamic electrical connections of candidate node associated edges, as well as the historical association characteristics of all edges. Then, the historical normal feature range obtained from the archived data is retrieved, and the extracted real-time operating features are compared one by one with the corresponding historical normal feature range of the health status. If the real-time features of a node exceed the historical normal range, the node is marked as a "candidate abnormal node". Next, for each candidate abnormal node, the static topology connection attributes of all associated edges in the power grid status diagram are checked: if the static connection status of the edge is marked as connected, but the actual physical connection is detected as disconnected, or the inherent parameters of the line (such as resistance) do not match the ledger, it indicates that there is a topology anomaly, and the candidate node should be retained; if the real-time power transmission of the edge is 0, or the voltage phase angle difference exceeds the physical constraint range of 0-180°, it indicates that there is an electrical correlation anomaly. In this case, feature matching is further performed in conjunction with the anomaly case library. Specifically, a pre-built anomaly case library of similar equipment is retrieved, and then the feature deviation pattern of the candidate abnormal node is extracted, including the deviation magnitude of the real-time features from the historical normal range, the trend of feature changes over time, the start time of the deviation, and the current duration. Finally, based on the feature deviation pattern and the anomaly case library, the target power grid is identified as abnormal.

[0047] Optionally, in some embodiments of this application, the step of "identifying anomalies in the target power grid based on the power grid state diagram structure" may specifically include: Extract the real-time operating characteristics of each node from the power grid state diagram structure; Based on the aforementioned real-time operating characteristics, candidate nodes are determined from multiple nodes; The candidate nodes are identified as abnormal based on a pre-defined database of abnormal cases.

[0048] Optionally, in some embodiments of this application, the step of "identifying anomalies in the candidate nodes according to a preset anomaly case library" may specifically include: Obtain the feature deviation pattern of the candidate node; Based on the feature deviation pattern and the pre-fault characteristics of abnormal cases in the preset abnormal case library, the candidate nodes are identified as abnormal.

[0049] Optionally, in some embodiments of this application, the step of "identifying anomalies in the candidate nodes based on the feature deviation pattern and the pre-failure features of anomaly cases in a preset anomaly case library" may specifically include: Determine the deviation parameters corresponding to the feature deviation pattern; The determined deviation parameters are compared with the pre-fault characteristics of abnormal cases in the preset abnormal case library in multiple dimensions to identify the abnormal state of the candidate node.

[0050] To further understand the power grid anomaly identification scheme of this application, the following will describe the power grid anomaly identification system in more detail. For specifics, please refer to [link to relevant documentation]. Figure 3 The power grid anomaly identification system includes a data acquisition module, a processing module, an identification module, and an interaction module. The acquisition module is used to collect raw data from the acquisition unit and transmit it to the processing module; The processing module is used to analyze and process the raw data through the processing unit, generate analysis results, and transmit them to the recognition module; The identification module is used to identify anomalies in the analysis results, output the anomaly identification results, and transmit them to the interaction module. The interactive module is used to visualize and interact with the anomaly identification results.

[0051] In one embodiment, the system includes a data acquisition module, a processing module, an identification module, and an interaction module. The data acquisition module dynamically adjusts the sampling strategy according to the power grid equipment status through the acquisition unit, collects basic operating parameters and multi-source associated data, and transmits them to the processing module. The processing module performs hierarchical storage of the raw data and constructs a data traceability ledger, generating analysis results including power grid equipment operating characteristics and topology associations, which are then sent to the identification module. The identification module identifies anomalies based on the power grid status diagram, using a graph neural network combined with a historical case library, and outputs anomaly identification results including power grid equipment identifiers, feature parameters, and timestamps, and associates the topology version with the traceability ledger to generate a report. The interaction module visualizes the power grid status diagram, highlights abnormal nodes in red, and supports interactive operations such as hover traceability pop-ups, topology version retrieval, and anomaly alarm parameter push. The data acquisition module, processing module, identification module, and interaction module work together through data interfaces to achieve dynamic acquisition, intelligent processing, accurate identification, and efficient interaction of power grid anomalies.

