A method of segmenting display of a relay control circuit fault

Abstract

The application relates to the technical field of relay protection, and discloses a method for segmentally displaying relay control circuit faults, which comprises the following steps: taking the fault propagation logic in a relay control circuit knowledge base as a framework, taking a fault entity as a node, and taking a propagation relationship as an edge to construct a fault knowledge graph; based on abnormal features in the relay control circuit, bidirectional graph reasoning is performed on the fault knowledge graph to obtain potential subsequent symptoms and potential root causes; the potential subsequent symptoms and the potential root causes are subjected to causal path synthesis to obtain a complete causal chain; based on the complete causal chain, link priority of the relay control circuit is evaluated to obtain a main causal chain; based on the main causal chain, the relay control circuit is subjected to logical paragraph division to obtain a segmented logical structure; the segmented logical structure is subjected to visual interactive design to obtain segmented display data and triggerable detailed diagnostic information; and the application can improve the fault troubleshooting efficiency.

Description

Technical Field

[0001] This invention relates to the field of relay protection technology, and in particular to a method for segmented display of relay control circuit faults. Background Technology

[0002] Relay control circuits are widely used in industrial automation systems, and their fault diagnosis is crucial for ensuring stable equipment operation. Existing fault diagnosis methods for relay control circuits are mostly based on predefined rule bases or simple graph structure analysis. They locate fault points by detecting abnormal signals or state changes in the circuit and use empirical knowledge bases to infer fault causes. In recent years, some research has introduced knowledge graph technology, modeling fault entities and propagation relationships as graph nodes and edges to support basic fault reasoning and visualization. These methods have improved the automation level of fault diagnosis to some extent, providing maintenance personnel with auxiliary decision-making support.

[0003] Existing fault diagnosis methods for relay control circuits have limitations in technical implementation. First, the construction of fault knowledge graphs fails to fully integrate fault propagation logic, resulting in incomplete coverage of graph nodes and edges, affecting the integrity of fault propagation paths. Second, fault reasoning processes mostly employ unidirectional analysis, lacking bidirectional reasoning mechanisms, making it difficult to simultaneously trace root causes and predict potential symptoms, thus reducing diagnostic accuracy. Furthermore, redundant paths are not effectively filtered during causal path synthesis, resulting in lengthy causal chains with unclear key information. Additionally, the lack of effective link priority evaluation methods makes it impossible to identify the main propagation path from multiple causal chains, impacting diagnostic efficiency. Finally, the visualization display lacks logical segmentation and interactive design, leading to scattered fault information presentation, which is not conducive to users quickly capturing core fault logic. Therefore, improving the efficiency of fault diagnosis has become an urgent problem to be solved. Summary of the Invention

[0004] This invention provides a method for segmented display of relay control circuit faults to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides a method for segmented display of relay control circuit faults, comprising: S1. Using the fault propagation logic in the relay control circuit knowledge base as a framework, the fault entities in the relay control circuit knowledge base as nodes, and the propagation relationships in the relay control circuit knowledge base as edges, construct a fault knowledge graph for the relay control circuit. S2. Based on the abnormal characteristics in the relay control circuit, perform bidirectional graph reasoning on the fault knowledge graph to obtain the potential subsequent symptoms and potential root causes of the relay control circuit fault. S3. Perform causal path synthesis on the potential subsequent symptoms and the potential root causes to obtain the complete causal chain of the relay control circuit failure; S4. Based on the complete causal chain, perform link priority evaluation on the relay control circuit to obtain the main causal chain of the relay control circuit fault; S5. Based on the main causal chain, the relay control circuit is divided into logical segments to obtain the segmented logical structure of the relay control circuit fault. S6. Perform a visual interactive design on the segmented logic structure to obtain segmented display data of the relay control circuit fault and detailed diagnostic information that can be triggered.

[0006] In a preferred embodiment, the construction of a fault knowledge graph for the relay control circuit, using the fault propagation logic in the relay control circuit knowledge base as a framework, fault entities in the relay control circuit knowledge base as nodes, and propagation relationships in the relay control circuit knowledge base as edges, includes: Extract the fault propagation logic from the relay control circuit knowledge base to obtain the fault propagation logic framework of the relay control circuit. Based on the fault propagation logic framework, fault entity identification is performed on the relay control circuit knowledge base to obtain the fault entity nodes of the relay control circuit. Based on the propagation relationships in the relay control circuit knowledge base, construct the propagation relationship edges of the relay control circuit; Based on the fault propagation logic framework, the fault entity nodes and the propagation relationship edges are integrated into a graph to obtain the fault knowledge graph of the relay control circuit.

[0007] In a preferred embodiment, the step of performing bidirectional graph reasoning on the fault knowledge graph based on the abnormal characteristics in the relay control circuit to obtain the potential subsequent symptoms and potential root causes of the relay control circuit fault includes: The abnormal features in the relay control circuit are analyzed to obtain the set of associated nodes in the fault knowledge graph; Bidirectional graph reasoning is performed on the set of associated nodes to obtain the propagation influence values ​​of candidate nodes in the fault knowledge graph; Based on the propagation influence value, a forward priority traversal is performed on the fault knowledge graph to obtain the subsequent node sequence of the fault knowledge graph. Based on the propagation influence value, the fault knowledge graph is traversed in reverse priority to obtain the predecessor node sequence of the fault knowledge graph. Extract the origin entity from the predecessor node sequence and the representation entity from the subsequent node sequence, respectively; Based on the propagation influence value, the symptom intensity of the characterized entity is assessed to obtain candidate symptoms of the relay control circuit fault. Based on the propagation influence value, the root cause contribution of the originating entity is determined to obtain the candidate root causes of the relay control circuit fault. Verify the correlation logic of the propagation path in the candidate symptoms to obtain the potential subsequent symptoms of the relay control circuit failure; By verifying the reverse causal logic of the candidate root causes, the potential root causes of the relay control circuit failure can be obtained.

[0008] In a preferred embodiment, the formula for calculating the propagation influence value is: ; In the formula, This represents the propagation influence value of the candidate node. This refers to the set of associated nodes in the fault knowledge graph. For the set of associated nodes Size, For the set of associated nodes Each symptom node in the data, For the fault knowledge graph from candidate nodes To the symptom node The shortest path distance, This is a decay factor used to control the rate of decay of propagation influence. Candidate nodes To the symptom node The strength of the propagation relationship between them It is an exponentially decaying function.

[0009] In a preferred embodiment, the step of synthesizing causal paths from the potential subsequent symptoms and the potential root causes to obtain the complete causal chain of the relay control circuit fault includes: A reverse search is performed on the potential subsequent symptoms to obtain the source path of the potential subsequent symptoms; A forward search is performed on the potential root cause to obtain the endpoint path of the potential root cause; By connecting the associated paths between the source path and the destination path, a causal relationship network of the relay control circuit fault is obtained. By eliminating redundant and secondary paths in the causal relationship network, the complete causal chain of the relay control circuit fault is obtained.

