A knowledge graph-based fault tracing method, system, device and medium for a switch

Abstract

The application relates to the technical field of smart grid equipment fault analysis, and discloses a switch failure tracing method, system, device and medium based on a knowledge graph, which comprises the following steps: receiving switch equipment data; establishing a switch failure knowledge graph ontology; converting the received data into triples and writing the triples into a graph storage space to form a switch failure knowledge graph containing internal state evolution relations and external electrical connection relations; taking a specified node as the starting point, performing reverse path traversal in the knowledge graph in sequence, and limiting candidate paths; determining the nodes corresponding to the switch failure from the limited candidate tracing paths, and outputting the fault correlation objects and the corresponding tracing paths. The application can be used for the associated expression and semantic deduction of multi-source heterogeneous data, improves the accuracy and completeness of switch failure analysis, improves the fault positioning efficiency, and enhances the coherence and interpretability of the failure analysis results.

Description

Technical Field

[0001] This invention relates to the field of fault analysis technology for smart grid equipment, and in particular to a method, system, equipment, and medium for tracing faults in disconnectors based on knowledge graphs. Background Technology

[0002] As a crucial component of primary equipment in substations, disconnectors perform functions such as electrical connection switching, maintenance isolation, and operation mode adjustment. Their operational status directly impacts the safe and stable operation of the power system. During operation, disconnectors are susceptible to faults such as incomplete opening and closing, abnormal status indication, loss of synchronization during operation, and abnormal contact due to factors such as mechanical transmission wear, abnormal contact of contacts, inconsistent feedback from auxiliary contacts, inadequate operation execution, and changes in the external electrical connection environment. These types of faults typically involve changes in the state of internal components of the disconnector and changes in external electrical connections, exhibiting characteristics of multi-source information coupling, long fault chains, and complex correlations.

[0003] In existing technologies, the analysis of disconnector faults typically relies on a comprehensive judgment based on equipment ledgers, operation ticket records, alarm information, online monitoring data, and the experience of operating personnel. This approach suffers from several drawbacks. First, it requires manual processing and comparison of multi-source heterogeneous data, resulting in low analytical efficiency. Second, the lack of a unified data association model makes it difficult to integrate the internal state evolution of the disconnector with its external electrical connections, leading to inaccurate fault location and a lack of coherence and interpretability in the tracing process.

[0004] Furthermore, in existing fault analysis solutions, data from different sources is typically stored in scattered locations across ledger systems, monitoring systems, operation ticket systems, and topology model systems, lacking a unified semantic organization method. Especially when a disconnector malfunctions, current technology lacks a unified solution for modeling the internal state chain and external topology connection chain of the disconnector and performing reverse semantic derivation, addressing the challenge of quickly tracing back fault-related objects from multi-dimensional information such as alarm time, operation process, state changes, bus segmentation, and electrical connection topology. Summary of the Invention

[0005] In view of the aforementioned existing problems, the present invention provides a method, system, device and medium for tracing the fault source of disconnect switches based on knowledge graphs.

[0006] Therefore, the technical problem solved by the present invention is that the existing technology suffers from insufficient multi-source data association, incomplete expression of fault chains, low source tracing efficiency, and low fault location accuracy.

[0007] To solve the above technical problems, the present invention provides the following technical solution: a method for tracing the fault source of a disconnect switch based on a knowledge graph, comprising: receiving disconnect switch equipment data and establishing a disconnect switch fault knowledge graph ontology, wherein the disconnect switch fault knowledge graph ontology includes entity types and relation types; The data received by the body is converted into triples and written into the graph storage space to form a knowledge graph of switch faults that includes internal state evolution relationships and external electrical connection relationships; Starting from a specified node, a reverse path traversal is performed in the knowledge graph according to the order of operation ticket steps and alarm timestamps, and candidate paths are limited based on bus segmentation relationships and electrical connection topology. Identify the node corresponding to the switch fault from the limited candidate tracing paths, and output the fault-related object and the corresponding tracing path.

[0008] As a preferred embodiment of the knowledge graph-based method for tracing faults of disconnectors described in this invention, the establishment of the disconnector fault knowledge graph ontology includes: establishing a structural relationship between the entities of the disconnector ontology and the entities of the contact components, the entities of the transmission mechanism, and the entities of the auxiliary contacts, based on the component composition information in the disconnector equipment ledger data. Based on the busbar segmentation relationship data and electrical connection topology data, establish the attribution relationship between the entity of the disconnect switch body and the entity of the busbar segment, and establish the connection relationship between the entity of the disconnect switch body and the entity of adjacent electrical equipment. Based on the correspondence between operation ticket data and contact status data, a correspondence is established between the entities of operation events and the entities of status nodes; based on the association between alarm timestamp data and contact status and auxiliary contact position signals, a correspondence is established between the entities of abnormal events and contact status and auxiliary contact position signals.

[0009] As a preferred embodiment of the knowledge graph-based fault tracing method for disconnectors described in this invention, the step of converting the received data into triples by the entity includes: writing the positional state, contact state, and action stage information of the contacts of phases A, B, and C in the contact state quantity data into the attribute fields of the corresponding phase contact component entities; if the contact component entities are not distinguished separately, they are written into the attribute fields of the disconnector entity; and at the same time, generating corresponding state node entities based on the contact state quantity data of each phase at each sampling time. Write the step number, step content, and execution time corresponding to each step in the operation ticket data into the attribute field of the operation event entity; write the occurrence time of each abnormal event in the alarm timestamp data into the attribute field of the abnormal event entity.