[0052] The raw data includes basic operating parameters and multi-source correlated data; The data acquisition unit includes a lifecycle sensing unit for power grid equipment, which dynamically adjusts the sampling strategy based on the status of the power grid equipment. The status of power grid equipment includes healthy power grid equipment, aging power grid equipment, and power grid equipment in warning condition.

[0053] Sampling strategies include low-frequency sampling strategies and high-frequency sampling strategies; For grid equipment in a healthy state, a low-frequency sampling strategy is implemented to collect basic operating parameters; Basic operating parameters include the effective values ​​of voltage and current; For aging or early warning power grid equipment, a high-frequency sampling mode is automatically triggered to collect multi-source correlated data. Multi-source correlated data includes vibration signals, real-time oil temperature curves, and load fluctuation data; Vibration signals include the peak value of the vibration signal.

[0054] In one embodiment, a data acquisition unit integrating a lifecycle sensing unit achieves differentiated acquisition of raw data. The acquisition unit incorporates a status monitoring sensor and an intelligent decision-making module to acquire real-time health status tags of power grid equipment, including healthy, aging, and warning-prone equipment. It dynamically adjusts the sampling strategy based on preset rules, with a low-frequency sampling strategy set to once every 10 minutes (based on the low data redundancy requirements of conventional power grid equipment health status monitoring). This strategy is suitable for healthy power grid equipment, acquiring basic operating parameters such as effective voltage (V) and effective current (A), and transmitting them to the processing module via RS485 or Ethernet interface. When the power grid equipment is determined to be aging equipment... The system automatically triggers a high-frequency sampling mode (1 time / second to meet the real-time capture requirements of fault symptoms) for early warning power grid equipment (such as vibration signal peak exceeding twice the average value of power grid equipment in a healthy state, oil temperature fluctuation exceeding the historical average by ±10%, based on the abnormal characteristic threshold standard in the field of power grid equipment condition monitoring). In addition to basic operating parameters, it additionally collects multi-source correlated data such as vibration signals (including peak data, unit: m / s²), real-time oil temperature curve (resolution 0.1℃), and load fluctuation data (power change rate, unit: kW / s), and uploads them in encrypted form. The life cycle sensing unit realizes the status judgment by integrating historical fault records of power grid equipment, preventive test data, and real-time operating parameters. The acquisition unit supports the Modbus and IEC 61850 communication protocols and can be seamlessly connected to monitoring terminals in different scenarios to build a differentiated data acquisition system for all power grid equipment.

[0055] The processing module includes a hierarchical storage and archiving unit and a data processing traceability unit; Tiered storage and archiving units are used to establish archiving strategies associated with power grid topology change events, specifically including: When a power grid device is decommissioned or a topology reconfiguration event is detected, the historical data of the associated power grid device is automatically marked as archived, and a topology version number including a timestamp is generated. Historical data in the archived state is stored using a three-tier storage architecture, which includes a real-time data area, a recent data area, and a historical data area. The real-time data area is used to store historical data with a storage period less than or equal to the first threshold, and is cached. The recent data area is used to store historical data with a storage period of the second threshold. It uses solid-state drives for storage and supports second-level retrieval. The historical data area is used to store historical data with a storage period greater than the third threshold. This data is transferred to Blu-ray storage media and indexed according to the topology version number.

[0056] In one embodiment, an archiving strategy is established to associate with power grid topology change events (including power grid equipment decommissioning and topology reconfiguration). When such events are detected, the historical data of the associated power grid equipment is automatically marked as archived, and a topology version number including a timestamp accurate to the second is generated to identify the corresponding power grid structure version. A three-level storage architecture is used to classify and store the archived data. The real-time data area stores historical data with a storage period of less than or equal to 1 day (the first threshold is set to 1). Based on the high-frequency access requirements of real-time power grid monitoring for the latest data, a high-speed cache medium is used to achieve nanosecond-level reading. The recent data area stores historical data with a storage period of 1 day to 1 year (the second threshold is set to 1 day to 1 year). Considering the second-level retrieval requirements of data from the past year in power grid operation and maintenance, solid-state drives (SSDs) are used for storage and a B+ tree index is built. The historical data area stores historical data with a storage period of more than 1 year (the third threshold is set to 1 year). Due to the cost control and data reliability requirements of long-term archiving, the data is transferred to Blu-ray storage medium and a hash index is built according to the topology version number to support batch data backtracking based on the power grid structure version.