[0010] In a preferred embodiment, the step of eliminating redundant and secondary paths in the causal relationship network to obtain the complete causal chain of the relay control circuit fault includes: Based on the causal relationship network, the propagation relationship in the relay control circuit knowledge base is intensity quantized to obtain the path weight distribution of the causal relationship network; Extracting low-weight paths from the path weight distribution yields candidate redundant paths for the causal relationship network; Verify the critical connectivity of the candidate redundant paths to obtain the redundant paths of the causal relationship network; By removing redundant paths from the causal relationship network, the complete causal chain of the relay control circuit fault is obtained.

[0011] In a preferred embodiment, the step of performing link priority evaluation on the relay control circuit based on the complete causal chain to obtain the main causal chain of the relay control circuit fault includes: The complete causal chain is subjected to comprehensive feature quantification to obtain the evaluation parameters of the complete causal chain; The evaluation parameters are used to construct a feature tensor to obtain the fused feature vector of the complete causal chain; Based on the preset fault impact level rules, the fused feature vector is subjected to rule matching to obtain the rule matching result of the complete causal chain; The influence weight distribution of the complete causal chain is obtained by performing weight scaling mapping on the rule matching results. Based on the influence weight distribution, the complete causal chain is prioritized to obtain the main causal chain of the relay control circuit fault.

[0012] In a preferred embodiment, prioritizing the complete causal chain based on the influence weight distribution to obtain the main causal chain of the relay control circuit fault includes: The influence weight distribution is sorted in descending order to obtain the weight ranking list of the complete causal chain; Extract the high-weight causal chains from the weighted list; Based on the threshold values ​​defined in the preset fault impact level rules, the high-weight causal chains are matched and filtered to obtain the candidate main causal chains of the relay control circuit. Based on the influence weight distribution and the fault influence level classification, the root cause determination of the candidate main causal chain is performed to obtain the main causal chain of the relay control circuit fault.

[0013] In a preferred embodiment, the step of dividing the relay control circuit into logical segments based on the main causal chain to obtain a segmented logical structure for relay control circuit faults includes: Identify the logical boundaries in the main causal chain to obtain the logical splitting points of the main causal chain; Based on the logical segmentation points, the main causal chain is logically segmented to obtain the initial logical segments of the main causal chain; The initial logic paragraph is subjected to semantic consistency verification to obtain the logic paragraph after verification of the relay control circuit fault. Based on the topological relationships in the fault knowledge graph, the verified logic segments are integrated to obtain the segmented logic structure of the relay control circuit fault.

[0014] In a preferred embodiment, the step of performing a visual interactive design on the segmented logic structure to obtain segmented display data of the relay control circuit fault and triggerable detailed diagnostic information includes: The segmented logical structure is mapped to visual elements to obtain the visual element configuration of the segmented logical structure. Based on the configuration of the visual elements, the segmented logical structure is hierarchically laid out to obtain the topological layout scheme of the segmented logical structure. An interactive response mechanism is designed for the topology layout scheme to obtain the interactive response rules of the topology layout scheme; Based on the interactive response rules, the segmented logic structure is encapsulated for display data to obtain segmented display data of the relay control circuit fault and detailed diagnostic information that can be triggered.

[0015] Compared with the prior art, the present invention has the following beneficial effects: 1. Significantly improves the accuracy of fault diagnosis. This method first uses fault propagation logic as a framework to integrate fault entities and propagation relationships in the knowledge base to construct a complete fault knowledge graph, ensuring comprehensive coverage of nodes and edges and solving the problem of incomplete coverage in traditional graphs. Then, it conducts bidirectional graph reasoning based on circuit anomaly characteristics. By parsing the set of associated nodes and calculating the propagation influence value, it simultaneously traces the potential root cause of the fault and predicts subsequent symptoms, making up for the limitations of traditional unidirectional reasoning. Finally, in the causal path synthesis, invalid paths are filtered out through steps such as propagation relationship strength quantification and redundant path verification and elimination, making the core information of the causal chain more prominent and effectively improving the accuracy of diagnostic results.

[0016] 2. Significantly improves fault diagnosis efficiency and ease of use. By comprehensively quantifying features, matching rules, and mapping weight scales for the complete causal chain, it achieves link priority assessment and quickly filters out the main causal chain from multiple causal chains, solving the problem that traditional methods struggle to identify the core path. Subsequently, based on the main causal chain, it identifies logical boundaries and segments and verifies logical segments. Then, through visualization element mapping, hierarchical layout, and interactive response mechanism design, it transforms the segmented logical structure into intuitive segmented display data. Coupled with triggerable detailed diagnostic information, it solves the shortcomings of traditional scattered visualization information, helps users quickly capture core fault logic, and improves the convenience of diagnostic operations. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating a method for segmented display of relay control circuit faults according to an embodiment of the present invention. The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0018] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0019] This application provides a method for segmented display of relay control circuit faults. The execution subject of this method includes, but is not limited to, at least one of the following electronic devices that can be configured to execute the method provided in this application: a server, a terminal, etc. In other words, the method for segmented display of relay control circuit faults can be executed by software or hardware installed on a terminal device or a server device. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster. The server can be an independent server or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.

[0020] Reference Figure 1 The diagram shown is a flowchart illustrating a method for segmented display of relay control circuit faults according to an embodiment of the present invention. In this embodiment, the method for segmented display of relay control circuit faults includes: S1. Using the fault propagation logic in the relay control circuit knowledge base as a framework, the fault entities in the relay control circuit knowledge base as nodes, and the propagation relationships in the relay control circuit knowledge base as edges, construct a fault knowledge graph for the relay control circuit. In this embodiment of the invention, the construction of a fault knowledge graph for the relay control circuit, using the fault propagation logic in the relay control circuit knowledge base as a framework, fault entities in the relay control circuit knowledge base as nodes, and propagation relationships in the relay control circuit knowledge base as edges, includes: Extract the fault propagation logic from the relay control circuit knowledge base to obtain the fault propagation logic framework of the relay control circuit. Based on the fault propagation logic framework, fault entity identification is performed on the relay control circuit knowledge base to obtain the fault entity nodes of the relay control circuit. Based on the propagation relationships in the relay control circuit knowledge base, construct the propagation relationship edges of the relay control circuit; Based on the fault propagation logic framework, the fault entity nodes and the propagation relationship edges are integrated into a graph to obtain the fault knowledge graph of the relay control circuit.

[0021] The work of extracting fault propagation logic was carried out, and all kinds of fault-related information stored in the relay control circuit knowledge base were comprehensively sorted out. The inherent transmission rules and correlation patterns between different faults were deeply explored, and the core logical framework that can run through various fault situations was accurately extracted. In this way, a fault propagation logic framework for relay control circuits was formed, which provides basic guidance for subsequent graph construction.

[0022] After obtaining the fault propagation logic framework, a comprehensive investigation and screening of the relay control circuit knowledge base is carried out based on it. Each independent element related to the fault in the knowledge base is identified one by one. These elements cover the components where the fault occurs, the form of the fault, and the various factors that cause the fault. These elements are clearly defined as fault entity nodes to ensure that each node can accurately correspond to a specific fault-related content in the knowledge base.