[0010] As a preferred embodiment of the knowledge graph-based switch fault tracing method described in this invention, after generating operation event entities and status node entities, a sequential relationship representing the order of execution is established between operation event entities according to the step number in the operation ticket data from smallest to largest or the execution time from first to last, and the operation event entities are connected sequentially to form an operation event sequence. According to the temporal sequence of changes in each state quantity in the contact state quantity data, a state evolution relationship representing the direction of state transition is established between adjacent state node entities, and each state node entity is connected sequentially to form a state node sequence; and according to the correspondence, the operation event entity in the operation event sequence is associated with the corresponding state node entity in the state node sequence to form a relationship network for the collaborative expression of operation steps and state changes.

[0011] As a preferred embodiment of the knowledge graph-based disconnector fault tracing method described in this invention, the formation of a disconnector fault knowledge graph containing internal state evolution relationships and external electrical connection relationships includes: establishing topological connection relationships between the disconnector body entity, the bus segment entity, and adjacent electrical equipment entities based on bus segment relationship data and electrical connection topology data. The sequence of state nodes and state evolution relationships formed by the operation ticket order and the changes in contact state quantities will be linked to the corresponding bus section entities and the electrical connection paths composed of the disconnector body, adjacent electrical equipment entities and topological connection relationships through the correspondence between the disconnector body entity and the bus section entity, thus forming a knowledge graph structure that unifies the internal state evolution and external electrical connection of the disconnector.

[0012] As a preferred embodiment of the knowledge graph-based disconnector fault tracing method described in this invention, the limited candidate path includes: the specified node being the entity node corresponding to the abnormal event and / or the state node corresponding to the abnormal contact state; Based on the chronological relationship between the alarm timestamp and the execution time of the operation event, operation event entities whose execution time is before the occurrence time of the abnormal event are selected as candidate operation events; starting from the state node corresponding to the candidate operation event, backtracking to the previous state node in the reverse direction of the state evolution relationship; During the backtracking process, based on the bus segment relationship data and electrical connection topology data, only candidate tracing paths that are located within the bus segment entity associated with the abnormal event or have a direct electrical connection with the disconnector body entity corresponding to the abnormal event are retained.

[0013] As a preferred embodiment of the knowledge graph-based method for tracing the fault of a disconnect switch as described in this invention, the step of determining the node corresponding to the disconnect switch fault from the limited candidate tracing paths includes: performing sequential consistency checks of operation steps, continuity checks of state evolution, and continuity checks of electrical connections on each candidate tracing path. The operation step sequence consistency check determines whether the sequential relationship between operation event entities in the path is consistent with the step number sequence of the operation ticket. The state evolution continuity check determines whether the state evolution relationship between adjacent state node entities in the path is continuous and without any breaks; The electrical connection continuity check determines whether the topological connection relationship between the disconnector body entity, busbar segment entity and adjacent electrical equipment entity involved in the path is electrically connected. The content represented by the nodes corresponding to the candidate tracing paths that pass the above three checks is identified as the fault-related objects, and the current path is output.

[0014] This invention provides a knowledge graph-based fault tracing system for disconnectors.

[0015] As a preferred embodiment of the knowledge graph-based disconnector fault tracing system described in this invention, it includes a data access unit for receiving disconnector equipment ledger data, operation ticket data, contact status data, auxiliary contact position signal data, alarm timestamp data, busbar segmentation relationship data, and electrical connection topology data. The ontology construction unit is used to establish a knowledge graph ontology of disconnect switch faults. The knowledge graph ontology includes at least the entity types of disconnect switch body, contact component, transmission mechanism, auxiliary contact, bus section, adjacent electrical equipment, operation event, abnormal event and state node, as well as relationship types such as composition relationship, connection relationship, ownership relationship, correspondence relationship, sequential relationship and state evolution relationship. The graph generation unit is used to convert various types of data received by the data access unit into corresponding entity, relation and attribute triples based on the knowledge graph ontology, and write them into the graph storage space to form a knowledge graph of switch faults that simultaneously includes the internal state evolution relationship of the switch and the external electrical connection relationship. The semantic path deduction unit is used to perform reverse path traversal in the switch fault knowledge graph according to the operation ticket step sequence and alarm timestamp sequence, starting from the entity node corresponding to the abnormal event and / or the state node corresponding to the abnormal contact state, and to limit the candidate tracing path by combining the bus segment relationship and electrical connection topology. The result determination unit is used to determine the component nodes, status nodes, and electrical connection-related nodes corresponding to the switch fault from the limited candidate tracing paths, and output the fault-related object and the corresponding tracing path.

[0016] The present invention provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of a knowledge graph-based method for tracing the fault source of a disconnect switch.

[0017] The present invention provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of a knowledge graph-based method for tracing the source of switch faults.

[0018] The beneficial effects of this invention are as follows: By constructing a knowledge graph that simultaneously represents the internal state evolution relationship and external electrical connection relationship of the disconnector, a unified semantic organization of multi-source heterogeneous data is achieved; by performing reverse path traversal according to the operation ticket step sequence and alarm timestamp sequence, and by combining the bus segmentation relationship and electrical connection topology to limit the candidate tracing path, the pertinence and accuracy of disconnector fault association object identification are improved; by performing consistency verification on the candidate tracing path and outputting the fault association object and corresponding tracing path, the fault analysis results have better coherence and interpretability, making it easier for operation and maintenance personnel to locate and handle faults. Attached Figure Description

[0019] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a schematic flowchart of a knowledge graph-based method for tracing faults in disconnectors, provided as an embodiment of the present invention.

