Equipment fault processing method and device, equipment and medium

Abstract

The invention relates to the technical field of rail transit equipment operation and maintenance, and discloses an equipment fault processing method and device applied to rail transit equipment operation and maintenance, equipment and a medium, and the method comprises the steps: obtaining an unstructured equipment operation and maintenance document, and constructing the equipment operation and maintenance document into a structured standard operation and maintenance document according to a preset standardized knowledge template; based on the natural language processing function of the large model, equipment fault related information multi-tuples are extracted from the standard operation and maintenance document, workflow data capable of being called are generated based on the information multi-tuples, and equipment fault related information comprises fault phenomena and fault handling operation steps; and receiving a fault phenomenon description input by the operation and maintenance personnel, performing retrieval matching in the workflow data based on the fault phenomenon description by using the large model, and obtaining a recommended fault handling scheme based on the matched information for the operation and maintenance personnel to perform fault handling based on the recommended fault handling scheme. The labor cost can be effectively reduced, and the equipment operation and maintenance efficiency is greatly improved.

Description

Technical Field

[0001] The present invention relates to the technical field of rail transit equipment operation and maintenance, and in particular to an equipment fault processing method, device, equipment and medium. Background Art

[0002] In the field of equipment operation and maintenance in the rail transit industry, equipment failure handling has always faced severe challenges.

[0003] The industry currently uses paper or electronic documents as troubleshooting guides. While these documents contain a wealth of troubleshooting knowledge, they are typically written in free text format, with the relationship between fault symptoms and corresponding treatment steps implicitly embedded in natural language paragraphs. When troubleshooting, operations and maintenance personnel often spend a significant amount of time manually searching and analyzing relevant information in the documents, resulting in low operational efficiency. Furthermore, new personnel require training, leading to high labor costs. Summary of the Invention

[0004] In view of this, the present invention provides a method, apparatus, device and medium for handling equipment failures, which can effectively reduce labor costs while greatly improving equipment operation and maintenance efficiency.

[0005] In a first aspect, the present invention provides a method for handling equipment failures, which is applied to the operation and maintenance of rail transit equipment. The data processing method includes:

[0006] Obtaining unstructured equipment operation and maintenance documents, and constructing the equipment operation and maintenance documents into structured standard operation and maintenance documents according to a preset standardized knowledge template;

[0007] Extracting equipment failure-related information tuples from the standard operation and maintenance documents based on the natural language processing function of the large model, and generating callable workflow data based on the extracted information tuples; wherein the equipment failure-related information includes the failure phenomenon and the failure handling operation steps;

[0008] Receive the fault phenomenon description input by the operation and maintenance personnel, use the big model to search for matches in the workflow data based on the fault phenomenon description, and obtain a recommended fault handling solution based on the matched information, so that the operation and maintenance personnel can handle the fault based on the recommended fault handling solution.

[0009] In an optional embodiment, the equipment operation and maintenance document is constructed into a structured standard operation and maintenance document according to a preset standardized knowledge template, including:

[0010] Retrieve the current standardized knowledge template; wherein the standardized knowledge template defines a hierarchical knowledge structure, with titles / texts at different levels corresponding to the fault phenomenon, fault identification problem, and fault handling steps of the equipment fault;

[0011] The content of the equipment operation and maintenance document is added to the corresponding position of the standardized knowledge template according to the titles / texts of different levels in the hierarchical knowledge structure to obtain a standard operation and maintenance document.

[0012] In an optional embodiment, a natural language processing function based on a large model extracts a tuple of information related to equipment failure from the standard operation and maintenance document, including:

[0013] The natural language processing function of the large model is used to identify the correspondence between each title and the corresponding text in the standard operation and maintenance document, and based on the correspondence, triples containing fault phenomena, fault identification problems and fault handling operation steps are extracted.

[0014] In an optional embodiment, generating workflow data that can be called based on the extracted information tuple includes:

[0015] Based on the extracted information tuple, the fields of the fault node, the associated parent node, the fault handling parameters and the fault handling operation steps are determined, and based on the determined fields, a structured knowledge graph that can be called is generated as workflow data.

[0016] In an optional embodiment, using the large model to search for matches in the workflow data based on the fault phenomenon description, and obtaining a recommended fault handling solution based on the matched information, includes:

[0017] The large model is used to retrieve and match corresponding fault nodes in the workflow data based on the fault phenomenon description, and the associated parent node, fault handling parameters and fault handling operation steps are determined based on the fault node, and a fault handling plan containing the determined data is generated.

[0018] In an optional embodiment, after generating the callable workflow data based on the extracted information tuple, the method further includes:

[0019] The various data in the workflow data are displayed in a visual flow chart on a preset page, and the operation and maintenance experts are accepted to dynamically arrange the displayed various data, and the work data flow is updated based on the accepted dynamic arrangement.

[0020] In an optional embodiment, before the preset page displays the data in the workflow data in the form of a visual flowchart, the method further includes:

[0021] Receive a display instruction and verify whether the identity information contained in the display instruction is legal. If it is legal, execute the step of displaying each data in the workflow data in the form of a visual flowchart on a preset page. Otherwise, display the information that you are not authorized to update the workflow data flow.