[0023] The association information between different fault entities recorded in the relay control circuit knowledge base is systematically sorted out to clarify how different fault entities interact and influence each other, thus leading to fault propagation. Based on these actual associations, propagation relationship edges connecting each fault entity node are established so that each edge can accurately reflect the real fault propagation association between two nodes.

[0024] Guided by the previously established fault propagation logic framework, all identified fault entity nodes and constructed propagation relationship edges are organically integrated. Following the logical order and inherent connections of fault propagation, nodes and edges are rationally arranged and combined to ensure that the entire graph can completely and accurately represent the propagation path and connections of faults in the relay control circuit, ultimately forming a complete fault knowledge graph for the relay control circuit. The beneficial effects are as follows: Extracting the fault propagation logic can extract core fault transmission patterns from the complex information in the knowledge base, providing clear guidance for graph construction, avoiding directional deviations, and improving the targeting of graph construction; Identifying fault entity nodes can clearly define the scattered fault-related elements in the knowledge base, making fault information more organized and laying the foundation for establishing connections between nodes; Constructing propagation relationship edges can accurately present the interactions between fault entities, making the fault propagation path intuitively visible, providing a connection basis for understanding fault spread and finding the root cause; Graph integration integrates nodes and edges according to a unified logic to form a complete fault knowledge graph, making it easier for staff to grasp the fault association situation, providing reliable support for subsequent fault reasoning and analysis, and improving fault handling efficiency.

[0025] S2. Based on the abnormal characteristics in the relay control circuit, perform bidirectional graph reasoning on the fault knowledge graph to obtain the potential subsequent symptoms and potential root causes of the relay control circuit fault. In this embodiment of the invention, the step of performing bidirectional graph reasoning on the fault knowledge graph based on the abnormal characteristics in the relay control circuit to obtain the potential subsequent symptoms and potential root causes of the relay control circuit fault includes: The abnormal features in the relay control circuit are analyzed to obtain the set of associated nodes in the fault knowledge graph; Bidirectional graph reasoning is performed on the set of associated nodes to obtain the propagation influence values ​​of candidate nodes in the fault knowledge graph; Based on the propagation influence value, a forward priority traversal is performed on the fault knowledge graph to obtain the subsequent node sequence of the fault knowledge graph. Based on the propagation influence value, the fault knowledge graph is traversed in reverse priority to obtain the predecessor node sequence of the fault knowledge graph. Extract the origin entity from the predecessor node sequence and the representation entity from the subsequent node sequence, respectively; Based on the propagation influence value, the symptom intensity of the characterized entity is assessed to obtain candidate symptoms of the relay control circuit fault. Based on the propagation influence value, the root cause contribution of the originating entity is determined to obtain the candidate root causes of the relay control circuit fault. Verify the correlation logic of the propagation path in the candidate symptoms to obtain the potential subsequent symptoms of the relay control circuit failure; By verifying the reverse causal logic of the candidate root causes, the potential root causes of the relay control circuit failure can be obtained.

[0026] The formula for calculating the value of the dissemination influence is: ; In the formula, This represents the propagation influence value of the candidate node. This refers to the set of associated nodes in the fault knowledge graph. For the set of associated nodes Size, For the set of associated nodes Each symptom node in the data, For the fault knowledge graph from candidate nodes To the symptom node The shortest path distance, This is a decay factor used to control the rate of decay of propagation influence. Candidate nodes To the symptom node The strength of the propagation relationship between them It is an exponentially decaying function.

[0027] First, the abnormal characteristics exhibited during the operation of the relay control circuit are analyzed. Here, "analysis" refers to a comprehensive breakdown and detailed analysis of various abnormal behaviors of the circuit—such as the fluctuation range of voltage, the range of sudden increases or decreases in current, the frequency of abnormal sounds emitted by the contactor, and the overheating temperature of the coil. The specific manifestations, locations in the circuit, and durations of these abnormal characteristics are identified one by one. After the analysis of the abnormal characteristics is completed, the analyzed abnormal information is compared with all nodes in the constructed fault knowledge graph. All nodes directly related to the current abnormal characteristics are selected, and then these selected nodes are integrated to form the set of associated nodes in the fault knowledge graph.

[0028] A comprehensive disassembly and detailed analysis of various abnormal characteristics exhibited during the operation of relay control circuits is conducted. This comprehensive disassembly and detailed analysis refers to systematically and thoroughly investigating all abnormal behaviors of the circuit, such as voltage fluctuations, current abrupt changes, abnormal contactor noise frequencies, and coil overheating temperatures. Through this operation, the specific manifestations, locations, and durations of abnormal characteristics in the circuit are clearly identified. After the comprehensive disassembly and detailed analysis of abnormal characteristics is completed, the abnormal information obtained from the disassembly and analysis is compared and matched with all nodes in the constructed fault knowledge graph. This comparison and matching involves meticulously comparing the attributes and characteristics of each abnormal information with those of the nodes in the knowledge graph to identify matching nodes. Then, all nodes directly related to the current abnormal characteristic are selected. This selection process is based on the comparison and matching results, eliminating unrelated or indirectly related nodes and retaining directly related nodes. Finally, the selected nodes are summarized and integrated, which means consolidating the scattered related nodes into a complete set, ultimately obtaining the set of related nodes in the fault knowledge graph.

[0029] A bidirectional inference operation is performed on the obtained set of associated nodes, both probing forward to deduce possible development directions and tracing backward to possible sources. This bidirectional inference, on the one hand, explores the potential subsequent impacts of associated nodes along the fault propagation direction, and on the other hand, traces back the initial initiating factors of nodes against the fault propagation direction. Through this bidirectional inference, the propagation influence data of each candidate node in the fault knowledge graph is obtained, forming a propagation influence value. During the bidirectional inference process, the propagation influence value of the candidate node is determined through specific calculation logic, as follows: ; in, Representative candidate nodes The propagation influence value is used to measure the candidate node. The higher the value, the more critical its role in fault propagation. The set of associated nodes representing the fault knowledge graph is a selection of nodes from the knowledge graph that are directly related to the current circuit anomaly characteristics, providing fundamental data for calculating the propagation impact. Represents the set of associated nodes The total number of symptom nodes is used to normalize the summation result, avoiding incomparable influence values ​​due to differences in the number of associated nodes in different scenarios, and ensuring... It has significance for horizontal comparison. Represents the set of associated nodes Each symptom node in the algorithm, that is, the specific fault manifestation node related to the current abnormal feature, serves as a reference for calculating the influence of candidate nodes. Representative fault knowledge graph from candidate nodes To the symptom node The shortest path distance, when When it increases, This decreases accordingly, leading to The decrease reflects the natural attenuation of fault impact as propagation distance increases, where λ represents the attenuation factor that controls the rate of attenuation of propagation impact. The larger the value, Follow The faster the rate of increase or decrease, the more significant the attenuation of the fault's impact. This can be flexibly set according to the fault propagation characteristics of the relay control circuit type to ensure that the attenuation law closely matches reality. Representative candidate nodes To the symptom node The strength of the propagation relationship is determined based on the fault propagation logic of the relay control circuit knowledge base. The larger the right The greater the contribution, the stronger the correlation between fault propagation among nodes. Representing the exponential decay function, it transforms the decay effect of the shortest path distance on the influence into a smooth numerical change, avoiding abrupt changes in the influence value caused by small changes in distance, and ensuring that the calculation results closely match the gradual process of actual fault propagation.

