A power grid fault plan generation method and system based on a knowledge graph and a storage medium

Abstract

The application discloses a power grid fault plan generation method and system based on a knowledge graph, and a storage medium, simulates a power transformation accident expectation, establishes a power grid accident plan generation knowledge graph database on the basis of simulating the power transformation accident expectation, and automatically analyzes a power loss range; constructs the knowledge graph database, constructs a power grid accident knowledge graph database according to a power grid accident plan typical template, quickly obtains relevant knowledge by calling an association relationship constructed by the knowledge graph; simulates power grid load transfer after an accident, adopts a path reasoning algorithm based on the knowledge graph to perform path reasoning and optimization according to the knowledge graph database, and automatically generates a power grid load transfer path; automatically generates a power grid accident plan, adopts an inference algorithm based on the knowledge graph to perform inference and optimization according to the power grid accident knowledge graph and event information, automatically generates the power grid accident plan, provides the automatically generated power grid accident plan to relevant dispatching and control personnel, and performs corresponding emergency handling operations according to plan steps.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power grid fault handling, in particular to a power grid fault plan generation method and system based on a knowledge graph and a storage medium. BACKGROUND

[0002] Compiling an accident plan is the core business process of a power grid company. Current plans are mainly formed by manual operation and records, and there is a lack of a platform for information aggregation. Most power grid knowledge is stored in different systems, files, or even in individuals' minds. It is difficult to dynamically maintain rule-based knowledge for accident handling, and it cannot be accurately matched when applied, which cannot adapt to the complex operating environment of the power grid. The information of the power grid is highly specialized, the rules are complex, and the understanding of knowledge varies from person to person, which lacks induction and summary, affecting the safety of power grid handling. The value of power grid data is lost, and there is a lack of technical support for knowledge mining and application, which cannot meet the needs of rapid response of the power grid. Therefore, it is urgent to promote the intelligent generation of power grid fault plans based on a knowledge graph, to centrally manage and apply power grid data, and to promote the transformation of the power grid from experience-based to intelligent. SUMMARY

[0003] The purpose of the present application is to provide a power grid fault plan generation method and system based on a knowledge graph, which can improve the reliability of the power grid, help staff identify potential fault points, and effectively and quickly handle fault types.

[0004] Technical solution: In order to achieve the above-mentioned purpose of the application, a power grid fault plan generation method based on a knowledge graph according to the present application comprises the following steps:

[0005] Simulate substation accident anticipation, including simulating substation full-station power outage anticipation, substation high-voltage side single-mother power outage anticipation, and substation single-main transformer power outage anticipation. On the basis of simulating substation accident anticipation, establish a power grid accident plan generation knowledge graph database, and automatically analyze power outage ranges;

[0006] Construct a knowledge graph database. According to a power grid accident plan typical template, construct a power grid accident knowledge graph database. The knowledge graph database includes a descriptive basic knowledge graph database, a descriptive conventional knowledge graph database, and a rule-based knowledge graph database. The descriptive basic knowledge graph database, the descriptive conventional knowledge graph database, and the rule-based knowledge graph database are knowledge fused. Through the correlation relationship constructed by calling the knowledge graph, relevant knowledge is quickly obtained, and information support is provided for intelligent plan generation;

[0007] Simulate power grid load transfer after an accident. According to the knowledge graph database, a path reasoning algorithm based on the knowledge graph is used for path reasoning and optimization, and a power grid load transfer path is automatically generated;

[0008] The power grid accident plan is automatically generated according to the power grid accident knowledge graph and event information, and a reasoning algorithm based on the knowledge graph is used for inference and optimization to automatically generate the power grid accident plan. The automatically generated power grid accident plan is provided to the relevant control personnel, including plan prompts and operation guidance, and the control personnel performs corresponding emergency handling operations according to the plan steps.

[0009] Further, the method for constructing the power grid accident plan typical template comprises,

[0010] The typical accidents of the power grid equipment failure assumption, pre-accident mode analysis, post-protection action mode analysis and load transfer strategy pushing are sorted and classified, and the corresponding power grid accident plan typical template is established.

[0011] Further, the model of the power grid accident knowledge graph library adopts a bottom layer data model constructed by a triple technology, the triple technology is abstracted into the form of entities, attributes and relationships, and is used for storing and querying power grid fault plans and other related information.