[0021] Figure 2 This is a time sequence diagram of the formation of a knowledge graph for a disconnector fault tracing method based on a knowledge graph, provided in one embodiment of the present invention.

[0022] Figure 3 This is a flowchart illustrating the reverse path traversal and candidate path limitation of a knowledge graph-based disconnector fault tracing method, as provided in one embodiment of the present invention. Detailed Implementation

[0023] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0024] Example 1, referring to Figure 1 This is the first embodiment of the present invention, which provides a method for tracing the fault source of a disconnector based on a knowledge graph, including: S1: Receive data from the disconnector device and establish a disconnector fault knowledge graph ontology. The disconnector fault knowledge graph ontology includes entity types and relationship types.

[0025] S2: Convert the data received by the host into triples and write them into the graph storage space to form a knowledge graph of switch faults that includes internal state evolution relationships and external electrical connection relationships.

[0026] S3: Starting from the specified node, perform a reverse path traversal in the knowledge graph according to the order of operation ticket steps and alarm timestamps, and limit the candidate paths based on the bus segmentation relationship and electrical connection topology.

[0027] S4: Determine the node corresponding to the switch fault from the limited candidate tracing paths, and output the fault-related object and the corresponding tracing path.

[0028] Example 2, refer to Figure 2 and Figure 3 As an embodiment of the present invention, based on the above embodiment, a method for tracing the fault source of a disconnect switch based on a knowledge graph is provided.

[0029] Furthermore, in this embodiment of the application, step S1 receives disconnector device data and establishes a disconnector fault knowledge graph ontology, wherein the disconnector fault knowledge graph ontology includes entity types and relationship types, and the specific steps include: S101: Based on the component composition information in the disconnector equipment ledger data, establish the compositional relationship between the entity of the disconnector body and the entities of the contact components, the transmission mechanism, and the auxiliary contacts. The data access unit receives disconnector equipment ledger data, operation ticket data, contact status data, auxiliary contact position signal data, alarm timestamp data, busbar segmentation relationship data, and electrical connection topology data.

[0030] The disconnector equipment ledger data is used to characterize the basic information, component composition information, bay information, busbar segment information, and static association information with adjacent electrical equipment of the disconnector equipment; the operation ticket data is used to characterize the step number, step content, execution time, and related operation instruction information during the disconnector operation process; the contact status data is used to characterize the disconnector contact's position status, contact status, and action stage during the operation process; the auxiliary contact position signal data is used to characterize the position feedback information of the disconnector's auxiliary contacts at different times; the alarm timestamp data is used to characterize the time of occurrence and abnormality record of disconnector operation or abnormality; the busbar segment relationship data is used to characterize the attribution and association relationship between the disconnector and the corresponding busbar segment; and the electrical connection topology data is used to characterize the electrical connection relationship and topological path relationship between the disconnector body, busbar segment, and adjacent electrical equipment.

[0031] An ontology of knowledge graph of disconnect switch faults is established through ontology construction unit. The ontology of knowledge graph includes at least the entity types of disconnect switch body, contact component, transmission mechanism, auxiliary contact, bus section, adjacent electrical equipment, operation event, abnormal event and state node, as well as the relationship types of composition relationship, connection relationship, ownership relationship, correspondence relationship, sequential relationship and state evolution relationship.

[0032] S102: Based on the busbar segmentation relationship data and electrical connection topology data, establish the attribution relationship between the entity of the disconnect switch body and the entity of the busbar segment, and establish the connection relationship between the entity of the disconnect switch body and the entity of the adjacent electrical equipment. Based on the correspondence between operation ticket data and contact status data, a correspondence is established between the entities of operation events and the entities of status nodes; based on the association between alarm timestamp data and contact status and auxiliary contact position signals, a correspondence is established between the entities of abnormal events and contact status and auxiliary contact position signals.

[0033] It should be noted that the disconnector body entity is used to represent the disconnector equipment object to be analyzed for fault tracing; the contact component entity is used to represent the contact structure object inside the disconnector; the transmission mechanism entity is used to represent the mechanical transmission part that drives the disconnector to perform opening or closing actions; the auxiliary contact entity is used to represent the auxiliary contact object used for feedback of position status; the busbar segment entity is used to represent the busbar segment object associated with the disconnector; the adjacent electrical equipment entity is used to represent other electrical equipment that has a direct or indirect electrical connection with the disconnector; the operation event entity is used to represent the various operation step events in the operation ticket; the abnormal event entity is used to represent abnormal events generated during the operation of the disconnector; and the status node is used to represent the corresponding status representation node during the operation of the disconnector.

[0034] Among them, the composition relationship is used to characterize the compositional relationship between the disconnector body and the contact components, transmission mechanism, and auxiliary contacts; the connection relationship is used to characterize the electrical connection relationship between the disconnector body and the bus section and adjacent electrical equipment; the affiliation relationship is used to characterize the affiliation relationship between the disconnector body and the bus section or between the disconnector body and other operating units; the correspondence relationship is used to characterize the correspondence between operating events and state nodes, as well as the correspondence between abnormal events and contact states and auxiliary contact position signals; the sequential relationship is used to characterize the sequential relationship between different operating events according to the step number and execution time; and the state evolution relationship is used to characterize the changing relationship between adjacent state nodes during the disconnector operation.

