Equipment operation and maintenance decision support method and system, computer equipment and storage medium

Abstract

The invention discloses an equipment operation and maintenance decision support method and system, computer equipment and a storage medium, and relates to the technical field of equipment operation and maintenance intelligent management. Constructing a semantic association model of the structured data and the unstructured data; and a reasoning rule base is constructed to perform automatic reasoning on the equipment data, and fault early warning and operation optimization suggestions are provided. According to the method, through dynamic management and expansion of equipment entity information, the accuracy and relevance of equipment data are improved, and problems caused by data omission or redundancy are reduced; the problem that unstructured data is difficult to use is effectively solved, and the richness and accuracy of the knowledge graph are improved; through automatic verification and complementation, the integrity and accuracy of the data are improved, and the complexity of manual maintenance is reduced; and through dynamic generation and automatic reasoning of the reasoning rule base, the intelligent level of equipment operation and maintenance is improved, and accurate fault early warning and operation optimization are realized.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent management of equipment operation and maintenance, and specifically to an equipment operation and maintenance decision support method, system, computer device and storage medium. Background Art

[0002] In recent years, with the rapid development of industrial Internet and Internet of Things technologies, equipment operation and maintenance has gradually moved towards digitalization and intelligence. Traditional equipment operation and maintenance methods mainly rely on manual monitoring, recording and experience judgment. This method cannot meet the needs of diversified equipment types and complex operating conditions in modern industry. Knowledge graph technology has gradually become an important tool in the field of equipment operation and maintenance. By converting scattered data into structured knowledge and displaying the relationship between devices in the form of semantic relationships, it provides support for equipment status monitoring and management. At the same time, the application of big data analysis and machine learning technology makes it possible to mine fault modes and optimization rules from massive operating data. In addition, the inference rule engine, as the core of intelligent decision support, can evaluate equipment status and predict potential failures through rule logic. However, although these technologies have made great progress, the coordination and integration between technologies have not yet been fully realized, and there are still technical difficulties in the intelligent management of equipment operation and maintenance.

[0003] Existing equipment operation and maintenance systems usually adopt a decentralized data storage mode. Equipment data from different sources are stored in heterogeneous databases, resulting in a lack of relevance between data. In addition, the update of equipment data relies on manual operations, and it is difficult to ensure real-time and consistency, which easily generates data islands. Although some systems use knowledge graph technology for equipment management, most of them only stay at the basic mapping of structured data and fail to effectively process the implicit knowledge in unstructured data (such as operation logs, images, etc.). At the same time, for problems such as synonym ambiguity and data missing, the existing technology lacks effective solutions, resulting in limitations in the construction and application of semantic models. Most of the existing inference rules rely on manual writing and fixed rule logic, which cannot meet the needs of complex changes in equipment operation status. The rule base is difficult to dynamically adjust and optimize. The results of the inference engine lack a feedback mechanism, and the rule base cannot be iteratively updated through actual operation data, thereby limiting the adaptability and prediction ability of the system. Most of the current systems use a simple threshold trigger mechanism, which is difficult to handle multi-dimensional anomaly detection under complex working conditions. At the same time, the generation of operation optimization decisions mostly stays at the empirical rule, lacks a deep combination of the real-time operation status of the equipment and historical data, and cannot meet the refined needs of equipment operation and maintenance. Summary of the invention

[0004] In view of the above-mentioned problems, the present invention is proposed.

[0005] Therefore, the technical problems solved by the present invention are: the existing equipment data management methods have the problems of data dispersion and lack of dynamic update capabilities, the semantic modeling technology lacks flexibility and depth, the generation and optimization capabilities of the inference rule library are weak, and how to achieve fault warning and operation optimization based on graph editing, semantic modeling and inference rule library.

[0006] To solve the above technical problems, the present invention provides the following technical solutions: an equipment operation and maintenance decision support method, including graph editing to manage equipment data; building a semantic association model for structured data and unstructured data; building an inference rule library to automatically reason about equipment data, and provide fault warnings and operation optimization suggestions.

[0007] As a preferred solution of the equipment operation and maintenance decision support method described in the present invention, wherein: the graph editing includes entity editing and relationship editing.

[0008] As a preferred solution of the equipment operation and maintenance decision support method described in the present invention, the entity editing includes: equipment entity data management based on multi-condition retrieval, searching according to entity name, definition, alias, attribute, category, and adding, modifying, and deleting entity attributes of search results; expansion and multimodal association management of entity data, adding pictures, documents and multimodal resources to entities, and automatically associating related entities.