[0012] More specifically, the entity represents an individual that objectively exists, the attribute represents the relationship between entities, and the relationship represents the connection or interaction between entities.

[0013] Further, the specific method for constructing the descriptive basic knowledge graph library comprises,

[0014] A power grid container model is constructed to divide the power grid into substations, power plants and lines;

[0015] A power grid equipment model is constructed, including busbars, main transformers, main transformer windings, switches, circuit breakers, line segments and line segment endpoints;

[0016] A power grid topology model is constructed, including topology nodes and topology relationships;

[0017] A power grid operation mode model is constructed, including switch states and loads under multiple time states.

[0018] Further, the specific method for constructing the descriptive general knowledge graph library comprises,

[0019] Important user information, risk level information, special transfer path information, equipment limit information and various data sources of regulatory system files are collected to provide basic data for knowledge graph construction;

[0020] Natural language processing and deep learning artificial intelligence technologies are used for semantic analysis to convert text into structured data, and entities, attributes and mutual relationships between entities are extracted from the text;

[0021] According to the identification information, the concepts of entities, attributes, and logical relationships between entities and attributes are abstracted, and the knowledge representation modeling of the ontology is performed;

[0022] Based on the collected data, the construction of the knowledge graph is carried out, and the entities, attributes and relationships are stored in the form of triples in the graph database, and an index is established to support efficient query and reasoning;

[0023] Continuously update and maintain the knowledge graph library to ensure the accuracy and integrity of the information to meet the needs of different fields and provide strong support for decision-making.

[0024] Further, the specific method for constructing the rule-based knowledge graph library includes,

[0025] Determine the target and application scenario of the knowledge graph, and clearly define the entities, attributes and relationships to be covered;

[0026] Collect and process related data, including data cleaning, deduplication and standardization steps;

[0027] Based on the ontology model, define the entity classes, attributes and relationships in the knowledge graph, and establish the hierarchical structure and logical relationships between elements;

[0028] Formulate power loss impact analysis rules, identify potential fault points based on power grid parameters and equipment state data, and predict the power loss impact range and duration information;

[0029] Formulate load transfer path rules, determine the feasibility and optimal scheme of load transfer according to the voltage and current parameters between each node in the power system;

[0030] Combine the formulated rules with the ontology model, and represent the knowledge as rules to support rule-based reasoning application scenarios;

[0031] Verify the correctness and effectiveness of the rules, adjust and optimize the rule library, and improve the accuracy and reliability of the knowledge graph.

[0032] Further, the descriptive basic knowledge graph library, the descriptive general knowledge graph library and the rule-based knowledge graph library are fused, including the following steps,

[0033] Use entity fusion technology to unify the identification of related entities in the constructed descriptive basic knowledge graph library, descriptive general knowledge graph library and rule-based knowledge graph library, and associate them;

[0034] According to the characteristics and attributes of the power grid fault knowledge elements, the related entities in the constructed descriptive basic knowledge graph library, descriptive regular knowledge graph library and rule knowledge graph library are normalized according to each classification to form standardized power grid fault knowledge elements;

[0035] The triple technology is used to represent various entities, attributes and relationships in the power grid fault plan graph in the power grid accident knowledge graph library.

[0036] Further, the power grid load transfer path transfer method further comprises,

[0037] The method of load transfer knowledge graph key node path analysis is adopted to identify the key nodes affecting load transfer in the power system, and according to the path characteristics and electrical parameter information between nodes, a load transfer strategy is formulated, and the load transfer strategy takes "not generating new over-limit" as the basic principle;

[0038] When performing load transfer, the method of load transfer key node path determination is adopted, by comparing the transferred load value with the sum of the current load value of the equipment to be transferred in, it is judged whether it is lower than the load limit value of the equipment to be transferred in. Only when this condition is met, load transfer can be performed, otherwise the load transfer strategy needs to be re-formulated or the operation state of the power system needs to be adjusted.

[0039] Further, it comprises a simulation substation accident module, a knowledge graph library generation module, a power grid load transfer path generation module and a power grid accident plan automatic generation module,

[0040] The simulation substation accident module is used to simulate substation full station power outage anticipation, substation high voltage side single mother power outage anticipation and substation single main transformer power outage anticipation. On the basis of simulating substation accident anticipation, the power grid accident plan generation knowledge graph library is established, and the power outage range is automatically analyzed.