[0030] The calculation logic is derived from the constructed fault knowledge graph, combined with the characteristics of the associated node set and the objective laws of fault propagation in relay control circuits. By integrating factors such as node distance, association strength, and attenuation characteristics, it can objectively and accurately quantify the propagation influence of each candidate node, providing a quantifiable basis for subsequent priority traversal, symptom assessment, and root cause determination, thus avoiding the subjective bias of relying solely on experience.

[0031] Based on the obtained propagation influence values, the nodes in the fault knowledge graph are examined one by one in descending order of value, following the direction of fault propagation. This operation of examining nodes in the direction of propagation and sorting them by influence value is called forward priority traversal. During the traversal, nodes with high priority are selected and arranged in order to obtain the subsequent node sequence of the fault knowledge graph. At the same time, based on the propagation influence values, the nodes in the fault knowledge graph are examined one by one in descending order of value, following the direction of fault propagation. This operation of examining nodes in the direction of propagation and sorting them by influence value is called reverse priority traversal. Similarly, nodes with high priority are selected and arranged in order to obtain the predecessor node sequence of the fault knowledge graph.

[0032] The process of identifying the core entity that initially caused the fault from the organized sequence of predecessor nodes is called extracting the origin entity. The process of identifying the core entity that manifests the fault from the subsequent node sequence is called extracting the representation entity. Through these two extraction operations, the origin entity in the sequence of predecessor nodes and the representation entity in the sequence of subsequent nodes are identified respectively.

[0033] Based on the previously obtained propagation influence values, the severity of the fault manifestation is assessed for the extracted representative entities. This assessment is called symptom intensity assessment. The higher the propagation influence value, the more obvious the fault manifestation of the representative entity and the more prominent its impact on circuit operation. Through this assessment, candidate symptoms of relay control circuit faults are identified. At the same time, based on the propagation influence values, the magnitude of the driving force of the extracted origin entity on the occurrence of the fault is analyzed. This analysis of the magnitude of the driving force is called root cause contribution determination. The higher the propagation influence value, the greater the contribution of the origin entity to the occurrence of the fault and the higher the probability of causing the fault. Through this determination, candidate root causes of relay control circuit faults are identified.

[0034] Carefully examine whether the propagation paths corresponding to candidate symptoms conform to the objective laws of circuit fault propagation, and whether there is a reasonable correlation between each link. This operation of verifying the rationality of the correlation is to verify the correlation logic of the propagation paths in candidate symptoms. By verifying and eliminating candidate symptoms that are illogical or do not conform to the actual propagation laws, the potential subsequent symptoms of relay control circuit faults are finally obtained. At the same time, carefully examine whether the reverse derivation causal relationship corresponding to the candidate root cause is valid and whether it can be reasonably traced back to the initial cause of the fault. This operation of verifying the rationality of reverse causation is to verify the reverse causal logic of the candidate root cause. By verifying and eliminating candidate root causes that are causal or cannot be traced back to the initial cause, the potential root cause of relay control circuit faults is finally obtained.

[0035] The beneficial effects are as follows: By dissecting and analyzing the abnormal characteristics of relay control circuits, and combining fault knowledge graph node matching, screening, and integration, the accuracy and reliability of the associated node set are ensured, providing basic data for subsequent reasoning, avoiding analytical bias, and improving the reliability of the starting point for fault analysis. The system bidirectionally derives and calculates the propagation influence value of the associated node set, explores the node's influence potential, integrates key factors to fit fault patterns, provides quantitative basis for subsequent analysis, overcomes the limitations of experience-based judgment, and improves the scientific rigor of the results. Based on the propagation influence value, a forward and reverse priority traversal is performed, focusing on key nodes according to their influence magnitude, avoiding blind screening from a massive number of nodes. This allows for the rapid identification of nodes important for fault development and source tracing, improving the efficiency of subsequent extraction of origin and characterization entities, ensuring extraction accuracy, reducing interference from irrelevant nodes, and enabling… Fault analysis is more targeted, extracting origin and representative entities from node sequences to clarify the core of the fault, simplifying subsequent analysis objects, avoiding inefficiency and judgment errors, clarifying the analysis direction, helping to accurately locate fault elements, assessing symptom intensity and determining root cause contribution based on propagation influence values, quantitatively screening candidate objects, avoiding subjective errors, ensuring practicality, improving reliability, providing high-quality materials for verification, verifying the path logic of candidate symptoms and the causal logic of root causes, eliminating unreasonable objects, providing high-quality data for causal chain synthesis, avoiding result distortion, ensuring the rigor of analysis conclusions, clarifying the symbol meaning and trend of propagation influence calculation logic, deducing based on graphs and circuit characteristics, making it understandable and reproducible, accurately reflecting the impact of nodes, providing mathematical support, and promoting the transformation of influence assessment from qualitative to quantitative.

[0036] S3. Perform causal path synthesis on the potential subsequent symptoms and the potential root causes to obtain the complete causal chain of the relay control circuit failure; In this embodiment of the invention, the step of synthesizing causal paths from the potential subsequent symptoms and the potential root causes to obtain the complete causal chain of the relay control circuit fault includes: A reverse search is performed on the potential subsequent symptoms to obtain the source path of the potential subsequent symptoms; A forward search is performed on the potential root cause to obtain the endpoint path of the potential root cause; By connecting the associated paths between the source path and the destination path, a causal relationship network of the relay control circuit fault is obtained. By eliminating redundant and secondary paths in the causal relationship network, the complete causal chain of the relay control circuit fault is obtained.

[0037] The process of eliminating redundant and secondary paths in the causal network to obtain the complete causal chain of the relay control circuit fault includes: Based on the causal relationship network, the propagation relationship in the relay control circuit knowledge base is intensity quantized to obtain the path weight distribution of the causal relationship network; Extracting low-weight paths from the path weight distribution yields candidate redundant paths for the causal relationship network; Verify the critical connectivity of the candidate redundant paths to obtain the redundant paths of the causal relationship network; By removing redundant paths from the causal relationship network, the complete causal chain of the relay control circuit fault is obtained.

[0038] When performing a reverse search to obtain the source path of potential subsequent symptoms, firstly, the specific node position corresponding to each potential subsequent symptom in the fault knowledge graph should be identified. The reverse search means starting from the node and tracing back along the reverse direction of fault propagation, checking one by one the preceding nodes that are directly or indirectly related to it, recording the connection relationship and path direction between each node in detail, without omitting any possible tracing direction, and finally sorting out the complete set of paths that can point to the source of the potential subsequent symptom, which is the source path.