[0022] The device fault handling method provided by the present invention has the following advantages:

[0023] By forcing the template to structure and construct workflow data, the automatic conversion of unstructured equipment operation and maintenance documents into structured data is achieved, enabling the machine to identify fault handling logic and greatly shortening the time for matching fault phenomena with corresponding handling operation steps. In addition, the semantic understanding ability of the large model can process natural language input and achieve accurate matching of fuzzy descriptions, thereby reducing the need for manual experience and allowing new employees to quickly obtain fault handling solutions. It can be seen that the present invention can effectively reduce labor costs while greatly improving equipment operation and maintenance efficiency.

[0024] In a second aspect, the present invention provides an equipment fault processing device, which is applied to rail transit equipment operation and maintenance; the data processing device includes:

[0025] A standard document construction module is used to: obtain unstructured equipment operation and maintenance documents, and construct the equipment operation and maintenance documents into structured standard operation and maintenance documents according to a preset standardized knowledge template;

[0026] A workflow generation module is configured to extract information tuples related to equipment failures from the standard operation and maintenance documents based on the natural language processing function of the large model, and generate workflow data that can be called based on the extracted information tuples; wherein the information related to equipment failures includes the failure symptoms and the failure handling steps;

[0027] The intelligent interaction module is used to: receive a fault phenomenon description input by an operation and maintenance personnel, use the large model to search for matches in the workflow data based on the fault phenomenon description, and obtain a recommended fault handling solution based on the matched information, so that the operation and maintenance personnel can handle the fault based on the recommended fault handling solution.

[0028] The device for handling equipment failures provided by the present invention has the following advantages:

[0029] By forcing the template to structure and construct workflow data, the automatic conversion of unstructured equipment operation and maintenance documents into structured data is achieved, enabling the machine to identify fault handling logic and greatly shortening the time for matching fault phenomena with corresponding handling operation steps. In addition, the semantic understanding ability of the large model can process natural language input and achieve accurate matching of fuzzy descriptions, thereby reducing the need for manual experience and allowing new employees to quickly obtain fault handling solutions. It can be seen that the present invention can effectively reduce labor costs while greatly improving equipment operation and maintenance efficiency.

[0030] In a third aspect, the present invention provides a computer device comprising: a memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, and the processor executing the device fault handling method of the first aspect or any corresponding embodiment thereof by executing the computer instructions.

[0031] In a fourth aspect, the present invention provides a computer-readable storage medium having computer instructions stored thereon, the computer instructions being used to enable a computer to execute the device fault handling method of the first aspect or any corresponding embodiment thereof. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the specific embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0033] Figure 1 is a flow chart of a method for handling equipment failures according to an embodiment of the present invention;

[0034] Figure 2 This is a structural block diagram of a device failure processing apparatus according to an embodiment of the present invention;

[0035] Figure 3 Schematic diagram of the hardware structure of a computer device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0036] To make the purpose, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making creative efforts shall fall within the scope of protection of the present invention.

[0037] In the existing technology, the field of rail transit equipment operation and maintenance has long relied on paper or electronic documents as troubleshooting guides. These documents use free text format to describe fault phenomena and processing steps. The correlation between fault phenomena and corresponding processing steps is implicit in natural language paragraphs. Operation and maintenance personnel need to manually retrieve and analyze this information before they can handle equipment failures based on it. New employees need to be trained before they can complete equipment operation and maintenance work, resulting in high labor costs and low equipment maintenance efficiency.

[0038] In order to solve the above technical problems, the embodiment of the present invention provides a method for handling equipment failures, which is applied to the operation and maintenance of rail transportation equipment; Figure 1 As shown, the data processing method may specifically include:

[0039] S11: Obtain unstructured equipment operation and maintenance documents, and construct the equipment operation and maintenance documents into structured standard operation and maintenance documents according to a preset standardized knowledge template.

[0040] Unstructured equipment operation and maintenance documents refer to text data that is not organized in a machine-readable format. Specifically, they can be implemented as troubleshooting guides in Word or PDF format or other preset formats. Their free text characteristics make it difficult for machines to recognize the correspondence between fault phenomena and fault handling steps. The preset standardized knowledge template refers to a formatting framework that defines a hierarchical knowledge structure. Specifically, it can be implemented using a tree-like hierarchical template that includes fault phenomena, fault identification problems, and fault handling steps. Text differences are eliminated by forcing a unified document format. Among them, the fault phenomenon may include the specific manifestations of the fault, the fault identification problem may include the faulty equipment determined based on the fault phenomenon, the specific problem of the fault, and the cause of the fault. The fault handling steps are the various steps required to troubleshoot the fault.

[0041] S12: extracting information tuples related to equipment failure from the standard operation and maintenance documents based on the natural language processing function of the large model, and generating workflow data that can be called based on the extracted information tuples; wherein, the information related to equipment failure includes the failure phenomenon and the failure handling operation steps.