[0041] The knowledge graph library generation module is used to construct the power grid accident knowledge graph library according to the power grid accident plan typical template, and the knowledge graph library comprises a descriptive basic knowledge graph library, a descriptive regular knowledge graph library and a rule knowledge graph library. The descriptive basic knowledge graph library, the descriptive regular knowledge graph library and the rule knowledge graph library are knowledge fused, the related knowledge is quickly acquired through calling the associated relationship constructed by the knowledge graph, and information support is provided for intelligent plan generation;

[0042] The power grid load transfer path generation module is used to generate the power grid load transfer path automatically according to the knowledge graph library and by using the path reasoning algorithm based on the knowledge graph.

[0043] The automatic power grid accident contingency plan generation module is used to infer and optimize based on the power grid accident knowledge graph and event information, and automatically generate power grid accident contingency plans. The automatically generated power grid accident contingency plans are provided to relevant dispatching personnel, including plan prompts and operation guidance. The dispatching personnel perform corresponding emergency handling operations according to the plan steps.

[0044] Furthermore, the computer-readable storage medium stores a computer program that is executed by a processor to implement a knowledge graph-based method for generating power grid fault contingency plans.

[0045] Beneficial Effects: The advantage of this invention lies in utilizing knowledge graph technology to anticipate power grid accidents such as substation-wide power outages, single busbar outages on the high-voltage side of substations, and single main transformer outages. By simulating these substation accident scenarios, the scope of power loss can be automatically analyzed. These anticipations can help staff identify potential fault points and predict possible power outages in the power grid.

[0046] The advantage of this invention lies in its use of knowledge graph technology to simulate the operation of automatic safety devices in power grids and automatic safety devices in case of accidents, thereby simulating various faults and equipment damage scenarios that may occur in actual power systems. This allows for the assessment of the safety and stability of the power system and provides a reference for subsequent fault handling.

[0047] The advantage of this invention lies in its use of knowledge graph technology to automatically deduce changes in the power grid topology after a fault. Based on preset rules and conditions, it infers the new connection relationships and electrical parameters between various nodes in the power system after the fault. This can help staff quickly restore the operation of the power system and prevent the fault from spreading and affecting a wider area.

[0048] The advantage of this invention lies in the use of knowledge graph technology. The power grid load transfer method formulates a reasonable load transfer strategy by applying knowledge graphs and path analysis of key nodes in load transfer, and strictly controls the conditions and restrictions of the transfer process in actual operation, which can improve the reliability and stability of the power grid.

[0049] The advantages of this invention also lie in the knowledge graph-based power grid fault contingency plan generation method, which improves the efficiency and accuracy of fault contingency plan generation and realizes intelligent fault contingency plan preparation.

[0050] The advantages of this invention also lie in the knowledge graph-based method for generating power grid fault contingency plans. This method intelligently manages relevant information for contingency plans by constructing a descriptive basic knowledge base, a descriptive conventional knowledge base, and a rule-based knowledge base. Based on relevant policy requirements, it intelligently selects the corresponding handling process according to the fault type and intelligently generates contingency plan content and provides a visual display.

[0051] The advantages of this invention also lie in the knowledge graph-based power grid fault contingency plan generation method provided. It employs artificial intelligence technology, that is, it uses knowledge graph technology to associate and process complex information during the fault contingency plan preparation. Based on constraints such as equipment limits and power grid operation, it uses machine learning to continuously conduct rehearsals and exercises during the fault contingency plan preparation, gradually achieving full coverage of the fault contingency plan handling process. Combined with online platform management, it effectively improves the efficiency of fault contingency plan preparation and the level of control. Attached Figure Description

[0052] Figure 1 This is a schematic diagram of the knowledge graph-based power grid fault contingency plan generation method in Example 1;

[0053] Figure 2 This is a schematic diagram of the structure of the knowledge graph-based power grid fault contingency plan generation system in Example 2. Detailed Implementation

[0054] Example 1

[0055] like Figure 1 The method for generating power grid fault contingency plans based on knowledge graphs includes the following steps:

[0056] The simulation of substation accident scenarios includes scenarios of a complete substation blackout, a single busbar blackout on the high-voltage side of the substation, and a single main transformer blackout. Based on these simulations, a knowledge graph database for generating power grid accident contingency plans is established to automatically analyze the scope of power outages. These scenarios help staff identify potential fault points and predict possible power outages in the power grid. Specifically, the complete substation blackout scenario simulates a blackout caused by damage or malfunction of various equipment in the substation, and the plan can include the activation of the substation's backup power supply, investigation and repair of the blackout cause, etc. The single busbar blackout scenario simulates a blackout of a single busbar on a high-voltage line in the substation, and the plan can include line inspection and maintenance, busbar tripping protection measures, etc. The single main transformer blackout scenario simulates a blackout caused by the failure of a single main transformer in the substation, and the plan can include the activation of the backup main transformer, investigation and repair of the blackout cause, etc. Based on simulated substation accident scenarios, a power grid accident contingency plan generation strategy library can be established to automatically analyze the scope of power outages and promptly implement emergency measures to reduce the impact of accidents. Simultaneously, it is necessary to strengthen equipment maintenance and monitoring to improve the reliability and security of the power grid. Simulating the operation of automatic safety devices in the power grid and automatic safety devices during accidents simulates various faults and equipment damage scenarios that may occur in actual power systems. This allows for the assessment of the power system's safety and stability and provides a reference for subsequent fault handling. Automatically extrapolating post-accident power grid topology changes, based on preset rules and conditions, infers new connection relationships and electrical parameters between various nodes in the power system after the fault. This helps personnel quickly restore power system operation and prevent the fault from escalating and affecting a wider area.

[0057] A knowledge graph database is constructed based on typical templates for power grid accident contingency plans. This database includes a descriptive basic knowledge graph database, a descriptive general knowledge graph database, and a rule-based knowledge graph database. These databases are integrated, and relevant knowledge is quickly acquired by invoking the relationships built from the knowledge graphs, providing information support for the generation of intelligent contingency plans. The method for constructing typical templates for power grid accident contingency plans includes: sorting and classifying typical accidents in the power grid, such as hypothetical equipment failures, pre-accident mode analysis, post-protection action mode analysis, and load transfer strategy pushes, to establish corresponding typical templates for power grid accident contingency plans. For hypothetical equipment failures, based on industry needs, hypothetical equipment failures are input, and the system automatically matches the corresponding power grid operating status to find typical situations for equipment failure simulation. For pre-accident scenarios... Precision Analysis: This involves collecting data on the pre-accident grid operation mode, substation operation mode, substation load status, and related plant / substation load status, and statistically analyzing the operation within the relevant scope. Post-Protection Action Analysis: This identifies affected substations and lines, generates a post-accident grid wiring diagram based on load changes, identifies substations and lines requiring priority power restoration based on a list of important users in the affected area, implements power supply control for key substations and lines through grid topology, and provides specific control methods. Post-Accident Strategy Push: Through effective analysis of various data, this enables intelligent auxiliary decision support and regional self-healing control in various scenarios such as load transfer, accidental power outages, load exceeding limits, trial power restoration, and planned maintenance. Post-Accident Impact Analysis: After adjustments, the temporary operating state needs to be restored to normal operation, and the impact of the fault needs to be summarized, analyzing information such as lost load and risk level, and generating record reports. The grid accident knowledge graph database uses a triplet technique to construct the underlying data model. Triplet techniques abstract entities, attributes, and relationships for storing and querying grid fault contingency plans and other relevant information. An entity represents an objectively existing individual, an attribute represents the relationship between entities, and a relationship represents the connection or interaction between entities.

[0058] The specific methods for constructing a descriptive basic knowledge graph library include: constructing a power grid container model, dividing the power grid into containers such as substations, power plants, and lines, and defining the relationships and attributes between them; constructing a power grid equipment model, including various power grid equipment such as buses, main transformers, main transformer windings, switches, circuit breakers, line segments, and line segment endpoints, and defining the relationships and attributes between them; constructing a power grid topology model, including topology nodes and topology relationships, and associating it with the equipment model; and constructing a power grid operation mode model, including information such as switch states and loads under multiple time states, and associating it with the equipment and topology models.