[0039] When performing a forward search to find the endpoint path for a potential root cause, first determine the node corresponding to each potential root cause in the fault knowledge graph. The forward search means starting from the node, following the natural direction of fault propagation, sequentially finding subsequent nodes that have a direct or indirect propagation relationship with it, recording the propagation order and relationship between nodes in detail, and comprehensively covering all subsequent nodes that the root cause may affect, forming a complete set of paths extending from the potential root cause to each potentially affected node, which is the endpoint path.

[0040] When connecting the related paths between the source path and the destination path to obtain the causal relationship network, all nodes in the source path and the destination path are compared one by one to identify the nodes that overlap or have logical connections between the two types of paths. Path connection refers to using these related nodes as connection points to connect the source path, the intermediate paths corresponding to the related nodes, and the destination path in an orderly manner to ensure the logical coherence and correlation between the paths, and finally form a comprehensive relationship network covering the source of the fault, the intermediate propagation process, and the subsequent symptoms, that is, the causal relationship network.

[0041] When quantifying the propagation relationships in the relay control circuit knowledge base based on causal relationship networks to obtain path weight distribution, the strength quantification refers to analyzing and evaluating each propagation relationship in the causal relationship network based on this information, assigning a corresponding quantitative value to each propagation relationship according to the preset evaluation criteria. This value directly reflects the strength of the propagation relationship. Then, the quantitative values ​​of all propagation relationships are organized according to the corresponding paths to form the weight data set corresponding to each path, i.e., the path weight distribution.

[0042] When extracting low-weight paths from the path weight distribution to obtain candidate redundant paths, a reasonable weight threshold is first set by considering factors such as the degree of impact of relay control circuit failures and the normal propagation law. Extraction means screening all paths in the path weight distribution one by one against the threshold, and filtering out all paths with weight values ​​lower than the threshold. These filtered paths are the candidate redundant paths.

[0043] When verifying the critical connectivity of candidate redundant paths to obtain redundant paths, the position and connection role of each candidate redundant path in the causal relationship network are analyzed. Verification means checking whether the path is the only connection path between two key nodes, or whether there are other paths with higher weights that can replace it to achieve the same connectivity function. If the candidate path is not necessary for network connectivity and there are alternative paths, then the path is determined to be a redundant path.

[0044] When removing redundant paths from a causal network to obtain a complete causal chain, the identified redundant paths are explicitly marked. Removal means completely eliminating these paths from the causal network according to the marking results. At the same time, the logical connections between the remaining paths are checked to ensure that the removal operation does not disrupt the coherence of the core path. After removal, the remaining paths form a complete set of paths without redundancy and with clear logic, which is the complete causal chain of relay control circuit faults.

[0045] The beneficial effects include: accurately locating the source-related paths of potential subsequent symptoms, avoiding omissions in source investigation, providing a reliable starting point for constructing subsequent causal chains, presenting the subsequent propagation trajectory of potential root causes, fully covering the paths they may affect, ensuring the integrity of the causal chain in the direction of propagation, integrating source paths, related paths, and endpoint paths to form a unified causal relationship network, clearly presenting the complete propagation path of the fault from its source to subsequent symptoms, avoiding causal relationship breaks, converting the importance of propagation relationships into intuitive quantitative data, providing a scientific basis for redundant path identification, effectively improving the accuracy of redundant path elimination, quickly filtering low-weight paths based on weight thresholds, narrowing the scope of redundant path investigation, reducing invalid operations, improving the efficiency of preliminary identification of redundant paths, accurately distinguishing redundant paths from necessary paths in candidate paths, avoiding accidental deletion of critical paths that would damage the causal relationship network, ensuring the integrity of the network structure, simplifying the causal relationship network structure, highlighting the core propagation path of the fault, reducing interference from irrelevant paths, and further improving the efficiency and accuracy of fault diagnosis.

[0046] S4. Based on the complete causal chain, perform link priority evaluation on the relay control circuit to obtain the main causal chain of the relay control circuit fault; In this embodiment of the invention, the step of performing link priority evaluation on the relay control circuit based on the complete causal chain to obtain the main causal chain of the relay control circuit fault includes: The complete causal chain is subjected to comprehensive feature quantification to obtain the evaluation parameters of the complete causal chain; The evaluation parameters are used to construct a feature tensor to obtain the fused feature vector of the complete causal chain; Based on the preset fault impact level rules, the fused feature vector is subjected to rule matching to obtain the rule matching result of the complete causal chain; The influence weight distribution of the complete causal chain is obtained by performing weight scaling mapping on the rule matching results. Based on the influence weight distribution, the complete causal chain is prioritized to obtain the main causal chain of the relay control circuit fault.

[0047] The prioritization of the complete causal chain based on the influence weight distribution yields the main causal chain of the relay control circuit fault, including: The influence weight distribution is sorted in descending order to obtain the weight ranking list of the complete causal chain; Extract the high-weight causal chains from the weighted list; Based on the threshold values ​​defined in the preset fault impact level rules, the high-weight causal chains are matched and filtered to obtain the candidate main causal chains of the relay control circuit. Based on the influence weight distribution and the fault influence level classification, the root cause determination of the candidate main causal chain is performed to obtain the main causal chain of the relay control circuit fault.

[0048] A comprehensive analysis of all key attributes involved in the complete causal chain is conducted. These attributes include the propagation coverage, the total number of fault nodes, and the speed-related attributes of fault propagation. For each attribute, corresponding quantitative standards are developed based on its actual importance in relay control circuit fault diagnosis. For the propagation coverage, specific values ​​are determined according to the number of circuit functional modules it covers; for the total number of fault nodes, a direct count is performed and converted into a corresponding quantitative value; for the speed-related attributes of fault propagation, the propagation time in historical fault data is referenced and converted into a quantifiable indicator. Through this comprehensive analysis and targeted quantitative conversion, evaluation parameters that can fully reflect the characteristics of all aspects of the complete causal chain are obtained.

[0049] The acquired evaluation parameters are categorized by dimension, such as parameters related to the degree of fault impact, parameters related to propagation characteristics, and parameters related to the importance of associated nodes. Based on the specific number of parameters in each category and the relationships between them, the dimensionality of the feature tensor is determined. Evaluation parameters within the same category are arranged in a predetermined order, and parameters from different categories are assigned to different dimensions. Then, all evaluation parameters are individually filled into the designated positions of the corresponding dimensions of the feature tensor, completing the construction of the feature tensor. Through this structured integration operation, a fused feature vector capable of comprehensively carrying all evaluation parameter information is formed.

[0050] The preset fault impact level rules include various feature standards corresponding to different impact levels, covering the numerical range of evaluation parameters in each dimension, the proportional relationship between different parameters, etc. When performing rule matching, the parameter values ​​of each dimension and the interrelationship between parameters are first extracted from the fused feature vector. In order from high impact level to low impact level, the extracted information is compared and checked one by one with each standard in the rule. It is checked whether the parameter value of each dimension is within the numerical range defined by the corresponding level rule, and whether the proportional relationship between parameters meets the rule requirements. If the complete causal chain corresponding to a certain fused feature vector satisfies all the standards of a certain level rule, then it is determined that the complete causal chain matches the level rule, and the corresponding matching level is recorded, thus obtaining the rule matching result of the complete causal chain.