[0042] A large language model (LLM) refers to an artificial intelligence model trained with massive data and a large parameter scale. Its core is natural language processing (NLP) technology based on deep learning. An information tuple refers to a structured data unit that describes fault-related information. Fault-related information includes but is not limited to fault phenomena, fault identification problems, and fault handling steps. The embodiment of the present invention uses the NLP capability of the large model to analyze standard operation and maintenance documents, extract information tuples related to equipment failures contained therein, and support machine calls by explicitly expressing fault handling logic. Workflow data refers to a structured knowledge set that can be called by the system. Specifically, it can be implemented using a knowledge graph containing information related to equipment failures, thereby supporting intelligent retrieval by establishing associations between information related to equipment failures.

[0043] S13: Receive the fault phenomenon description input by the operation and maintenance personnel, use the large model to search for matches in the workflow data based on the fault phenomenon description, and obtain a recommended fault handling solution based on the matched information, so that the operation and maintenance personnel can handle the fault based on the recommended fault handling solution.

[0044] The system receives a fault phenomenon described by an operator in natural language, uses a large model to perform semantic parsing and extract key information from the described fault phenomenon, and obtains relevant information about the fault phenomenon. It then searches and matches the obtained fault phenomenon information in the workflow data. Finally, based on the matched information, it obtains a recommended fault resolution plan, which is then fed back to the operator, allowing them to resolve the fault based on the obtained fault resolution plan. The system also records the operator's actual feedback on the fault resolution process, which can be used to update and optimize the relevant information in the workflow data.

[0045] Specifically, the system first converts unstructured equipment operation and maintenance documents into structured, standard operation and maintenance documents according to standardized knowledge templates, eliminating differences in text formatting. The system then uses the big model to identify the corresponding relationships between the titles and text in the standard operation and maintenance documents, extracting information tuples related to equipment failures. These information tuples are then converted into workflow data that supports invocation. When the operator enters a description of the fault phenomenon, the big model retrieves matching equipment failure-related information from the workflow data and generates a fault resolution plan based on the matched information. This solution replaces manual retrieval with semantic matching, enabling automated guidance of the fault handling process.

[0046] The technical solution of the present invention realizes the automatic conversion of unstructured equipment operation and maintenance documents into structured data through template-enforced structuring and workflow data construction, enabling the machine to identify fault handling logic and greatly shortening the matching time between fault phenomena and corresponding handling operation steps; in addition, the semantic understanding ability of the large model can process natural language input and realize accurate matching of fuzzy descriptions, thereby reducing manual experience and allowing new employees to quickly obtain fault handling solutions; it can be seen that the present invention can effectively reduce labor costs while greatly improving equipment operation and maintenance efficiency.

[0047] An embodiment of the present invention provides a method for handling equipment failures, wherein constructing the equipment operation and maintenance document into a structured standard operation and maintenance document according to a preset standardized knowledge template may include:

[0048] Retrieve the current standardized knowledge template; wherein the standardized knowledge template defines a hierarchical knowledge structure, with titles / texts at different levels corresponding to the fault phenomenon, fault identification problem, and fault handling steps of the equipment fault;

[0049] The content of the equipment operation and maintenance document is added to the corresponding position of the standardized knowledge template according to the titles / texts of different levels in the hierarchical knowledge structure to obtain a standard operation and maintenance document.

[0050] A standardized knowledge template is a digital template with a predefined document format framework. This can be implemented using a fixed Word outline title format template. This template establishes a machine-readable knowledge framework by establishing corresponding relationships between titles and text at different levels. A hierarchical knowledge structure is a multi-level title system divided by logical relationships. This can be implemented by having first-level titles correspond to fault symptoms, second-level titles correspond to fault identification questions, and text correspond to fault handling steps. This makes fault handling logic explicit through hierarchical nesting.

[0051] In a specific implementation, the standardized knowledge template is a pre-designed fixed outline title format template, which defines the first-level title as a description of the specific fault phenomenon of the operation and maintenance equipment (i.e., the fault phenomenon), the second-level title as the fault identification problem of the operation and maintenance equipment (i.e., the fault identification problem), and the main text content as detailed operation steps for structured operation and maintenance equipment troubleshooting (i.e., the fault handling operation steps). When the equipment operation and maintenance document is imported, the free text content in the original equipment operation and maintenance document is disassembled and reorganized according to the hierarchical relationship defined by the template; for example, the paragraph describing the fault phenomenon is filled in the first-level title position, the paragraph corresponding to the fault identification problem is filled in the second-level title position, and the fault handling operation steps are filled in the main text area, forming a clear mapping relationship between the fault phenomenon, the fault identification problem and the fault handling operation steps, laying a structured foundation for the automatic generation of workflow data. This template-based forced structuring process converts the fault handling logic originally implied in the natural language paragraph into machine-parseable data with clear hierarchical labels, establishing a unified input standard for subsequent information extraction.