[0059] The specific methods for constructing a descriptive general knowledge graph base include: collecting various data sources such as important user information, risk level information, special transfer path information, equipment quota information, and regulatory documents to provide basic data for knowledge graph construction; using natural language processing and deep learning artificial intelligence technologies for semantic analysis to transform text into structured data, and extracting entities, attributes, and relationships between entities; abstracting the concepts of entities, attributes, and logical relationships between entities based on the identified information to model the knowledge representation of the ontology; constructing the knowledge graph based on the collected data, storing entities, attributes, and relationships in the form of triples in the graph database, and establishing indexes to support efficient querying and reasoning; and continuously updating and maintaining the knowledge graph base to ensure the accuracy and completeness of its information to meet the needs of different domains and provide strong support for decision-making.

[0060] The specific methods for constructing a rule-based knowledge graph base include: determining the goals and application scenarios of the knowledge graph, and clarifying the entities, attributes, and relationships that need to be covered; collecting and processing relevant data, including data cleaning, deduplication, and standardization; defining entity classes, attributes, and relationships in the knowledge graph based on an ontology model, and establishing a hierarchical structure and logical relationships between elements; formulating rules for analyzing the impact of power outages, identifying potential fault points through rule-based reasoning based on grid parameters and equipment status data, and predicting the scope and duration of the power outage impact; formulating rules for load transfer paths, determining the feasibility and optimal solution for load transfer based on voltage and current parameters between nodes in the power system; combining the formulated rules with the ontology model to represent knowledge in rule form to support rule-based reasoning application scenarios; verifying the correctness and effectiveness of the knowledge graph, adjusting and optimizing the knowledge model and rule base to improve the accuracy and reliability of the knowledge graph. The algorithm for analyzing the impact of power outages is as follows:

[0061] In the knowledge graph, device connection relationships are set as g1(GA,GB), g2(GB,GC), ..., gn(Gn,G0). Assuming a power grid failure occurs in g1, the power-loss equipment and related loads are analyzed based on voltage levels. The algorithm for analyzing the impact of power-loss equipment is as follows:

[0062] Sn(g,v) = g1->n;

[0063] P = P(Sn);

[0064] Where V1, V2, ..., Vn are the equipment voltage levels, and P(Sn) is the load value of the relevant de-energized line.

[0065] The analysis considers various scenarios of load transfer within and outside the station. However, for main transformer or single bus operation, load transfer within the station is prioritized. If load transfer within the station is not feasible, load transfer outside the station is considered. In the event of a complete station power outage, only load transfer outside the station can be considered. If a portion of the line load is not supported by the grid structure for external load transfer, it is considered as de-energized equipment and will subsequently serve as the data source for calculating the impact on important users and load loss.

[0066] In generating power grid emergency response plans, if specific equipment failure scenarios need to be considered, basic information about the equipment, its operating principles, failure modes, and handling methods can be obtained through descriptive basic knowledge graphs and descriptive general knowledge graphs. Simultaneously, specific rules can be obtained from rule-based knowledge graphs to determine when and what measures should be taken to address the failure. By integrating these different types of knowledge, smart contingency plans can be generated more accurately, improving emergency response efficiency.

[0067] The knowledge fusion of descriptive basic knowledge graphs, descriptive general knowledge graphs, and rule-based knowledge graphs includes the following steps: First, entity fusion technology is used to uniformly identify and associate relevant entities in the constructed descriptive basic knowledge graph, descriptive general knowledge graph, and rule-based knowledge graph. Second, based on the characteristics and attributes of power grid fault knowledge elements, the relevant entities in the constructed descriptive basic knowledge graph, descriptive general knowledge graph, and rule-based knowledge graph are classified and normalized according to each classification to form standardized power grid fault knowledge elements. Third, triplet technology is used to represent various entities, attributes, and relationships in the power grid fault contingency plan graph in the power grid accident knowledge graph.

[0068] The simulation of power grid load transfer after an accident involves using a knowledge graph-based path reasoning algorithm for path inference and optimization, automatically generating power grid load transfer paths. The transfer method also includes using a key node path analysis method based on the load transfer knowledge graph to identify key nodes in the power system affecting load transfer. Based on the path characteristics and electrical parameter information between each node, a load transfer strategy is formulated, with the basic principle of "not generating new limit violations." During load transfer, a key node path determination method is used. By comparing the sum of the transferred load value and the current load value of the equipment to be transferred into, it is determined whether it is lower than the load limit of the equipment to be transferred into. Load transfer can only proceed if this condition is met; otherwise, a new load transfer strategy needs to be formulated or the operating status of the power system needs to be adjusted. The specific analysis algorithm for power grid load transfer paths is as follows:

[0069]

[0070] In the formula: PX is the load to be transferred, PK is the active power of the transformer in the substation to which the load is transferred, F(PK,PX) is the calculated value of the transformer load in the substation after the load is transferred, and n is the number of nodes in the final path; the load is transferred at node n.