[0051] A pre-defined weight scale standard is set to correspond one-to-one with each fault impact level. Each impact level corresponds to a fixed weight range, and the higher the impact level, the larger the corresponding weight range value. When performing weight scale mapping, the specific impact level corresponding to the rule matching result is first determined. Then, according to the weight scale standard corresponding to the level, the weight value corresponding to the complete causal chain is determined. If the matching result is in the middle range of a certain impact level, the specific weight value is accurately determined according to the preset weight allocation ratio within that range. Through this correspondence and conversion, the impact weight value corresponding to each complete causal chain is obtained, and then integrated to form the impact weight distribution of all complete causal chains.

[0052] First, summarize the influence weight values ​​corresponding to all complete causal chains and establish a one-to-one correspondence between complete causal chains and weight values. Arrange all complete causal chains in descending order of weight values. The larger the weight value, the earlier the complete causal chain is arranged, and the higher its priority level. If there are complete causal chains with the same weight value, further consider their corresponding fault propagation range and the number of core circuit modules involved. Causal chains with a wider propagation range and more core modules involved are given priority. Through this orderly arrangement, the priority level of each complete causal chain is determined, and then the causal chain with the highest priority level is selected as the main causal chain of relay control circuit fault.

[0053] Extract the weight value corresponding to each complete causal chain in the influence weight distribution, and record the one-to-one correspondence between each complete causal chain and its weight value. Using the weight value as the core sorting criterion, start with the weight with the largest value and arrange the corresponding complete causal chains in order, followed by the complete causal chains with smaller values. If two or more complete causal chains have the same weight value, arrange them according to their order in the generation sequence of the complete causal chains. Finally, generate a weight sorting list containing all complete causal chains and their corresponding weight values, arranged in descending order of weight.

[0054] Based on the actual needs of relay control circuit fault diagnosis and the statistical results of historical fault data, the range of high weight intervals is determined in advance. Usually, this range is set to the interval corresponding to the top 30% of the weight values ​​in the weight ranking list. During the extraction process, the weight value of each complete causal chain in the weight ranking list is checked one by one to determine whether it is within the preset high weight interval. If the weight value of the complete causal chain is greater than or equal to the lowest value of the high weight interval, the complete causal chain is filtered out and organized into a high weight causal chain set.

[0055] The thresholds defined in the preset fault impact level rules include multiple dimensions such as weight threshold, fault propagation depth threshold, and core node association threshold. During the matching and filtering process, each indicator of a high-weight causal chain is checked one by one to see if its weight value reaches the preset weight threshold, whether the fault propagation depth meets the threshold requirement, and whether the number of associated core nodes exceeds the core node association threshold. Only when all indicators of a high-weight causal chain meet the corresponding threshold requirements can it be retained. After checking and filtering all high-weight causal chains one by one, a set of candidate main causal chains for relay control circuits is finally formed.

[0056] Examine the specific weight value of each candidate main causal chain in the influence weight distribution to clarify its weight ranking among all causal chains. At the same time, combine the fault influence level classification to determine the influence level corresponding to each candidate main causal chain. Conduct in-depth analysis of the propagation path of each candidate main causal chain, trace the fault root cause it points to, and determine whether the root cause is the initial core root cause in the entire fault propagation process and whether it can fully cover the main propagation path of the fault. Compare the weight ranking, influence level, and root cause tracing ability of all candidate main causal chains, and select the candidate main causal chain with the highest weight ranking, the highest influence level, and the most accurate pointing to the initial core root cause, and determine it as the main causal chain of the relay control circuit fault.

[0057] The beneficial effects are as follows: It sorts out and quantifies the key attributes of the complete causal chain, ensuring that the evaluation parameters fully reflect its characteristics, providing a reliable basis for priority assessment, avoiding evaluation bias; it integrates evaluation parameters to construct a feature tensor, allowing the fused feature vector to fully carry information and avoid loss, while providing a clear structure for rule matching, improving processing efficiency and accuracy; it compares against preset fault impact level rules, ensuring consistent rule matching standards and avoiding subjective bias; it quickly locates the impact level of the causal chain, providing a basis for weight scaling mapping, improving evaluation efficiency; it converts the impact level into specific weight values, achieving quantitative representation, allowing for intuitive comparison of the impact degree of different causal chains, providing a basis for priority arrangement, ensuring objective and accurate evaluation; and it arranges priorities by combining weights, propagation scope, and the number of core module associations, ensuring reasonable ranking. The system highlights the most influential causal chain; the main causal chain reflects the core path of the fault, improving the targeting and efficiency of diagnosis. It organizes the chains into a list according to their weight from highest to lowest, making the weight relationships of each causal chain clear, facilitating the rapid identification of high-weight chains, reducing screening operations, and improving processing efficiency. It presets high-weight intervals to screen causal chains, quickly focusing on chains with significant impact, eliminating interference from low-weight chains, reducing subsequent workload, and improving the targeting and effectiveness of the work. Multi-dimensional threshold comparisons are used to screen high-weight causal chains, eliminating chains that do not meet key indicators, ensuring that candidate main causal chains meet diagnostic requirements, and improving the accuracy of main causal chain determination. A comprehensive judgment is made based on weight, impact level, and root cause tracing capability, ensuring that the main causal chain has a significant impact and points to the core root cause of the fault, providing crucial evidence for subsequent work and improving the efficiency and accuracy of fault handling.

[0058] S5. Based on the main causal chain, the relay control circuit is divided into logical segments to obtain the segmented logical structure of the relay control circuit fault. In this embodiment of the invention, the step of dividing the relay control circuit into logical segments based on the main causal chain to obtain a segmented logical structure for the fault of the relay control circuit includes: Identify the logical boundaries in the main causal chain to obtain the logical splitting points of the main causal chain; Based on the logical segmentation points, the main causal chain is logically segmented to obtain the initial logical segments of the main causal chain; The initial logic paragraph is subjected to semantic consistency verification to obtain the logic paragraph after verification of the relay control circuit fault. Based on the topological relationships in the fault knowledge graph, the verified logic segments are integrated to obtain the segmented logic structure of the relay control circuit fault.

[0059] The main causal chain reflects the complete sequence of relationships from the root cause to the final symptom of a relay control circuit fault. Logical boundaries are the natural divisions between different functional modules and fault propagation stages within the main causal chain. During identification, it is necessary to consider the actual structural characteristics of the relay control circuit, including the connection relationships of components, the path of signal transmission, and the changes in impact during fault propagation. When the degree of impact of fault propagation shows a significant turning point, or when the involved circuit functional modules switch, these key locations are the logical boundaries. For example, nodes where a power module fault propagates to the control module, and transition nodes where a transient fault transforms into a persistent fault, are all logical boundaries. By accurately defining these key locations in the main causal chain, specific points are determined as logical dividing points, ensuring that each dividing point accurately separates different fault logic units.