[0052] This embodiment enforces the structure of equipment operation and maintenance documents by pre-defining hierarchical standard knowledge templates. This converts unstructured text into structured data with a clear title hierarchy, achieving standardized reorganization of fault knowledge and making fault handling logic explicit in a machine-readable manner. This effectively addresses the issue of implicit correspondence between fault symptoms and handling steps caused by unstructured documents, significantly improving the efficiency and accuracy of subsequent large-scale model analysis of fault handling logic and reducing the cost of manual intervention.

[0053] An embodiment of the present invention provides a method for handling equipment failures, which extracts information tuples related to equipment failures from the standard operation and maintenance documents based on the natural language processing function of a large model, and may include:

[0054] The natural language processing function of the large model is used to identify the correspondence between each title and the corresponding text in the standard operation and maintenance document, and based on the correspondence, triples containing fault phenomena, fault identification problems and fault handling operation steps are extracted.

[0055] The large model's natural language processing function refers to semantic understanding based on a deep learning architecture. This is achieved by combining a pre-trained language model with sequence tagging technology. By analyzing the grammatical structure and contextual semantic relationships within a document, it identifies the logical relationship between the title hierarchy and the main text. The corresponding relationship between titles and their corresponding main text refers to the subordinate relationship between different levels of titles and their subordinate text paragraphs in standard operation and maintenance documents. This is achieved using a hierarchical relationship parsing algorithm based on an attention mechanism, which establishes a mapping relationship by calculating the semantic similarity between titles and paragraphs. A triple is a logical unit consisting of three elements: a fault phenomenon, a fault identification question, and a fault resolution procedure. This is achieved by combining pattern matching-based entity extraction technology with a relational classification model to extract structured data triples from standard operation and maintenance documents. For example, in a train brake system fault document, the "brake pressure abnormality" phenomenon, its corresponding "pressure sensor calibration method" identification question, and the "perform sensor calibration procedure" procedure can be automatically integrated into a complete triple, providing an accurate data foundation for the subsequent generation of visualization workflows.

[0056] Specifically, the pre-trained language model is used to perform semantic analysis on the standard operation and maintenance documents, and the semantic boundaries of different levels of titles in the document and their corresponding text paragraphs are identified. Sequence labeling technology is used to classify titles hierarchically. For example, the first-level titles are labeled as fault phenomenon categories, the second-level titles are labeled as fault identification problem categories, and the third-level titles are labeled as fault handling operation step categories. The semantic correlation between each title and its subordinate paragraphs is calculated based on the attention mechanism, and a mapping relationship between the title hierarchy and the text content is established. The relational classification model is used to identify the logical association between fault phenomena and fault identification problems, and the corresponding fault handling operation steps are extracted to form a structured triple containing three elements. This achieves the automatic extraction of machine-recognizable fault handling knowledge units from standardized documents.

[0057] The embodiment of the present invention, through the semantic understanding ability of the large model, can adaptively identify the semantic features of titles at different levels and their association with the text, effectively realizing the explicit structured expression of implicit information in unstructured text, so that fault phenomena, identification problems and handling steps originally scattered in different locations of the document can be accurately associated.

[0058] An embodiment of the present invention provides a method for handling a device failure, which generates callable workflow data based on the extracted information tuple, and may include:

[0059] Based on the extracted information tuple, the fields of the fault node, the associated parent node, the fault handling parameters and the fault handling operation steps are determined, and based on the determined fields, a structured knowledge graph that can be called is generated as workflow data.

[0060] In embodiments of the present invention, workflow data can be formatted in JSON. A fault node refers to an independent entity (device) corresponding to a specific fault phenomenon. This can be achieved by using natural language processing techniques to identify the entity description of the fault phenomenon in standard operation and maintenance documents and assigning a unique identifier. This is used to create an indexable independent node in the knowledge graph. The associated parent node refers to the parent device of the fault node. This can be obtained by analyzing the associations between different devices in the document and used to construct the tree-like topology of the knowledge graph. Fault handling parameters refer to the device parameters or operating conditions involved in the operation steps (such as "voice input keywords" and "step execution result feedback"). This can be achieved by extracting numerical data from text descriptions using regular expression matching or named entity recognition techniques, thereby converting natural language descriptions into executable instruction parameters. The fields of the fault handling operation steps refer to standardized operating instructions arranged in sequence. This can be achieved by using a sequence labeling model to perform structured extraction of action verbs and operation objects from the text, thereby generating a programmatic process that can be called by the system. In embodiments of the present invention, a large-scale prompt instruction project is constructed to achieve dual processing, including the formation of triples and the generation of workflow data.

[0061] In one specific implementation, the predefined hierarchical knowledge structure in the standardized knowledge template is converted into a node-relationship framework within a knowledge graph. Fault phenomenon entities within information tuples are mapped as nodes within the graph, their parent devices are mapped as their parent nodes, and fault handling parameters and operation steps serve as the data associated with each node, thereby generating a corresponding knowledge graph. The resulting knowledge graph leverages the relationships between nodes, enabling the system to quickly locate faulty nodes and their associated handling processes using a graph traversal algorithm.