[0071] If the load transfer within the station fails to meet the limit conditions, then the load transfer outside the station will be carried out. The specific formula is as follows:

[0072]

[0073] Where limP K For the transformer limit within the station, when P A1 P A2 P A... P An The loads are transferred in ascending order of size, with load A being transferred. m The condition must be met for the main transformer load F(P) after the transfer. Am ,P K ) less than limP K Transformer quota.

[0074] If F(P) An ,P Kn If the limit is exceeded, off-site load transfer will be carried out. The off-site load transfer conditions will also simultaneously meet the power grid limit conditions. The specific formula is as follows:

[0075]

[0076] Where limP Kn For the transformer quota of substations planned for load transfer, P An The loads to be transferred are arranged in descending order, and the load of the main transformer after the transfer must be satisfied. An ,P Kn ) less than limP Kn If the transformer capacity does not meet the load transfer conditions, it cannot transfer power.

[0077] After analyzing the power grid structure, data on transferable power grid lines and related loads are obtained. Based on the power grid topology analysis, transfer paths are identified. The main transformer load and line load within these transfer paths are obtained according to pre-set conditions. If the sum of these loads does not exceed equipment limits, the transferable path is deemed feasible. Following the transferable path analysis, it is necessary to determine if multiple power sources within the relevant substations can provide transfer services. This requires identifying special transfer path sequences and non-transferable paths. Through these steps, knowledge graphs and graph reasoning technologies can be used to automate and intelligently transfer power grid loads, improving the power grid's disaster resilience and emergency response efficiency.

[0078] The system automatically generates power grid accident contingency plans. Based on the power grid accident knowledge graph and event information, it uses a knowledge graph-based reasoning algorithm for inference and optimization to automatically generate power grid accident contingency plans. The automatically generated power grid accident contingency plans are then provided to relevant dispatching personnel, including plan prompts and operation instructions. Dispatching personnel then follow the plan steps to perform corresponding emergency response operations.

[0079] Example 2

[0080] like Figure 2 As shown, a knowledge graph-based power grid fault contingency plan generation system includes a substation accident simulation module, a knowledge graph database generation module, a power grid load transfer path generation module, and an automatic power grid fault contingency plan generation module.

[0081] The simulated substation accident module is used to simulate the anticipation of a substation-wide power outage, a single busbar outage on the high-voltage side of the substation, and a single main transformer outage. Based on the simulated substation accident scenarios, a knowledge graph database for generating power grid accident contingency plans is established to automatically analyze the scope of power outages.

[0082] The knowledge graph generation module is used to construct a power grid accident knowledge graph based on typical templates of power grid accident contingency plans. The knowledge graph includes a descriptive basic knowledge graph, a descriptive regular knowledge graph, and a rule-based knowledge graph. The module integrates the knowledge from the descriptive basic knowledge graph, the descriptive regular knowledge graph, and the rule-based knowledge graph, and quickly obtains relevant knowledge by calling the relationships constructed by the knowledge graph, thus providing information support for the generation of intelligent contingency plans.

[0083] The power grid load transfer path generation module is used to automatically generate power grid load transfer paths by using a knowledge graph-based path reasoning algorithm to perform path reasoning and optimization based on a knowledge graph database.

[0084] The automatic power grid accident contingency plan generation module is used to infer and optimize based on the power grid accident knowledge graph and event information, and automatically generate power grid accident contingency plans. The automatically generated power grid accident contingency plans are provided to relevant dispatching personnel, including plan prompts and operation instructions. Dispatching personnel perform corresponding emergency response operations according to the plan steps.