[0035] In the embodiment, in the knowledge graph ontology of the disconnector fault established by the ontology construction unit, the disconnector body and contact components, transmission mechanism and auxiliary contacts are established in a compositional relationship; the disconnector body and bus section and adjacent electrical equipment are established in a connection relationship and / or affiliation relationship; operation events are established in a correspondence relationship with corresponding status nodes; and abnormal events are established in a correspondence relationship with contact status and auxiliary contact position signals.

[0036] Furthermore, in this embodiment, step S2 converts the data received by the body into triples and writes them into the graph storage space to form a switch fault knowledge graph containing internal state evolution relationships and external electrical connection relationships. Specific steps include: S201: Write the position status, contact status and action stage information of the contacts of phases A, B and C in the contact status data into the attribute field of the corresponding phase contact component entity. If the contact component entity is not distinguished separately, write it into the attribute field of the disconnector body entity. At the same time, generate the corresponding status node entity based on the contact status data of each phase at each sampling time. It should be noted that the graph generation unit generates triples to convert information in the disconnector equipment ledger data that represents equipment objects, component objects, connection objects, and belonging objects into corresponding entity triples and relation triples; information in the operation ticket data that represents operation steps, step content, and execution sequence into operation event entities and their attribute triples; information in the contact status quantity data that represents state changes into status nodes and their attribute triples; information in the auxiliary contact position signal data that represents position feedback into auxiliary contact-related entity, attribute, or relation triples; information in the alarm timestamp data that represents the anomaly type and occurrence time into anomaly event entities and their attribute triples; and information in the bus segmentation relationship data and electrical connection topology data that represents topology connections and belonging ranges into topology connection relation triples.

[0037] S202: Write the step number, step content, and execution time corresponding to each step in the operation ticket data into the attribute field of the operation event entity; write the occurrence time of each abnormal event in the alarm timestamp data into the attribute field of the abnormal event entity.

[0038] When generating triples, the graph generation unit writes the positional state, contact state, and action stage of the contact state quantity into the attribute items of the contact component entity or the disconnector body entity, and generates a state node corresponding to the contact state quantity; it writes the step number, step content, and execution time of the operation ticket data into the attribute items of the operation event entity, and writes the alarm timestamp data into the attribute items of the abnormal event entity.

[0039] S203: According to the step number in the operation ticket data in ascending order or the execution time in the order of first to last, establish a sequential relationship between the operation event entities to represent the order of execution, and connect each operation event entity in sequence to form an operation event sequence; According to the temporal sequence of changes in each state quantity in the contact state quantity data, a state evolution relationship representing the direction of state transition is established between adjacent state node entities, and each state node entity is connected sequentially to form a state node sequence; and according to the correspondence, the operation event entity in the operation event sequence is associated with the corresponding state node entity in the state node sequence to form a relationship network for the collaborative expression of operation steps and state changes.

[0040] Furthermore, the graph generation unit establishes sequential relationships between operation event entities based on the operation ticket step number and execution time, and establishes state evolution relationships between adjacent state nodes based on contact state data, so as to form a directed state chain of the disconnector operation process.

[0041] In establishing the sequential relationship, the operation event entities can be connected sequentially according to the step numbering order or execution time order in the operation ticket. In establishing the state evolution relationship, adjacent state nodes can be connected sequentially according to the acquisition time order, state change order, or action stage order of the contact state data, thus forming a directed state chain reflecting the disconnector operation process. The directed state chain is used to characterize the internal state evolution process of the disconnector from a certain initial state to subsequent states.

[0042] S204: Based on the busbar segmentation relationship data and electrical connection topology data, establish topology connection relationships between the disconnector body entity, the busbar segment entity, and adjacent electrical equipment entities; The sequence of state nodes and state evolution relationships formed by the operation ticket order and the changes in contact state quantities will be linked to the corresponding bus section entities and the electrical connection paths composed of the disconnector body, adjacent electrical equipment entities and topological connection relationships through the correspondence between the disconnector body entity and the bus section entity, thus forming a knowledge graph structure that unifies the internal state evolution and external electrical connection of the disconnector.

[0043] It should be noted that the map generation unit establishes topological connection relationships between the disconnector body entity, the bus section entity, and the adjacent electrical equipment entity based on the bus segment relationship data and the electrical connection topology data, and maps the directed state chain of the disconnector operation process to the corresponding bus section entity and electrical connection path, forming a two-layer semantic relationship structure.

[0044] The first layer of the two-layer semantic relationship structure can be used to characterize the state evolution relationship inside the disconnector, which consists of contact state variables, action stages and related state nodes; the second layer can be used to characterize the external electrical connection relationship between the disconnector body, busbar sections and adjacent electrical equipment.

[0045] By mapping the directed state chain of the disconnector operation process to the corresponding bus segment entity and electrical connection path, the internal state changes of the disconnector can be associated with the external electrical connection environment, thereby providing a unified semantic association basis for subsequent fault tracing.