[0062] The embodiment of the present invention, through the construction of a structured knowledge graph, not only realizes the explicit expression of fault classification relationships, but also converts the handling steps described in natural language into executable instruction chains containing parameter fields, so that workflow data has programmatic calling capabilities, and the correspondence between fault phenomena and fault handling operation steps and other information can be automatically parsed by the machine, and then the parameters and operation steps in the knowledge graph can be directly called to generate fault handling solutions, avoiding operational delays caused by manual document retrieval.

[0063] An embodiment of the present invention provides a method for handling equipment faults, which utilizes the large model to search for matches in the workflow data based on the fault phenomenon description, and obtains a recommended fault handling solution based on the matched information, and may include:

[0064] The large model is used to retrieve and match corresponding fault nodes in the workflow data based on the fault phenomenon description, and the associated parent node, fault handling parameters and fault handling operation steps are determined based on the fault node, and a fault handling plan containing the determined data is generated.

[0065] The large model in the embodiments of the present invention refers to a pre-trained language model with natural language processing capabilities, specifically a model based on the Transformer architecture. It is used to semantically parse input fault phenomenon descriptions, identify implicit fault characteristics, and overcome the limitations of traditional keyword matching. Workflow data refers to information related to equipment failures stored in the form of a structured knowledge graph. It can be generated through triple extraction and field mapping to provide machine-readable hierarchical relationships and operational parameters.

[0066] In a specific implementation, when the operation and maintenance personnel input a description of the fault phenomenon through voice (which can be converted into text through ASR technology) or text, the big model first performs semantic analysis on the description to identify the key fault features; then, the fault nodes that match the features are retrieved in the knowledge graph of the workflow data. During the retrieval process, the big model can filter out the most relevant fault nodes by calculating the semantic similarity; after a successful match, the parent node corresponding to the fault node is located according to the node association relationship preset in the knowledge graph to confirm the potential cause of the fault, and at the same time, the fault handling parameters and fault handling operation steps bound to the node are extracted from the structured field. Finally, the fault root cause analysis, fault handling parameters and fault handling operation steps corresponding to the parent node are dynamically combined to generate a complete handling plan including operation instructions, parameter configuration and associated logic.

[0067] The embodiment of the present invention achieves accurate mapping of fault phenomena and treatment solutions through the collaboration of the semantic understanding ability of the large model and the structured knowledge graph, and at the same time traces the root cause of the fault through the parent node to avoid the one-sidedness of the treatment solution caused by the matching of a single phenomenon; and can automatically convert the fault phenomenon described in natural language into executable standardized operation instructions, reducing the time consumption of manual retrieval and experience judgment, improving the matching accuracy of treatment solutions in complex fault scenarios, and ensuring the integrity and executable nature of the operation steps and equipment parameters.

[0068] An embodiment of the present invention provides a method for handling a device failure, which, after generating workflow data that can be called based on the extracted information tuple, may further include:

[0069] The various data in the workflow data are displayed in a visual flow chart on a preset page, and the operation and maintenance experts are accepted to dynamically arrange the displayed various data, and the work data flow is updated based on the accepted dynamic arrangement.

[0070] Among them, visual flowchart refers to the conversion of structured workflow data into graphical interface elements for display. Specifically, it can be implemented by using Web front-end technology combined with a flowchart drawing library. For example, nodes and connecting lines are rendered through SVG or Canvas. This feature enables complex data structures to be presented in an intuitive form. Dynamic orchestration refers to allowing users to interactively operate on nodes in the flowchart. Specifically, it can be implemented by using drag event monitoring and data binding technology. For example, operation and maintenance experts can adjust the order of steps or modify node attributes / information by dragging the mouse. This feature gives users the ability to flexibly adjust standardized processes. Updating the work data flow refers to synchronizing the orchestrated flowchart data to the underlying storage of the system. Specifically, it can be implemented by using JSON format conversion and database transaction mechanism, so that the latest workflow data corresponds to the modified flowchart. For example, the modified flowchart is serialized into a JSON string and submitted to the server to update the database record. This feature ensures that user operations can be fed back to the system knowledge base in real time.

[0071] Specifically, after the workflow data is generated, the visualization engine can be called to render the fault nodes, fault handling parameters, fault handling operation steps and their logical relationships into a flowchart. The operation and maintenance expert accesses the preset page through the browser to view the flowchart. If it is found that the process contains a step sequence that does not match the actual scenario or a missing branch or a parameter exceeds the limit, the node can be dragged to adjust the execution order or a new handling step node can be inserted at the specified location or the fault handling parameters can be modified. For example, when a certain type of fault adds an on-site emergency handling requirement, the expert can directly add a label node to the corresponding position of the flowchart and associate the fault handling parameters. All orchestration operations trigger data change events, and the system automatically converts the updated flowchart into a structured data format, overwriting the original workflow data storage content, and realizing dynamic optimization of the knowledge base. For example, when a new device is connected, the expert can directly expand the fault node branch in the visualization interface without waiting for the system version iteration, shortening the knowledge update cycle from months to hours, while reducing the risk of misoperation caused by process delays.