[0085] In particular, according to embodiments disclosed in this invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this disclosure include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication component, and / or installed from a removable medium. When the computer program is executed by a central processing unit (CPU), it performs the functions defined in the methods of this application. It should be noted that the computer-readable medium described above in this application can be a computer-readable signal medium or a computer-readable storage medium, or any combination of the two. A computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wire segments, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this application, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in connection with an instruction execution system, apparatus, or device. In this application, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can also be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on a computer-readable medium may be transmitted using any suitable medium, including but not limited to: wireless segments, wire segments, optical fibers, RF, etc., or any suitable combination thereof.

[0086] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0087] The above are merely preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for generating power grid fault contingency plans based on knowledge graphs, characterized in that, Includes the following steps: Simulated substation accident scenarios include simulated substation-wide power outage scenarios, simulated single busbar power outage scenarios on the high-voltage side of the substation, and simulated single main transformer power outage scenarios. Based on the simulated accident scenarios, a knowledge graph database for generating power grid accident contingency plans is established to automatically analyze the scope of power outages. A knowledge graph database is constructed based on typical templates of power grid accident contingency plans. The knowledge graph database includes a descriptive basic knowledge graph database, a descriptive conventional knowledge graph database, and a rule-based knowledge graph database. The descriptive basic knowledge graph database, the descriptive conventional knowledge graph database, and the rule-based knowledge graph database are fused together. By calling the association relationships constructed by the knowledge graph, relevant knowledge can be quickly obtained, and information support can be provided for the generation of intelligent contingency plans. The system simulates the power grid load transfer after an accident. Based on a knowledge graph database, a knowledge graph-based path reasoning algorithm is used to perform path reasoning and optimization, and automatically generate power grid load transfer paths. Automatically generate power grid accident contingency plans. Based on the power grid accident knowledge graph and event information, use knowledge graph-based reasoning algorithms to infer and optimize, automatically generate power grid accident contingency plans, and provide the automatically generated power grid accident contingency plans to relevant dispatching personnel, including plan prompts and operation guidance. Dispatching personnel can then perform corresponding emergency response operations according to the plan steps. The knowledge fusion of the descriptive basic knowledge graph library, the descriptive general knowledge graph library, and the rule-based knowledge graph library includes the following steps: Entity fusion technology is used to uniformly identify and associate related entities in the constructed descriptive basic knowledge graph library, descriptive regular knowledge graph library, and rule-based knowledge graph library; Based on the characteristics and attributes of power grid fault knowledge elements, the relevant entities in the constructed descriptive basic knowledge graph library, descriptive regular knowledge graph library, and rule-based knowledge graph library are normalized according to each category to form standardized power grid fault knowledge elements. The triplet technique is used to represent various entities, attributes and relationships in the power grid fault contingency plan graph in the power grid accident knowledge graph database.

2. The method for generating power grid fault contingency plans based on knowledge graphs according to claim 1, characterized in that, The method for constructing the typical template of the power grid accident contingency plan includes, We will sort out and classify typical accidents in the power grid, including hypothetical equipment failures, pre-accident mode analysis, post-protection action mode analysis, and load transfer strategy push, and establish corresponding typical templates for power grid accident contingency plans.

3. The method for generating power grid fault contingency plans based on knowledge graphs according to claim 1, characterized in that, The model of the power grid accident knowledge graph database adopts the underlying data model constructed by triplet technology. The triplet technology is abstracted into the form of entities, attributes and relations, which are used to store and query power grid fault contingency plans and other related information.

4. The method for generating power grid fault contingency plans based on knowledge graphs according to claim 1, characterized in that, The specific methods for constructing the descriptive basic knowledge graph library include: Construct a power grid container model, dividing the power grid into substations, power plants, and lines; Construct a model of power grid equipment, including busbars, main transformers, main transformer windings, switches, circuit breakers, line segments, and line segment endpoints; Construct a power grid topology model, including topology nodes and topology relationships; Construct a power grid operation model, including switching states and loads at multiple time points.

5. The method for generating power grid fault contingency plans based on knowledge graphs according to claim 1, characterized in that, The specific methods for constructing the descriptive general knowledge graph library include: Collects multiple data sources, including important user information, risk level information, special transfer path information, equipment quota information, and regulatory documents; Semantic analysis using natural language processing and deep learning artificial intelligence techniques transforms text into structured data, and extracts entities, attributes, and relationships between entities from it. Based on the identification information, the concepts of entities, attributes, and logical relationships between entities and attributes are abstracted to perform knowledge representation modeling of the ontology. Knowledge graphs are constructed based on the collected data, and entities, attributes, and relations are stored in the graph database in the form of triples. Continuously update and maintain the knowledge graph database.