[0057] A digital ledger covering the entire data processing process is constructed through a data processing traceability unit, specifically including: A visualization process for directed acyclic graphs (DAGs) is adopted, where DAGs include nodes and edges. Nodes represent processing steps, and edges represent data flow and transformation logic; Each node stores key information, including input / output feature summaries, parameter configuration details, processing timestamps, and the entity performing the operation. Digital ledgers support reverse tracing based on anomaly characteristics, specifically including: When a voltage anomaly is detected at any node, the digital ledger is used to locate the parameter configuration details of the corresponding processing step and the original measurement data source.

[0058] In one embodiment, the data processing traceability unit achieves traceability of the processing process by constructing a digital ledger of the entire data processing workflow. The data processing workflow is visualized using a directed acyclic graph (DAG), where nodes correspond to each processing step, and edges are used to represent the data flow and data transformation logic. Each node stores key information, including summary information of input and output features, parameter configuration details of the processing step, timestamp of the processing operation, and the subject executing the operation. The digital ledger supports reverse tracing of abnormal features. When a voltage abnormality is detected at any node, the parameter configuration details of the processing step corresponding to the abnormality can be directly located through the ledger, and the original measurement data source of the original data can be traced back to achieve full-link traceability from the abnormal result to the original data.

[0059] Map archived historical data, key information, and power grid equipment status into a power grid status diagram; The power grid state diagram includes node V attributes and edge E attributes; Node V attributes include basic parameters of power grid equipment, real-time operating characteristics, and health status labels; Edge attributes include static topological connectivity, dynamic electrical association, and historical association characteristics; Static topology connections include the physical connection methods between power grid devices, inherent attributes of the topology, and static connection status identifiers; Dynamic electrical correlations include real-time power transfer and voltage phase angle difference; Historical correlation characteristics include historical fault correlation, long-term operating parameter correlation trends, and maintenance record correlation; When a topology change event is received, the activation and freeze states of the corresponding static topology connection edges in the power grid state diagram are switched within a preset time.

[0060] In one embodiment, historical data of the archived status, key information of the data processing traceability unit (including processing steps and parameter configurations), and the status of power grid equipment (including healthy power grid equipment, aging power grid equipment, and early warning power grid equipment) are mapped into a graph structure. The node V attribute integrates the basic parameters of the power grid equipment (such as model and rated voltage), real-time operating characteristics (effective voltage value, peak vibration signal value), and health status label. The edge E attribute includes three types of association characteristics: static topology connection (physical connection method, inherent attributes of topology structure, and static connection status identifier), dynamic electrical association (real-time power transmission and voltage phase angle difference), and historical association characteristics (historical fault coordination, operating parameter association trend, and maintenance record association). When a topology change event (such as power grid equipment decommissioning or topology reconfiguration) is received, based on the requirements of real-time power grid monitoring for topology response speed, the activation and freezing states of the corresponding static topology connection edges in the power grid status graph are switched within a preset time of 50ms (the preset time is set to 50ms), ensuring that the graph structure reflects changes in the physical connection of the power grid in real time.

[0061] Anomaly identification is performed based on the power grid state diagram, and the anomaly identification results are output, specifically including: Extract the real-time operating characteristics of node V in the power grid status diagram, compare them with the historical normal characteristic range under the same health status label in the digital ledger, and screen out candidate abnormal nodes whose real-time operating characteristic values ​​exceed the historical normal characteristic range. Real-time operating characteristics include the effective value of voltage and the peak value of vibration signal; Verify the edge E attribute associated with the candidate abnormal node; If the static topological connections in the edge E attribute have abnormal physical connection relationships; If the parameter value of the dynamic electrical association corresponding to the static topology connection is zero or exceeds the theoretical constraint range, it is marked as a suspected anomaly.