[0072] This embodiment of the present invention delegates process adjustment authority to operations and maintenance experts through a visual interactive interface, enabling them to optimize processes without relying on IT support. For example, if a safety hazard is discovered in an existing handling step, the expert can immediately insert a safety confirmation node into the flowchart, and the system will complete the data update and take effect within minutes. This enables operations and maintenance experts to dynamically adjust standardized workflows in real time, resolving the adaptability issues of traditional systems to special failure scenarios caused by rigid processes.

[0073] A device fault handling method provided by an embodiment of the present invention may further include, before displaying various data in the workflow data in the form of a visual flowchart on a preset page:

[0074] Receive a display instruction and verify whether the identity information contained in the display instruction is legal. If it is legal, execute the step of displaying each data in the workflow data in the form of a visual flowchart on a preset page. Otherwise, display the information that you are not authorized to update the workflow data flow.

[0075] The display instruction in the embodiment of the present invention refers to the user's operation request to trigger the visual flowchart display function, which can be implemented by clicking a page button or entering a specific command. Its function is to obtain the user's operation intention before starting the data display process. Identity information legitimacy verification refers to the authentication of user login credentials or permission identifiers, which can be implemented by a role-based access control model or a digital certificate authentication mechanism. Its function is to confirm whether the user has the authority to access the structured knowledge graph. The unauthorized update prompt information refers to the insufficient permission notification returned when the identity authentication fails. It can be implemented by a preset text pop-up window or error code feedback method. Its function is to block illegal access and clearly prompt operation restrictions.

[0076] Specifically, when an operations and maintenance personnel initiates a visualization display request through the client interface, the request is automatically captured and the identity credentials are extracted. The extracted identity credentials are compared with the records in the permission database. If a match is successful, access to the knowledge graph data is allowed, and a visualization flowchart containing relevant information such as the fault node and the handling steps is generated on a preset page. If the verification fails, the display process is immediately terminated, and a warning message indicating that the operation is not authorized is returned on the interface. This mechanism embeds identity authentication into the data call link through pre-authentication operations, ensuring that only authorized users can trigger the knowledge graph visualization function.

[0077] The embodiment of the present invention realizes dynamic permission management of the knowledge graph visualization process, solves the security risk of unauthorized personnel viewing or tampering with the operation and maintenance knowledge base. By adding a secondary authentication step before starting the visualization process, a multi-level access control system is constructed to effectively prevent permission leakage or unauthorized operations. By blocking the access requests of illegal users and providing instant feedback on the permission status, it avoids the risk of sensitive data leakage and reduces the possibility of knowledge base pollution due to misoperation.

[0078] The technical solution provided by the embodiment of the present invention has the following features:

[0079] Structured knowledge conversion overcomes the pain points of non-standardization: By using pre-set Word templates to organize knowledge into a fixed hierarchical structure, unstructured documents are converted into machine-parseable JSON workflows, enabling the system to quickly locate the "fault symptom-handling steps" correspondence, improving retrieval efficiency.

[0080] Big models empower intelligent fault handling: Leveraging the natural language understanding capabilities of big models, they can parse spoken device fault descriptions entered by operations and maintenance personnel, resolving the issue of traditional rule bases being unable to cover complex scenarios.

[0081] Multimodal interaction improves operational efficiency: The combination of voice input and visual workflows eliminates the need for maintenance personnel to manually review documents. Especially in noisy environments on-site, they can obtain real-time troubleshooting guidance through voice commands. Structured step-by-step instructions also reduce the learning curve for new employees.

[0082] Dynamic knowledge update mechanism: This allows business experts to directly and visually modify JSON workflow data (such as adding new fault scenarios and adjusting the sequence of equipment troubleshooting steps) through a web interface without relying on technical development. This enables rapid iteration of the knowledge system and solves the high maintenance cost of traditional expert system rule bases.

[0083] In addition, it should be noted that the models involved above can all be implemented using existing models or models obtained based on existing technologies, and will not be elaborated on here.

[0084] The embodiment of the present invention also provides an equipment fault processing device, which is applied to the operation and maintenance of rail transportation equipment; Figure 2 As shown, the data processing device may include:

[0085] The standard document construction module 11 is used to: obtain unstructured equipment operation and maintenance documents, and construct the equipment operation and maintenance documents into structured standard operation and maintenance documents according to a preset standardized knowledge template;

[0086] The workflow generation module 12 is configured to extract information tuples related to equipment failures from the standard operation and maintenance documents based on the natural language processing function of the large model, and generate workflow data that can be called based on the extracted information tuples; wherein the information related to equipment failures includes the failure symptoms and the failure handling steps;

[0087] The intelligent interaction module 13 is used to: receive the fault phenomenon description input by the operation and maintenance personnel, use the large model to search for matches in the workflow data based on the fault phenomenon description, and obtain a recommended fault handling solution based on the matched information, so that the operation and maintenance personnel can handle the fault based on the recommended fault handling solution.