[0060] Based on the established logical segmentation points, the main causal chain is split according to the chronological order of fault propagation. During the splitting process, the defined scope of each segmentation point must be strictly adhered to, ensuring that each split part fully reflects the fault correlation within a specific stage or functional module. For example, the propagation process from the fault's root cause to the first intermediate symptom can be considered as one independent part, the propagation process from that intermediate symptom to the next level symptom as another independent part, and so on. Each independent part formed after splitting possesses relatively complete fault logic, i.e., an initial logical segment, and each segment maintains the continuity and correlation of fault propagation.

[0061] The core of semantic consistency verification is to check whether the fault descriptions and relationships between fault nodes within each initial logic segment are coherent and consistent, and whether they conform to the working principles of relay control circuits and the objective laws of fault propagation. During the verification process, each initial logic segment must be checked one by one to confirm whether the connection relationships between fault nodes within the segment are reasonable, whether the fault propagation process is smooth, and whether there are any logical contradictions or broken relationships. For example, if an initial segment describes the propagation process of signal abnormalities caused by component damage, the actual role of the component in the circuit and the normal path of signal transmission must be verified to ensure that the described propagation process matches the actual operation of the circuit. If logical contradictions or uncoordinated relationships are found within a segment, the segment's scope must be adjusted or the relationships between fault nodes corrected in a timely manner. After a comprehensive check and confirmation that there are no logical problems, the verified logic segment is formed.

[0062] The topological relationships of the fault knowledge graph clearly present the connection methods, correlation strength, and hierarchical distribution among all fault entities. During integration, this topological relationship must be used as the core basis to reasonably arrange and combine the verified logical segments according to the natural order of fault propagation and the tightness of their correlation. For example, based on the hierarchical relationship of fault nodes in the topological relationship, verified logical segments belonging to the same level can be grouped together; or, according to the connection path order of nodes in the topological relationship, the verified logical segments can be arranged sequentially. During integration, it is necessary to ensure that the transitions between segments are natural and smooth, conforming to the connection logic of fault entities in the topological relationship, ultimately forming a segmented logical structure that is hierarchical, logically coherent, and can completely present the entire process of the fault from its root cause to its final symptoms.

[0063] The beneficial effects are: accurately identifying logical boundaries and split points, avoiding logical confusion in faults, providing a reliable basis for subsequent segmentation, laying the foundation for fault analysis, splitting the main causal chain into logically independent initial segments, reducing the difficulty of analysis, providing a clear basic unit for subsequent verification and integration, eliminating logical contradictions and broken connections in the initial segments, improving segment accuracy, avoiding affecting the accuracy of fault diagnosis, and integrating to form an orderly and coherent segmented logical structure that retains the independence of segments while ensuring the integrity of fault logic, thereby improving the efficiency of fault diagnosis.

[0064] S6. Perform a visual interactive design on the segmented logic structure to obtain segmented display data of the relay control circuit fault and detailed diagnostic information that can be triggered. In this embodiment of the invention, the step of performing a visual interactive design on the segmented logic structure to obtain segmented display data of the relay control circuit fault and triggerable detailed diagnostic information includes: The segmented logical structure is mapped to visual elements to obtain the visual element configuration of the segmented logical structure. Based on the configuration of the visual elements, the segmented logical structure is hierarchically laid out to obtain the topological layout scheme of the segmented logical structure. An interactive response mechanism is designed for the topology layout scheme to obtain the interactive response rules of the topology layout scheme; Based on the interactive response rules, the segmented logic structure is encapsulated for display data to obtain segmented display data of the relay control circuit fault and detailed diagnostic information that can be triggered.

[0065] The entire logical structure is organized by identifying all its constituent modules, clarifying the functional attributes of each module and the distinguishing requirements between them. For each module's characteristics, a suitable presentation format is selected. For example, the module responsible for identifying the root cause of a problem uses a specific graphic style to reflect its core position; the module responsible for conveying symptoms uses a unique line style to illustrate the transmission process; and the modules that serve as logical boundaries use unique color markings to clearly define their scope. In this way, a fixed correspondence is established between each module and its corresponding visual element, ultimately resulting in a complete configuration of visual elements.

[0066] Based on the established configuration of visualization elements, we first analyze the main causal chain relationship of the segmented logical structure. The key logical segments in the main causal chain are designated as the core level, and this content is placed in the most easily visible core area of ​​the visualization interface. Then, the remaining supplementary logical segments for auxiliary explanations are classified as secondary levels and arranged logically around the core level. Simultaneously, based on the tightness of the relationships between each logical segment, the spacing and arrangement order within and between each level are flexibly adjusted to clearly demonstrate the inherent connections and hierarchical differences of the segmented logical structure, ultimately forming a complete topological layout.

[0067] This document comprehensively analyzes the types of user actions that might occur when viewing the fault visualization interface, covering common actions such as clicking, hovering, and dragging. For each action, and in conjunction with different visualization elements in the topology layout scheme, the corresponding feedback content is clearly defined. For example, when a user clicks the visualization element corresponding to the root cause of the fault, the interface immediately displays the detailed formation process of that root cause; when a user hovers the mouse over the visualization element corresponding to a symptom, the interface simultaneously presents the specific manifestations of that symptom and its related logical relationships. Furthermore, the triggering conditions for each action are clearly defined to ensure that feedback is only triggered when the action meets the preset scenario, and reasonable response timelines are set to ensure timely feedback, ultimately forming a complete and feasible set of interactive response rules.

[0068] First, basic display data for each module in the segmented logical structure is collected, including a detailed description of each logical segment and the relationships between modules. Next, various feedback data defined in the interaction response rules are precisely bound to the corresponding visual elements, ensuring that specific elements trigger specific operations and can access the corresponding feedback data. Finally, in accordance with visualization requirements, all basic display data and bound feedback data are formatted uniformly to form a structured dataset. This ensures that the encapsulated data can be accurately recognized and quickly accessed by the visualization system, enabling timely output of corresponding segmented display content and accurate triggering of detailed diagnostic information when the user performs relevant operations.

[0069] The beneficial effects include: clarifying the execution basis of visualization presentation, improving visualization recognition, helping users quickly distinguish different functional modules in the fault logic, clearly presenting the hierarchy and relationship of segmented logical structure, avoiding interface clutter, improving information reading efficiency, helping users quickly grasp the overall framework of fault logic, breaking the limitations of traditional passive visualization viewing, enhancing user participation and operational convenience, helping users deeply understand the ins and outs of fault logic, achieving organic data integration, ensuring the stable operation of visualization interactive functions, and providing users with a smooth fault viewing and diagnosis experience.

[0070] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0071] This application embodiment can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence is the theory, method, technology, and application system that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0072] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended 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.