[0062] In one embodiment, the real-time operating characteristics (including effective voltage value and peak vibration signal) of node V are extracted from the power grid state diagram and compared with the historical normal characteristic range under the same health status label (healthy power grid equipment, aging power grid equipment, and early warning power grid equipment) in the digital ledger. Nodes whose characteristic values ​​exceed the historical normal range are selected as candidate abnormal nodes. For candidate abnormal nodes, the associated edge E attribute is verified. If the static topology connection in the edge E attribute has an abnormal physical connection method (such as disconnection and incorrect status identification), or the dynamic electrical association parameters (real-time power transmission and voltage phase angle difference) corresponding to the static topology connection have a value of zero or exceed the theoretical constraint range of the power system (such as voltage phase angle difference exceeding ±30°), then the candidate node is marked as a suspected abnormality, and the preliminary abnormality identification is completed.

[0063] Based on the abnormal case library of similar power grid equipment in the historical data area, if the feature deviation pattern of the candidate abnormal node matches the pre-fault features above the fourth threshold in the abnormal case library, and the duration is greater than or equal to the fifth threshold, it is judged as abnormal. The anomaly identification results include the anomaly power grid equipment identifier, anomaly characteristic parameters, and anomaly occurrence timestamp; The anomaly identification results are then linked with the topology version number and digital ledger records to generate a traceability report.

[0064] In one embodiment, a database of abnormal cases of similar power grid equipment in the historical data area is used to make a final judgment on candidate abnormal nodes. If the matching degree between its feature deviation pattern and the pre-fault features in the case database reaches more than 80% (the fourth threshold is set to 80% to ensure that the feature similarity is sufficient to support the judgment based on the accuracy requirements of fault mode recognition), and the duration of the deviation state is greater than or equal to 10 seconds (the fifth threshold is set to 10 seconds to exclude instantaneous interference, based on the characteristic that power grid abnormalities must have continuity), then it is judged as abnormal. The abnormal identification result includes the abnormal power grid equipment identifier, abnormal feature parameters (such as the voltage effective value exceeding the standard value and the vibration signal peak value) and the timestamp of the abnormal occurrence. The result is associated with the corresponding topology version number and the processing record of the digital ledger to automatically generate a traceability report including full-link information.

[0065] The anomaly identification results are visualized and interactively displayed through a human-computer interaction interface: The power grid status diagram is displayed visually, with abnormal nodes highlighted in red. A data processing and tracing pop-up window appears when the mouse is hovered over the node. The data processing traceability pop-up includes a complete processing log of the node characteristics of the abnormal node.

[0066] The human-computer interaction interface supports retrieving historical anomaly records by topology version number; When a description of a topology change event is entered, the system automatically selects the corresponding archived data area for the abnormal record and marks the sampling strategy at that time. When an abnormal alarm message is output, the parameter configuration details of the related data processing stage are pushed simultaneously.