[0088] In an apparatus for handling equipment failure provided by an embodiment of the present invention, a standard document construction module may include:

[0089] The standard document construction unit is configured to retrieve a current standardized knowledge template and add the content of the equipment operation and maintenance document to corresponding positions of the standardized knowledge template according to the titles / texts at different levels in the hierarchical knowledge structure, thereby obtaining a standard operation and maintenance document; wherein the standardized knowledge template defines a hierarchical knowledge structure, and the titles / texts at different levels correspond to the fault phenomenon, fault identification problem, and fault handling steps of the equipment fault, respectively;

[0090] An embodiment of the present invention provides a device for handling equipment failures, wherein the workflow generation module may include:

[0091] The triple generation unit is used to: use the natural language processing function of the large model to identify the correspondence between each title and the corresponding text in the standard operation and maintenance document, and extract triples containing fault phenomena, fault identification problems and fault handling operation steps based on the correspondence.

[0092] The workflow generation unit is used to: determine the fields of the fault node, the associated parent node, the fault handling parameters and the fault handling operation steps based on the extracted information tuple, and generate a callable structured knowledge graph as workflow data based on the determined fields.

[0093] An embodiment of the present invention provides a device for handling equipment failures, wherein the intelligent interaction module may include:

[0094] An intelligent interaction unit is used to: use the large model to retrieve and match corresponding fault nodes in the workflow data based on the fault phenomenon description, determine the associated parent node, fault handling parameters and fault handling operation steps based on the fault node, and generate a fault handling plan containing the determined data.

[0095] An apparatus for handling equipment failures provided by an embodiment of the present invention may further include:

[0096] The dynamic orchestration module is used to: generate callable workflow data based on the extracted information tuple, display each data in the workflow data in the form of a visual flow chart on a preset page, accept dynamic orchestration of the displayed data by operation and maintenance experts, and update the workflow data flow based on the accepted dynamic orchestration.

[0097] An apparatus for handling equipment failures provided by an embodiment of the present invention may further include:

[0098] The identity authentication module is used to: receive a display instruction before displaying the various data in the workflow data in the form of a visual flowchart on a preset page, and verify whether the identity information contained in the display instruction is legal; if so, execute the step of displaying the various data in the workflow data in the form of a visual flowchart on the preset page; otherwise, display information indicating that the user has no right to update the workflow data flow.

[0099] The device fault handling apparatus in this embodiment is presented in the form of a functional module, where the module refers to an application-specific integrated circuit (ASIC), a processor and memory that executes one or more software or fixed programs, and / or other devices that can provide the above functions.

[0100] The further functional description of each of the above modules and units is the same as that of the above corresponding embodiments and will not be repeated here.

[0101] The embodiment of the present invention also provides a computer device having the above Figure 2 The device troubleshooting device shown.

[0102] See also Figure 3 , Figure 3 is a structural diagram of a computer device provided by an optional embodiment of the present invention, such as Figure 3 As shown, the computer device includes: one or more processors 10, memory 20, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. Various components utilize different buses to communicate with each other and can be installed on a common mainboard or installed in other ways as needed. The processor can process the instructions executed in the computer device, including instructions stored in the memory or on the memory to display the graphical information of the GUI on an external input / output device (such as, a display device coupled to the interface). In some optional embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Equally, multiple computer devices can be connected, and each device provides part of the necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system). Figure 3 A processor 10 is taken as an example.

[0103] The processor 10 may be a central processing unit, a network processor, or a combination thereof. The processor 10 may further include a hardware chip. The hardware chip may be an application-specific integrated circuit, a programmable logic device, or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general purpose array logic, or any combination thereof.

[0104] The memory 20 stores instructions that can be executed by at least one processor 10, so as to enable at least one processor 10 to execute the method shown in the above embodiment.

[0105] The memory 20 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created based on the use of a computer device for displaying a small program landing page, etc. In addition, the memory 20 may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some optional embodiments, the memory 20 may optionally include a memory remotely located relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0106] The memory 20 may include a volatile memory, such as a random access memory; the memory may also include a non-volatile memory, such as a flash memory, a hard disk or a solid-state drive; the memory 20 may also include a combination of the above types of memory.

[0107] The computer device further includes an input device 30 and an output device 40. The processor 10, the memory 20, the input device 30 and the output device 40 may be connected via a bus or other means. Figure 3 The bus connection is taken as an example.

[0108] The input device 30 can receive input digital or character information and generate key signal input related to user settings and function control of the computer device, such as a touch screen, a keypad, a mouse, a trackpad, a touch pad, an indicator stick, one or more mouse buttons, a trackball, a joystick, etc. The output device 40 can include a display device, an auxiliary lighting device (e.g., an LED), and a tactile feedback device (e.g., a vibration motor). The above-mentioned display device includes but is not limited to a liquid crystal display, a light emitting diode, a display, and a plasma display. In some optional embodiments, the display device can be a touch screen.