6. The method for generating power grid fault contingency plans based on knowledge graphs according to claim 1, characterized in that, Specific methods for constructing rule-based knowledge graph bases include: Define the goals and application scenarios of the knowledge graph, and clarify the entities, attributes, and relationships that need to be covered; Collect and process relevant data, including data cleaning, deduplication, and standardization steps; Based on the ontology model, we define entity classes, attributes and relationships in the knowledge graph, and establish the hierarchical structure and logical relationships between elements; Develop rules for analyzing the impact of power outages, identify potential fault points through rule-based reasoning based on grid parameters and equipment status data, and predict the scope and duration of the power outage impact. Develop load transfer path rules and determine the feasibility and optimal solution for load transfer based on the voltage and current parameters between various nodes in the power system. By combining the established rules with the ontology model, knowledge is represented in rule form to support rule-based reasoning application scenarios.

7. The method for generating power grid fault contingency plans based on knowledge graphs according to claim 1, characterized in that, The power grid load transfer method also includes, By employing the critical node path analysis method of load transfer knowledge graph, the key nodes affecting load transfer in the power system are identified, and load transfer strategies are formulated based on the path characteristics and electrical parameter information between each node. The load transfer strategies do not generate new limits. When performing load transfer, the method of determining the critical node path of load transfer is adopted. By comparing the sum of the load value being transferred out and the current load value of the equipment to be transferred in, it is determined whether it is lower than the load limit of the equipment to be transferred in. Load transfer can only be carried out when this condition is met; otherwise, the load transfer strategy needs to be re-formulated or the operating status of the power system needs to be adjusted.

8. A power grid fault contingency plan generation system based on knowledge graphs, characterized in that, A method for generating power grid fault contingency plans based on a knowledge graph, as described in any one of claims 1-7, includes a substation accident simulation module, a knowledge graph database generation module, a power grid load transfer path generation module, and an automatic power grid fault contingency plan generation module. The simulated substation accident module is used to simulate the anticipation of a substation-wide power outage, a single busbar outage on the high-voltage side of the substation, and a single main transformer outage. Based on the simulated substation accident scenarios, a knowledge graph database for generating power grid accident contingency plans is established to automatically analyze the scope of power outages. The knowledge graph generation module is used to construct a power grid accident knowledge graph based on typical templates of power grid accident contingency plans. The knowledge graph includes a descriptive basic knowledge graph, a descriptive conventional knowledge graph, and a rule-based knowledge graph. The module integrates the knowledge from the descriptive basic knowledge graph, the descriptive conventional knowledge graph, and the rule-based knowledge graph. By calling the relationships constructed from the knowledge graph, the module can quickly obtain relevant knowledge and provide information support for the generation of intelligent contingency plans. The power grid load transfer path generation module is used to automatically generate power grid load transfer paths by using a knowledge graph-based path reasoning algorithm to perform path reasoning and optimization based on a knowledge graph database. The automatic power grid accident contingency plan generation module is used to infer and optimize based on the power grid accident knowledge graph and event information, and automatically generate power grid accident contingency plans. The automatically generated power grid accident contingency plans are provided to relevant control personnel, including contingency plan prompts and operation guidance. Control personnel perform corresponding emergency handling operations according to the steps in the contingency plan. The knowledge fusion of the descriptive basic knowledge graph library, the descriptive general knowledge graph library, and the rule-based knowledge graph library includes the following steps: Entity fusion technology is used to uniformly identify and associate related entities in the constructed descriptive basic knowledge graph library, descriptive regular knowledge graph library, and rule-based knowledge graph library; Based on the characteristics and attributes of power grid fault knowledge elements, the relevant entities in the constructed descriptive basic knowledge graph library, descriptive regular knowledge graph library, and rule-based knowledge graph library are normalized according to each category to form standardized power grid fault knowledge elements. The triplet technique is used to represent various entities, attributes and relationships in the power grid fault contingency plan graph in the power grid accident knowledge graph database.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that is executed by a processor to implement a knowledge graph-based power grid fault contingency plan generation method as described in any one of claims 1-7.

Images