[0046] Furthermore, in this embodiment, step S3 starts with a specified node, performs a reverse path traversal in the knowledge graph according to the order of operation ticket steps and alarm timestamps, and limits candidate paths based on bus segmentation relationships and electrical connection topology. Specific steps include: S301: The specified node is the entity node corresponding to the abnormal event and / or the status node corresponding to the abnormal contact state; Based on the chronological relationship between the alarm timestamp and the execution time of the operation event, operation event entities whose execution time is before the occurrence time of the abnormal event are selected as candidate operation events; starting from the state node corresponding to the candidate operation event, backtracking to the previous state node in the reverse direction of the state evolution relationship; The semantic path deduction unit takes the entity node corresponding to the abnormal event and / or the state node corresponding to the abnormal contact state as the starting node, performs reverse path traversal in the switch fault knowledge graph according to the operation ticket step sequence and alarm timestamp sequence, and limits the candidate tracing path by combining the bus segment relationship and electrical connection topology.

[0047] Specifically, when a disconnector malfunctions, if there is an abnormal alarm record, the entity node corresponding to the abnormal event can be used as the starting node; if there is an abnormal contact state but no abnormal alarm event is formed independently, the state node corresponding to the abnormal contact state can be used as the starting node; if both exist simultaneously, either one can be chosen as the starting node, or multiple starting nodes can be used simultaneously to perform reverse path traversal.

[0048] S302: During the backtracking process, based on the bus segment relationship data and electrical connection topology data, only candidate tracing paths that are located within the bus segment entity associated with the abnormal event or have a direct electrical connection with the disconnector body entity corresponding to the abnormal event are retained.

[0049] During the reverse path traversal, the semantic path deduction unit backtracks the relevant nodes in the graph according to the order of operation ticket steps and alarm timestamps. Specifically, it can combine the step number order of each step in the operation ticket and the time relationship between the occurrence time of the abnormal event and the execution time of the operation event to search backward from the abnormal event node or abnormal status node for corresponding status nodes, operation event nodes, switch body nodes, component nodes, and electrical connection related nodes, so as to form several candidate tracing paths.

[0050] S303: When performing reverse path traversal, the semantic path deduction unit first filters candidate operation events according to the chronological relationship between alarm timestamps and operation event execution times, then backtracks along the state evolution relationship corresponding to the candidate operation events, and retains only candidate tracing paths that are located within the bus segment range associated with the abnormal event or have a direct electrical connection with the disconnector corresponding to the abnormal event.

[0051] The process of screening candidate operation events can be based on the temporal relationship between the alarm timestamp and the execution time of each operation event, prioritizing operation events that are more temporally related to the abnormal event. The process of tracing back along the state evolution relationship corresponding to the candidate operation events can be carried out by tracing adjacent state nodes step by step in the reverse direction of the directed state chain to determine the state evolution path before the abnormality occurred. The requirement to retain only candidate tracing paths located within the bus segment associated with the abnormal event or having a direct electrical connection with the disconnector corresponding to the abnormal event refers to using bus segment relationship data and electrical connection topology data to limit the spatial and topological range of candidate paths, so as to exclude state paths and equipment paths that are unrelated to the target abnormality.

[0052] Furthermore, in this embodiment of the application, step S4 determines the node corresponding to the switch fault from the limited candidate tracing paths, and outputs the fault-related object and the corresponding tracing path. Specific steps include: S401: For each candidate tracing path, perform sequential checks on the consistency of operation steps, the continuity of state evolution, and the continuity of electrical connections. The result determination unit identifies the component nodes, status nodes, and electrical connection-related nodes corresponding to the switch fault from the limited candidate tracing paths, and outputs the fault-related objects and the corresponding tracing paths.

[0053] The result determination unit is used to perform consistency verification on candidate tracing paths. The consistency verification includes at least operation step sequence consistency verification, state evolution continuity verification, and electrical connection continuity verification. The contact components, transmission mechanisms, or auxiliary contacts represented by the nodes corresponding to the candidate tracing paths that simultaneously satisfy the consistency verification are determined as fault association objects.

[0054] S402: Operation step sequence consistency check is used to determine whether the operation events involved in the candidate tracing path conform to the operation ticket step number sequence and execution time sequence; The state evolution continuity check is used to determine whether a continuous and reasonable chain of state changes is formed between adjacent state nodes in the candidate tracing path; Electrical connection continuity check is used to determine whether the disconnector body nodes, busbar segment nodes and adjacent electrical equipment nodes involved in the candidate tracing path maintain continuity in topological connection relationship.

[0055] For candidate tracing paths that simultaneously meet the above consistency verification, the contact components, transmission mechanisms, or auxiliary contacts represented by their corresponding nodes can be identified as fault-associated objects, and the corresponding tracing paths can be output. The output results may include fault-associated objects, corresponding state nodes, electrical connection-related nodes, and node and relationship sequence information in the path.

[0056] Example 3 is the third embodiment of the present invention, which differs from the previous two embodiments in that: This embodiment also provides a knowledge graph-based switch fault tracing system, including: The data access unit is used to receive disconnector equipment ledger data, operation ticket data, contact status data, auxiliary contact position signal data, alarm timestamp data, busbar segmentation relationship data, and electrical connection topology data. The ontology construction unit is used to establish a knowledge graph ontology of disconnect switch faults. The knowledge graph ontology includes at least the entity types of disconnect switch body, contact component, transmission mechanism, auxiliary contact, bus section, adjacent electrical equipment, operation event, abnormal event and state node, as well as relationship types such as composition relationship, connection relationship, ownership relationship, correspondence relationship, sequential relationship and state evolution relationship. The graph generation unit is used to convert various types of data received by the data access unit into corresponding entity, relation and attribute triples based on the knowledge graph ontology, and write them into the graph storage space to form a knowledge graph of switch faults that simultaneously includes the internal state evolution relationship of the switch and the external electrical connection relationship. The semantic path deduction unit is used to perform reverse path traversal in the switch fault knowledge graph according to the operation ticket step sequence and alarm timestamp sequence, starting from the entity node corresponding to the abnormal event and / or the state node corresponding to the abnormal contact state, and to limit the candidate tracing path by combining the bus segment relationship and electrical connection topology. The result determination unit is used to determine the component nodes, status nodes, and electrical connection-related nodes corresponding to the switch fault from the limited candidate tracing paths, and output the fault-related object and the corresponding tracing path.