[0067] In one embodiment, the anomaly identification results are visualized and interacted with through a human-computer interaction interface. In the power grid status diagram visualization interface, anomaly nodes are marked with red highlights (RGB value #FF0000, following the general visual standard for power system anomaly warnings). When the mouse hovers over the node, a data processing traceability pop-up window automatically appears. The pop-up window covers the entire process log of the anomaly node, including the input and output data of node characteristics, processing step parameter configuration, operation timestamp, and associated original measurement data source. The human-computer interaction interface supports searching historical anomaly records by topology version number. When a description of a topology change event (such as "202508") is entered, the anomaly record can be retrieved. During the "01 busbar B1 reconfiguration" process, based on the hierarchical storage topology version number index, the system automatically matches the corresponding archived anomaly records in the historical data area and marks the sampling strategy at that time (e.g., low-frequency sampling once every 10 minutes for healthy grid equipment or high-frequency sampling once per second for aging grid equipment). When an anomaly alarm information is output, the system simultaneously pushes the parameter configuration details of the associated data processing steps, including data cleaning algorithm parameters, graph neural network model input feature list, and the main operation entity of the processing steps. This achieves efficient interaction between "anomaly visualization and location - historical record tracing - processing parameter verification," meeting the grid operation and maintenance needs for intuitive presentation and rapid response to anomaly information. In summary, the power grid anomaly identification method provided in this embodiment, after acquiring the operating data of the target power grid, archives the operating data and analyzes the archived operating data. Then, based on the power grid equipment status and data analysis results of the target power grid, a power grid state graph structure is constructed. Finally, anomalies in the target power grid are identified according to the power grid state graph structure. In the power grid anomaly identification scheme provided in this application, topological associations are established during data archiving, overcoming the pain points of traditional storage lacking topological associations and difficulty in fault reproduction. The constructed power grid state graph integrates equipment status with global topology and electrical association features, compensating for the shortcomings of traditional methods that ignore graph structures. Anomalies are identified multi-dimensionally based on the graph structure, avoiding false alarms and missed alarms due to single-node threshold judgments, ultimately improving the timeliness and accuracy of anomaly identification, thereby solving the problems of poor timeliness and low accuracy in current power grid anomaly identification schemes.

[0068] It should be understood that, although Figure 2 The steps in the flowchart are shown sequentially as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order in which these steps are executed, and they can be performed in other orders. Figure 2At least some of the steps in the process may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least some of the sub-steps or stages of other steps.

[0069] To facilitate better implementation of the power grid anomaly identification method of this application, this application also provides a power grid anomaly identification device based on the above-described power grid anomaly identification method. The meanings of the terms used are the same as in the power grid anomaly identification method described above, and specific implementation details can be found in the descriptions in the method embodiments.

[0070] Please see Figure 4 , Figure 4 The diagram below illustrates the structure of a power grid anomaly identification device provided in this embodiment. Specifically, the device may include an acquisition module 201, an analysis module 202, a construction module 203, and an identification module 204, as follows: The acquisition module 201 is used to acquire the operating data of the target power grid; Analysis module 202 is used to archive the running data and analyze the archived running data; The construction module 203 is used to construct a power grid state diagram structure based on the power grid equipment status and data analysis results of the target power grid; The identification module 204 is used to identify anomalies in the target power grid based on the power grid state diagram structure.

[0071] Specific limitations regarding the power grid anomaly detection device can be found in the limitations of the power grid anomaly detection method described above, and will not be repeated here. Each module in the aforementioned power grid anomaly detection device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the corresponding operations of each module.

[0072] The power grid anomaly identification device provided in this embodiment includes an acquisition module 201 acquiring the operating data of the target power grid, an analysis module 202 archiving the operating data and analyzing the archived data, a construction module 203 constructing a power grid state diagram structure based on the power grid equipment status and data analysis results of the target power grid, and finally, an identification module 204 identifying anomalies in the target power grid according to the power grid state diagram structure. In the power grid anomaly identification scheme provided in this application, topological associations are established during data archiving, overcoming the pain points of traditional storage lacking topological associations and difficulty in fault reproduction. The constructed power grid state diagram integrates equipment status with global topology and electrical association features, compensating for the shortcomings of traditional methods that ignore graph structures. Anomalies are identified multi-dimensionally based on the graph structure, avoiding false alarms and missed alarms due to single-node threshold judgments, ultimately improving the timeliness and accuracy of anomaly identification, thereby solving the problems of poor timeliness and low accuracy in current power grid anomaly identification schemes.