Claims

1. A method for segmented display of relay control circuit faults, characterized in that, The method includes: Using the fault propagation logic in the relay control circuit knowledge base as a framework, the fault entities in the relay control circuit knowledge base as nodes, and the propagation relationships in the relay control circuit knowledge base as edges, a fault knowledge graph of the relay control circuit is constructed. Based on the abnormal characteristics in the relay control circuit, bidirectional graph reasoning is performed on the fault knowledge graph to obtain the potential subsequent symptoms and potential root causes of the relay control circuit fault, including: The abnormal features in the relay control circuit are analyzed to obtain the set of associated nodes in the fault knowledge graph; Bidirectional graph reasoning is performed on the set of associated nodes to obtain the propagation influence value of candidate nodes in the fault knowledge graph. The formula for calculating the propagation influence value is as follows: ； In the formula, This represents the propagation influence value of the candidate node. This refers to the set of associated nodes in the fault knowledge graph. For the set of associated nodes Size, For the set of associated nodes Each symptom node in the data, For the fault knowledge graph from candidate nodes To the symptom node The shortest path distance, This is a decay factor used to control the rate of decay of propagation influence. Candidate nodes To the symptom node The strength of the propagation relationship between them It is an exponentially decaying function; Based on the propagation influence value, a forward priority traversal is performed on the fault knowledge graph to obtain the subsequent node sequence of the fault knowledge graph. Based on the propagation influence value, the fault knowledge graph is traversed in reverse priority to obtain the predecessor node sequence of the fault knowledge graph. Extract the origin entity from the predecessor node sequence and the representation entity from the subsequent node sequence, respectively; Based on the propagation influence value, the symptom intensity of the characterized entity is assessed to obtain candidate symptoms of the relay control circuit fault. Based on the propagation influence value, the root cause contribution of the originating entity is determined to obtain the candidate root causes of the relay control circuit fault. Verify the correlation logic of the propagation path in the candidate symptoms to obtain the potential subsequent symptoms of the relay control circuit failure; By verifying the reverse causal logic of the candidate root causes, the potential root causes of the relay control circuit failure can be obtained. By synthesizing the potential subsequent symptoms and the potential root causes, a complete causal chain of the relay control circuit failure is obtained. Based on the complete causal chain, the link priority of the relay control circuit is evaluated to obtain the main causal chain of the relay control circuit fault. Based on the main causal chain, the relay control circuit is divided into logical segments to obtain the segmented logical structure of the relay control circuit fault. A visual interactive design is performed on the segmented logic structure to obtain segmented display data of faults in the relay control circuit and detailed diagnostic information that can be triggered.

2. The method for segmented display of relay control circuit faults as described in claim 1, characterized in that, The method uses the fault propagation logic in the relay control circuit knowledge base as a framework, fault entities in the relay control circuit knowledge base as nodes, and propagation relationships in the relay control circuit knowledge base as edges to construct a fault knowledge graph for the relay control circuit, including: Extract the fault propagation logic from the relay control circuit knowledge base to obtain the fault propagation logic framework of the relay control circuit. Based on the fault propagation logic framework, fault entity identification is performed on the relay control circuit knowledge base to obtain the fault entity nodes of the relay control circuit. Based on the propagation relationships in the relay control circuit knowledge base, construct the propagation relationship edges of the relay control circuit; Based on the fault propagation logic framework, the fault entity nodes and the propagation relationship edges are integrated into a graph to obtain the fault knowledge graph of the relay control circuit.

3. The method for segmented display of relay control circuit faults as described in claim 1, characterized in that, The causal path synthesis of the potential subsequent symptoms and the potential root cause to obtain the complete causal chain of the relay control circuit fault includes: A reverse search is performed on the potential subsequent symptoms to obtain the source path of the potential subsequent symptoms; A forward search is performed on the potential root cause to obtain the endpoint path of the potential root cause; By connecting the associated paths between the source path and the destination path, a causal relationship network of the relay control circuit fault is obtained. By eliminating redundant and secondary paths in the causal relationship network, the complete causal chain of the relay control circuit fault is obtained.

4. The method for segmented display of relay control circuit faults as described in claim 3, characterized in that, The process of eliminating redundant and secondary paths in the causal network to obtain the complete causal chain of the relay control circuit fault includes: Based on the causal relationship network, the propagation relationship in the relay control circuit knowledge base is intensity quantized to obtain the path weight distribution of the causal relationship network; Extracting low-weight paths from the path weight distribution yields candidate redundant paths for the causal relationship network; Verify the critical connectivity of the candidate redundant paths to obtain the redundant paths of the causal relationship network; By removing redundant paths from the causal relationship network, the complete causal chain of the relay control circuit fault is obtained.

5. The method for segmented display of relay control circuit faults as described in claim 2, characterized in that, The step of evaluating the link priority of the relay control circuit based on the complete causal chain to obtain the main causal chain of the relay control circuit fault includes: The complete causal chain is subjected to comprehensive feature quantification to obtain the evaluation parameters of the complete causal chain; The evaluation parameters are used to construct a feature tensor to obtain the fused feature vector of the complete causal chain; Based on the preset fault impact level rules, the fused feature vector is subjected to rule matching to obtain the rule matching result of the complete causal chain; The influence weight distribution of the complete causal chain is obtained by performing weight scaling mapping on the rule matching results. Based on the influence weight distribution, the complete causal chain is prioritized to obtain the main causal chain of the relay control circuit fault.

6. The method for segmented display of relay control circuit faults as described in claim 5, characterized in that, The prioritization of the complete causal chain based on the influence weight distribution yields the main causal chain of the relay control circuit fault, including: The influence weight distribution is sorted in descending order to obtain the weight ranking list of the complete causal chain; Extract the high-weight causal chains from the weighted list; Based on the threshold values ​​defined in the preset fault impact level rules, the high-weight causal chains are matched and filtered to obtain the candidate main causal chains of the relay control circuit. Based on the influence weight distribution and the fault influence level classification, the root cause determination of the candidate main causal chain is performed to obtain the main causal chain of the relay control circuit fault.

7. The method for segmented display of relay control circuit faults as described in claim 1, characterized in that, The step of dividing the relay control circuit into logical segments based on the main causal chain to obtain a segmented logical structure for relay control circuit faults includes: Identify the logical boundaries in the main causal chain to obtain the logical splitting points of the main causal chain; Based on the logical segmentation points, the main causal chain is logically segmented to obtain the initial logical segments of the main causal chain; The initial logic paragraph is subjected to semantic consistency verification to obtain the logic paragraph after verification of the relay control circuit fault. Based on the topological relationships in the fault knowledge graph, the verified logic segments are integrated to obtain the segmented logic structure of the relay control circuit fault.

8. The method for segmented display of relay control circuit faults as described in claim 1, characterized in that, The visualization and interactive design of the segmented logic structure yields segmented display data and triggerable detailed diagnostic information for faults in the relay control circuit, including: The segmented logical structure is mapped to visual elements to obtain the visual element configuration of the segmented logical structure. Based on the configuration of the visual elements, the segmented logical structure is hierarchically laid out to obtain the topology layout scheme of the segmented logical structure. An interactive response mechanism is designed for the topology layout scheme to obtain the interactive response rules of the topology layout scheme; Based on the interactive response rules, the segmented logic structure is encapsulated for display data to obtain segmented display data of the relay control circuit fault and detailed diagnostic information that can be triggered.

Images