[0057] This embodiment also provides an electronic device, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to realize a knowledge graph-based method for tracing the fault source of a disconnect switch as proposed in the above embodiment.

[0058] This embodiment also provides a storage medium storing a computer program that, when executed by a processor, implements a knowledge graph-based method for tracing the fault source of a disconnector as proposed in the above embodiment.

[0059] The storage medium proposed in this embodiment belongs to the same inventive concept as the knowledge graph-based method for tracing faults of disconnectors proposed in the above embodiments. Technical details not described in detail in this embodiment can be found in the above embodiments, and this embodiment has the same beneficial effects as the above embodiments.

[0060] Based on the above description of the implementation methods, those skilled in the art can clearly understand that the present invention can be implemented using software and necessary general-purpose hardware, and of course, it can also be implemented using hardware. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory, hard disk, or optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of the various embodiments of the present invention.

[0061] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

[0062] Example 4 is an embodiment of the present invention, used to verify a knowledge graph-based method for tracing the fault source of a disconnect switch.

[0063] The following example illustrates an anomaly in a disconnector device at a substation. The disconnector device, numbered G202, belongs to bus section M1 and is electrically connected to adjacent electrical equipment circuit breaker QF201 and bus disconnector G203. The data access unit receives disconnector device ledger data, operation ticket data, contact status data, auxiliary contact position signal data, alarm timestamp data, bus section relationship data, and electrical connection topology data related to the disconnector device.

[0064] The disconnector equipment ledger data includes: equipment number G202, equipment name, associated bay, associated bus section M1, adjacent electrical equipment QF201 and G203, and information on the included A-phase contact components, B-phase contact components, C-phase contact components, transmission mechanism, and auxiliary contacts. The operation ticket data includes multiple operation steps related to the disconnector equipment, such as steps U1, U2, and U3. Step U1 corresponds to operation preparation, step U2 corresponds to performing the opening operation, and step U3 corresponds to confirming the operation result, corresponding to execution times t1, t2, and t3, respectively. The contact status data includes the separation status, contact status, and action stage information of phases A, B, and C, along with corresponding sampling times. The auxiliary contact position signal data includes the position status of auxiliary contact AJ202 at multiple sampling times. The alarm timestamp data includes the occurrence time t4 of the abnormal alarm E202, and t4 is later than the execution time t2 of step U2 and close to the execution time t3 of step U3. The bus segment relationship data indicates that the disconnector device G202 belongs to the bus segment M1, and the electrical connection topology data indicates that the disconnector device G202 has a direct electrical connection with the circuit breaker QF201 and the bus disconnector G203, respectively.

[0065] Based on the aforementioned data, the ontology construction unit establishes a knowledge graph ontology for disconnector faults, defining entity types such as disconnector body, contact components, transmission mechanisms, auxiliary contacts, busbar segments, adjacent electrical equipment, operating events, abnormal events, and state nodes. It also defines relationship types such as compositional relationships, connection relationships, attribution relationships, correspondence relationships, sequential relationships, and state evolution relationships. Specifically, the disconnector body entity "G202" establishes a compositional relationship with the contact component entities "A-phase contact component," "B-phase contact component," and "C-phase contact component," the transmission mechanism entity "transmission mechanism 202," and the auxiliary contact entity "AJ202." The disconnector body entity "G202" establishes an attribution relationship with the busbar segment entity "M1," and a connection relationship with the adjacent electrical equipment entities "QF201" and "G203." The operating event entities "U1," "U2," and "U3" establish correspondence relationships with their respective state nodes. The abnormal event entity "E202" establishes correspondence relationships with abnormal contact states and abnormal auxiliary contact position signals.

[0066] The graph generation unit converts the received data into corresponding entity, relation, and attribute triples based on the knowledge graph ontology and writes them into the graph storage space. Specifically, based on the disconnector equipment ledger data, the following entities and relations are generated: disconnector body entity (G202), contact component entities (A-phase contact component, B-phase contact component, C-phase contact component), transmission mechanism entity (transmission mechanism 202), auxiliary contact entity (AJ202), busbar segment entity (M1), and adjacent electrical equipment entities (QF201, G203). Triples such as “G202—Construction Relationship—A-phase Contact Component”, “G202—Construction Relationship—B-phase Contact Component”, “G202—Construction Relationship—C-phase Contact Component”, “G202—Construction Relationship—Transmission Mechanism 202”, “G202—Construction Relationship—AJ202”, “G202—Attribution Relationship—M1”, “G202—Connection Relationship—QF201”, and “G202—Connection Relationship—G203” are established.