[0073] Furthermore, embodiments of this application also provide an electronic device, such as... Figure 5 As shown, it illustrates a structural schematic diagram of the electronic device involved in the embodiments of this application, specifically: The electronic device may include components such as a processor 301 with one or more processing cores, a memory 302 with one or more computer-readable storage media, a power supply 303, and an input unit 304. Those skilled in the art will understand that... Figure 5 The electronic device structure shown does not constitute a limitation on the electronic device and may include more or fewer components than shown, or combine certain components, or have different component arrangements. Wherein: The processor 301 is the control center of the electronic device. It connects various parts of the electronic device via various interfaces and lines, and performs various functions and processes data by running or executing software programs and / or modules stored in the memory 302, and by calling data stored in the memory 302, thereby providing overall monitoring of the electronic device. Optionally, the processor 301 may include one or more processing cores; preferably, the processor 301 may integrate an application processor and a modem processor, wherein the application processor mainly handles the operating system, user interface, and applications, and the modem processor mainly handles wireless communication. It is understood that the modem processor may not be integrated into the processor 301.

[0074] The memory 302 can be used to store software programs and modules. The processor 301 executes various functional applications and power grid anomaly identification methods by running the software programs and modules stored in the memory 302. The memory 302 may mainly include a program storage area and a data storage area. The program storage area may store the operating system, application programs required for at least one function (such as sound playback function, image playback function, etc.), etc.; the data storage area may store data created according to the use of the electronic device, etc. In addition, the memory 302 may include high-speed random access memory, and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other volatile solid-state storage device. Accordingly, the memory 302 may also include a memory controller to provide the processor 301 with access to the memory 302.

[0075] The electronic device also includes a power supply 303 that supplies power to various components. Preferably, the power supply 303 can be logically connected to the processor 301 through a power management system, thereby enabling functions such as charging, discharging, and power consumption management through the power management system. The power supply 303 may also include one or more DC or AC power supplies, recharging systems, power fault detection circuits, power converters or inverters, power status indicators, and other arbitrary components.

[0076] The electronic device may also include an input unit 304, which can be used to receive input digital or character information and generate keyboard, mouse, joystick, optical or trackball signal inputs related to user settings and function control.

[0077] Although not shown, the electronic device may also include a display unit, etc., which will not be described in detail here. Specifically, in this embodiment, the processor 301 in the electronic device loads the executable files corresponding to the processes of one or more applications into the memory 302 according to the following instructions, and the processor 301 runs the applications stored in the memory 302 to realize various functions, as follows: Obtain operational data of the target power grid; archive the operational data and analyze the archived operational data; construct a power grid state diagram structure based on the power grid equipment status and data analysis results of the target power grid; and identify anomalies in the target power grid according to the power grid state diagram structure.

[0078] For details on the implementation of each of the above operations, please refer to the previous examples, which will not be repeated here.

[0079] This application embodiment, after acquiring the operational data of the target power grid, archives the operational data and analyzes the archived operational data. Then, based on the power grid equipment status and data analysis results of the target power grid, a power grid state graph structure is constructed. Finally, anomaly identification of the target power grid is performed based on the power grid state graph structure. In the power grid anomaly identification scheme provided in this application, topological associations are established during data archiving, overcoming the pain points of traditional storage lacking topological associations and difficulty in fault reproduction. The constructed power grid state graph integrates equipment status with global topology and electrical association features, compensating for the shortcomings of traditional methods that ignore graph structures. Anomalies are identified multi-dimensionally based on the graph structure, avoiding false alarms and missed alarms due to single-node threshold judgments, ultimately improving the timeliness and accuracy of anomaly identification, thereby solving the problems of poor timeliness and low accuracy in current power grid anomaly identification schemes.

[0080] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be performed by instructions, or by instructions controlling related hardware. These instructions can be stored in a computer-readable storage medium and loaded and executed by a processor.

[0081] Therefore, embodiments of this application provide a storage medium storing multiple instructions that can be loaded by a processor to execute steps in any of the power grid anomaly identification methods provided in this application. For example, the instructions can execute the following steps: Obtain operational data of the target power grid; archive the operational data and analyze the archived operational data; construct a power grid state diagram structure based on the power grid equipment status and data analysis results of the target power grid; and identify anomalies in the target power grid according to the power grid state diagram structure.