[0109] The embodiment of the present invention also provides a computer-readable storage medium. The above-mentioned method according to the embodiment of the present invention can be implemented in hardware, firmware, or implemented as a computer code that can be recorded in a storage medium, or implemented as a computer code that is originally stored in a remote storage medium or a non-temporary machine-readable storage medium and downloaded through a network and will be stored in a local storage medium, so that the method described herein can be stored in such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only storage memory, a random access memory, a flash memory, a hard disk or a solid-state drive, etc.; further, the storage medium can also include a combination of the above-mentioned types of memory. It can be understood that a computer, a processor, a microprocessor controller or programmable hardware includes a storage component that can store or receive software or computer code. When the software or computer code is accessed and executed by a computer, a processor or hardware, the method shown in the above embodiment is implemented.

[0110] Although the embodiments of the present invention have been described with reference to the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present invention. Such modifications and variations are all within the scope defined by the appended claims.

Claims

1. A method for handling equipment failure, characterized in that: Applied to rail transportation equipment operation and maintenance; the data processing method includes: Obtaining unstructured equipment operation and maintenance documents, and constructing the equipment operation and maintenance documents into structured standard operation and maintenance documents according to a preset standardized knowledge template; Extracting equipment failure-related information tuples from the standard operation and maintenance documents based on the natural language processing function of the large model, and generating callable workflow data based on the extracted information tuples; wherein the equipment failure-related information includes the failure phenomenon and the failure handling operation steps; Receive the fault phenomenon description input by the operation and maintenance personnel, use the big model to search for matches in the workflow data based on the fault phenomenon description, and obtain a recommended fault handling solution based on the matched information, so that the operation and maintenance personnel can handle the fault based on the recommended fault handling solution.

2. The method according to claim 1, characterized in that The equipment operation and maintenance document is constructed into a structured standard operation and maintenance document according to a preset standardized knowledge template, including: Retrieve the current standardized knowledge template; wherein the standardized knowledge template defines a hierarchical knowledge structure, with titles / texts at different levels corresponding to the fault phenomenon, fault identification problem, and fault handling steps of the equipment fault; The content of the equipment operation and maintenance document is added to the corresponding position of the standardized knowledge template according to the titles / texts of different levels in the hierarchical knowledge structure to obtain a standard operation and maintenance document.

3. The method according to claim 2, characterized in that The natural language processing function based on the large model extracts information tuples related to equipment failure from the standard operation and maintenance documents, including: The natural language processing function of the large model is used to identify the correspondence between each title and the corresponding text in the standard operation and maintenance document, and based on the correspondence, triples containing fault phenomena, fault identification problems and fault handling operation steps are extracted.

4. The method according to claim 3, characterized in that Generating workflow data that can be called based on the extracted information tuple includes: Based on the extracted information tuple, the fields of the fault node, the associated parent node, the fault handling parameters and the fault handling operation steps are determined, and based on the determined fields, a structured knowledge graph that can be called is generated as workflow data.

5. The method according to claim 4, characterized in that Using the large model to search for matches in the workflow data based on the fault phenomenon description, and obtaining a recommended fault handling solution based on the matched information, including: The large model is used to retrieve and match corresponding fault nodes in the workflow data based on the fault phenomenon description, and the associated parent node, fault handling parameters and fault handling operation steps are determined based on the fault node, and a fault handling plan containing the determined data is generated.

6. The method according to any one of claims 1 to 5, characterized in that After generating workflow data that can be called based on the extracted information tuple, the method further includes: The various data in the workflow data are displayed in a visual flow chart on a preset page, and the operation and maintenance experts are accepted to dynamically arrange the displayed various data, and the work data flow is updated based on the accepted dynamic arrangement.

7. The method according to claim 6, characterized in that Before the preset page displays the various data in the workflow data in the form of a visual flowchart, it also includes: Receive a display instruction and verify whether the identity information contained in the display instruction is legal. If it is legal, execute the step of displaying each data in the workflow data in the form of a visual flowchart on a preset page. Otherwise, display the information that you are not authorized to update the workflow data flow.

8. A device for handling equipment failure, characterized in that: Applicable to rail transportation equipment operation and maintenance; the data processing device includes: A standard document construction module is used to: obtain unstructured equipment operation and maintenance documents, and construct the equipment operation and maintenance documents into structured standard operation and maintenance documents according to a preset standardized knowledge template; A workflow generation module is configured to extract information tuples related to equipment failures from the standard operation and maintenance documents based on the natural language processing function of the large model, and generate workflow data that can be called based on the extracted information tuples; wherein the information related to equipment failures includes the failure symptoms and the failure handling steps; The intelligent interaction module is used to: receive a fault phenomenon description input by an operation and maintenance personnel, use the large model to search for matches in the workflow data based on the fault phenomenon description, and obtain a recommended fault handling solution based on the matched information, so that the operation and maintenance personnel can handle the fault based on the recommended fault handling solution.

9. A computer device, characterized in that: include: A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the device fault handling method according to any one of claims 1 to 7 by executing the computer instructions.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute the device failure processing method according to any one of claims 1 to 7.

Images