[0067] Furthermore, the graph generation unit writes the step number, step content, and execution time from the operation ticket data into the attribute items of the operation event entities "U1", "U2", and "U3", and establishes sequential relationships such as "U1 - sequential relationship - U2" and "U2 - sequential relationship - U3" based on the step number and execution time. The graph generation unit writes the position state, contact state, and action stage from the contact state quantity into the attribute items of the contact component entity or the disconnector body entity, and generates state nodes corresponding to each sampling time, such as state nodes N1, N2, N3, and N4. Here, N1 represents the initial state before disconnector operation, N2 represents the transition state after the disconnector operation, N3 represents the abnormal state of contact action, and N4 represents the state of auxiliary contact feedback not being in place. At the same time, based on the contact state quantity data, state evolution relationships such as "N1 - state evolution relationship - N2", "N2 - state evolution relationship - N3", and "N3 - state evolution relationship - N4" are established between adjacent state nodes to form a directed state chain of the disconnector operation process. The graph generation unit also writes alarm timestamp data into the attribute field of the abnormal event entity "E202" and establishes correspondences such as "E202—correspondence relationship—N3" and "E202—correspondence relationship—AJ202 abnormal position signal". Based on the bus segment relationship data and electrical connection topology data, a topological connection relationship is established between the disconnector body entity, the bus segment entity, and the adjacent electrical equipment entity, and the directed state chain is mapped to the bus segment entity "M1" and the corresponding electrical connection path, forming a two-layer semantic relationship structure that simultaneously includes the internal state evolution relationship of the disconnector and the external electrical connection relationship.

[0068] Based on this, the semantic path derivation unit uses the entity node corresponding to the abnormal event entity "E202" and / or the state node "N3" corresponding to the abnormal contact state as the starting node, and performs a reverse path traversal in the switch fault knowledge graph according to the order of operation ticket steps and alarm timestamps. Specifically, the semantic path derivation unit first filters candidate operation events according to the chronological relationship between alarm timestamp t4 and operation event execution times t1, t2, and t3, retaining operation events "U2" and "U3" that are more temporally relevant to the abnormal event; then it backtracks along the state evolution relationship corresponding to the candidate operation events, tracing back from state node N3 to state nodes N2 and N1, and further associating it with operation event "U2". After obtaining the initial candidate paths, the semantic path derivation unit combines the bus segment relationship and electrical connection topology to limit the candidate tracing paths, retaining only the candidate tracing paths that are located within the bus segment M1 associated with the abnormal event, or that have a direct electrical connection with the disconnector G202 corresponding to the abnormal event, thereby excluding irrelevant paths that do not belong to the bus segment M1 and do not have a direct electrical connection with the disconnector G202.

[0069] The result determination unit performs consistency checks on the defined candidate tracing paths. The consistency checks include at least: operation step sequence consistency checks, state evolution continuity checks, and electrical connection continuity checks. Specifically, the operation step sequence consistency check determines whether the operation events involved in the candidate path conform to the step sequence relationship from U1 to U3 in the operation ticket; the state evolution continuity check determines whether the state evolution chain formed from N1 to N4 is continuous and conforms to the logic of the disconnector changing from a normal state to an abnormal state; and the electrical connection continuity check determines whether the disconnector body G202, bus section M1, and adjacent electrical equipment QF201 and G203 involved in the candidate path are continuously reachable in terms of topological connection. For candidate tracing paths that simultaneously satisfy the above consistency checks, the result determination unit identifies the contact components, transmission mechanisms, or auxiliary contacts represented by the corresponding nodes as fault-associated objects. For example, in this embodiment, if after verification it is determined that the abnormal state node N3 directly corresponds to the abnormal contact state of the A-phase contact component, and the abnormal position feedback of the auxiliary contact AJ202 is consistent with the state evolution chain, then the A-phase contact component and / or the auxiliary contact AJ202 can be identified as fault-related objects, and the corresponding tracing path can be output.

[0070] Ultimately, the output of the result determination unit may include: fault-related object identifier, fault-related object type, corresponding status node, corresponding electrical connection related nodes, and a complete tracing path sequence. This tracing path sequence can be an ordered path starting from the abnormal event entity "E202," passing through the corresponding status node, operation event node, disconnector body node, component node, and electrical connection related nodes. Through this method, reverse semantic derivation from the abnormal event to the fault-related object of the disconnector fault can be achieved, thus providing a basis for disconnector fault location and operation and maintenance analysis.

Claims

1. A method for tracing the fault source of a disconnect switch based on a knowledge graph, characterized in that, include: Receive data from disconnector devices and establish a disconnector fault knowledge graph ontology, which includes entity types and relationship types; The data received by the body is converted into triples and written into the graph storage space to form a knowledge graph of switch faults that includes internal state evolution relationships and external electrical connection relationships; Starting from a specified node, a reverse path traversal is performed in the knowledge graph according to the order of operation ticket steps and alarm timestamps, and candidate paths are limited based on bus segmentation relationships and electrical connection topology. Identify the node corresponding to the switch fault from the limited candidate tracing paths, and output the fault-related object and the corresponding tracing path.

2. The knowledge graph-based method for tracing faults in disconnectors as described in claim 1, characterized in that, The establishment of the knowledge graph ontology for disconnector faults includes: establishing a structural relationship between the entities of the disconnector body and the entities of the contact components, the transmission mechanism, and the auxiliary contacts, based on the component composition information in the disconnector equipment ledger data; Based on the busbar segmentation relationship data and electrical connection topology data, establish the attribution relationship between the entity of the disconnect switch body and the entity of the busbar segment, and establish the connection relationship between the entity of the disconnect switch body and the entity of adjacent electrical equipment. Based on the correspondence between operation ticket data and contact status data, a correspondence is established between the entities of operation events and the entities of status nodes; based on the association between alarm timestamp data and contact status and auxiliary contact position signals, a correspondence is established between the entities of abnormal events and contact status and auxiliary contact position signals.