[0082] For details on the implementation of each of the above operations, please refer to the previous examples, which will not be repeated here.

[0083] The storage medium may include: read-only memory (ROM), random access memory (RAM), disk or optical disk, etc.

[0084] Since the instructions stored in the storage medium can execute the steps in any of the power grid anomaly identification methods provided in the embodiments of this application, the beneficial effects that any of the power grid anomaly identification methods provided in the embodiments of this application can achieve can be realized. For details, please refer to the previous embodiments, which will not be repeated here.

[0085] The above provides a detailed description of a power grid anomaly identification method and related equipment provided in the embodiments of this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The description of the above embodiments is only for the purpose of helping to understand the method and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for identifying power grid anomalies, characterized in that, include: Acquire operational data of the target power grid; The operational data is archived, and the archived operational data is analyzed. Based on the power grid equipment status and data analysis results of the target power grid, a power grid status diagram structure is constructed; Anomalies in the target power grid are identified based on the power grid state diagram structure.

2. The power grid anomaly identification method according to claim 1, characterized in that, The step of identifying anomalies in the target power grid based on the power grid state diagram structure includes: Extract the real-time operating characteristics of each node from the power grid state diagram structure; Based on the aforementioned real-time operating characteristics, candidate nodes are determined from multiple nodes; The candidate nodes are identified as abnormal based on a pre-defined database of abnormal cases.

3. The power grid anomaly identification method according to claim 2, characterized in that, The step of identifying anomalies in the candidate nodes based on a preset anomaly case library includes: Obtain the feature deviation pattern of the candidate node; Based on the feature deviation pattern and the pre-fault characteristics of abnormal cases in the preset abnormal case library, the candidate nodes are identified as abnormal.

4. The power grid anomaly identification method according to claim 3, characterized in that, The step of identifying anomalies in the candidate nodes based on the feature deviation pattern and the pre-fault characteristics of anomaly cases in a preset anomaly case library includes: Determine the deviation parameters corresponding to the feature deviation pattern; The determined deviation parameters are compared with the pre-fault characteristics of abnormal cases in the preset abnormal case library in multiple dimensions to identify the abnormal state of the candidate node.

5. The power grid anomaly identification method according to claim 1, characterized in that, The construction of the power grid state diagram structure based on the power grid equipment status and data analysis results of the target power grid includes: Obtain the basic graph structure; Map the state of the power grid equipment in the target power grid to the node attributes of the basic graph structure; The data analysis results are mapped to the node dynamic attributes and edge dynamic attributes of the basic graph structure; Based on the mapped graph structure, the power grid state graph structure of the target power grid is constructed.

6. The power grid anomaly identification method according to claim 5, characterized in that, The construction of the power grid state graph structure of the target power grid based on the mapped graph structure includes: Integrate the node and edge attributes of the mapped graph structure; The integrated graph structure is validated. When the integrated graph structure passes the verification, the integrated graph structure is determined as the power grid state graph structure of the target power grid.

7. The power grid anomaly identification method according to claim 1, characterized in that, The process of archiving the operational data and analyzing the archived operational data includes: The running data is archived according to the preset storage structure; Extract data characteristics from the archived runtime data; The archived operational data is analyzed based on the aforementioned data characteristics.

8. A power grid anomaly identification device, characterized in that, include: The acquisition module is used to acquire the operating data of the target power grid; The analysis module is used to archive the operational data and analyze the archived operational data. The construction module is used to construct a power grid state diagram structure based on the power grid equipment status and data analysis results of the target power grid; The identification module is used to identify anomalies in the target power grid based on the power grid state diagram structure.

9. An electronic device, characterized in that, include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the steps of the power grid anomaly identification method as described in any one of claims 1-7.

10. A storage medium, characterized in that, The computer program is stored and can be loaded by a processor and executed as described in any one of claims 1-7.

Images