3. The knowledge graph-based method for tracing faults in disconnectors as described in claim 2, characterized in that, The step of converting the data received by the body into a triplet includes: writing the positional state, contact state and action stage information of the contacts of phases A, B and C in the contact state quantity data into the attribute field of the corresponding phase contact component entity; if the contact component entity is not distinguished separately, it is written into the attribute field of the disconnector body entity; and generating the corresponding state node entity based on the contact state quantity data of each phase at each sampling time. Write the step number, step content, and execution time corresponding to each step in the operation ticket data into the attribute field of the operation event entity; write the occurrence time of each abnormal event in the alarm timestamp data into the attribute field of the abnormal event entity.

4. The knowledge graph-based method for tracing faults in disconnectors as described in claim 3, characterized in that, Also includes: After generating operation event entities and status node entities, establish a sequential relationship between operation event entities according to the step numbers in the operation ticket data in ascending order or the execution time in the order of first to last, and connect each operation event entity in sequence to form an operation event sequence. According to the temporal sequence of changes in each state quantity in the contact state quantity data, a state evolution relationship representing the direction of state transition is established between adjacent state node entities, and each state node entity is connected sequentially to form a state node sequence; and according to the correspondence, the operation event entity in the operation event sequence is associated with the corresponding state node entity in the state node sequence to form a relationship network for the collaborative expression of operation steps and state changes.

5. The method for tracing the fault source of a disconnect switch based on a knowledge graph as described in claim 4, characterized in that, The formation of the disconnector fault knowledge graph, which includes internal state evolution relationships and external electrical connection relationships, includes: establishing topological connection relationships between the disconnector body entity, the bus section entity, and adjacent electrical equipment entities based on bus segmentation relationship data and electrical connection topology data. The sequence of state nodes and state evolution relationships formed by the operation ticket order and the changes in contact state quantities will be linked to the corresponding bus section entities and the electrical connection paths composed of the disconnector body, adjacent electrical equipment entities and topological connection relationships through the correspondence between the disconnector body entity and the bus section entity, thus forming a knowledge graph structure that unifies the internal state evolution and external electrical connection of the disconnector.

6. The method for tracing the fault source of a disconnect switch based on a knowledge graph as described in claim 5, characterized in that, The limited candidate path includes: the specified node is the entity node corresponding to the abnormal event and / or the state node corresponding to the abnormal contact state; Based on the chronological relationship between the alarm timestamp and the execution time of the operation event, operation event entities whose execution time is before the occurrence time of the abnormal event are selected as candidate operation events; starting from the state node corresponding to the candidate operation event, backtracking to the previous state node in the reverse direction of the state evolution relationship; During the backtracking process, based on the bus segment relationship data and electrical connection topology data, only candidate tracing paths that are located within the bus segment entity associated with the abnormal event or have a direct electrical connection with the disconnector body entity corresponding to the abnormal event are retained.

7. The method for tracing the fault source of a disconnect switch based on a knowledge graph as described in claim 6, characterized in that, The step of determining the node corresponding to the disconnector fault from the limited candidate tracing paths includes: performing sequential consistency checks of operation steps, continuity checks of state evolution, and continuity checks of electrical connections for each candidate tracing path. The operation step sequence consistency check determines whether the sequential relationship between operation event entities in the path is consistent with the step number sequence of the operation ticket. The state evolution continuity check determines whether the state evolution relationship between adjacent state node entities in the path is continuous and without any breaks; The electrical connection continuity check determines whether the topological connection relationship between the disconnector body entity, busbar segment entity and adjacent electrical equipment entity involved in the path is electrically connected. The content represented by the nodes corresponding to the candidate tracing paths that pass the above three checks is identified as the fault-related objects, and the current path is output.

8. A knowledge graph-based disconnector fault tracing system, employing the knowledge graph-based disconnector fault tracing method as described in any one of claims 1 to 7, characterized in that, include: The data access unit is used to receive disconnector equipment ledger data, operation ticket data, contact status data, auxiliary contact position signal data, alarm timestamp data, busbar segmentation relationship data, and electrical connection topology data. Ontology building unit, used to establish the ontology of knowledge graph of disconnector faults; The graph generation unit is used to convert various types of data received by the data access unit into corresponding entity, relation and attribute triples based on the knowledge graph ontology, and write them into the graph storage space to form a knowledge graph of switch faults that simultaneously includes the internal state evolution relationship of the switch and the external electrical connection relationship. The semantic path deduction unit is used to perform reverse path traversal in the switch fault knowledge graph according to the operation ticket step sequence and alarm timestamp sequence, starting from the entity node corresponding to the abnormal event and / or the state node corresponding to the abnormal contact state, and to limit the candidate tracing path by combining the bus segment relationship and electrical connection topology. The result determination unit is used to determine the component nodes, status nodes, and electrical connection-related nodes corresponding to the switch fault from the limited candidate tracing paths, and output the fault-related object and the corresponding tracing path.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the knowledge graph-based switch fault tracing method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the knowledge graph-based switch fault tracing method according to any one of claims 1 to 7.

Images