[0009] As a preferred solution of the equipment operation and maintenance decision support method described in the present invention, the relationship editing includes: creating and adjusting dynamic relationships based on target entity pairs, creating new relationships according to entity types or attributes specified by the user, and updating relationship weights in real time; querying and managing relationship data through multi-dimensional conditions, filtering by relationship type, associated entity or attribute conditions, and editing or deleting the filtering results.

[0010] As a preferred solution of the equipment operation and maintenance decision support method described in the present invention, wherein: the construction of a semantic association model of structured data and unstructured data includes, by parsing structured data and unstructured data, identifying the entities, attributes and relationships of the equipment, and converting the structured data and unstructured data into a knowledge expression format; for structured data, parsing the table or database information into entity and relationship triples through a pattern matching algorithm; for unstructured data, using a natural language processing method to extract semantic information from text, images or videos, generate semantic associations between devices and store them in a knowledge base; the system automatically detects duplicate definitions of synonymous entities or attributes during the analysis phase, and the user makes adjustments through editing tools. For missing data, the suggested values ​​are automatically generated through context inference or completion algorithms, and the user is prompted to confirm the completion of data completion.

[0011] As a preferred solution of the equipment operation and maintenance decision support method described in the present invention, the construction of the inference rule library includes formulating and extracting logical rules from the equipment operation data, and dynamically sorting the rules based on the applicable scope and weight of the rules to form an inference rule library; extracting equipment attributes and associated parameters from the newly entered data sample set, entering the qualified data into the knowledge base through validity verification, generating logical rules based on the newly added inference rule library data and mapping them to equipment entities and relationships, and building triple semantic relationships between devices. For invalid data that does not comply with the rules, the system automatically deletes and updates the operation log, prompting the reason for deleting the invalid data, screening and querying the applicable scope of the rules, and dynamically sorting the rules based on the frequency of use and weight of the rules. By selecting the inference target and the current knowledge status, forward reasoning, reverse reasoning or machine learning strategy is selected according to the rules to generate adaptation rules.

[0012] As a preferred solution of the equipment operation and maintenance decision support method described in the present invention, the provision of fault warnings and operation optimization suggestions includes obtaining entity data and relationship data that have been added, deleted, checked and modified from the equipment map editing, generating a complete equipment operation status model, and based on the constructed semantic association model, storing structured data and unstructured data in a unified manner in the knowledge base, extracting equipment operation parameters and historical records, forming a semantic link between equipment status and failure mode, and using an inference rule base to automatically derive equipment operation optimization suggestions or fault warning information based on the equipment operation status model and semantic links, and generating targeted operation and maintenance decision plans. The decision plans are distributed to relevant responsible persons through an automated push mechanism, and the results of the plan execution are fed back to the knowledge base to update the map content and rule base weights.

[0013] Another object of the present invention is to provide an equipment operation and maintenance decision support system, which can extract logical rules from data and generate an inference rule base, dynamically sort the rules based on their scope of application and weight, and support real-time updating and optimization of the rule base, thereby solving the problem that the current intelligent management technology for equipment operation and maintenance contains weak generation and optimization capabilities of the inference rule base.

[0014] As a preferred solution of the equipment operation and maintenance decision support system described in the present invention, it includes: a graph editing module, a semantic association module, and a decision feedback module; the graph editing module includes an entity data management module and a relationship data management module, the entity data management module is used to support the addition, deletion, query and modification operations of equipment entity data, retrieve entities according to multiple conditions, and expand the multimodal associated resources of entities, the relationship data management module is used to support dynamic creation and adjustment of relationships between devices, filter and manage relationship data according to multi-dimensional conditions, and update relationship weights and logic in real time; the semantic association module includes a structured data parsing module and an unstructured data parsing module, the structured data parsing module is used to parse equipment attributes, parameters and associated information in a table or database through a pattern matching algorithm The unstructured data parsing module is used to extract semantic information from unstructured data using natural language processing technology and generate semantic associations between devices; the decision feedback module includes a rule reasoning module and a decision generation module. The rule reasoning module is used to extract logical rules from data and generate a reasoning rule base, and dynamically sort the rules based on their applicable scope and weight, while supporting real-time updating and optimization of the rule base. The decision generation module is used to select reasoning strategies based on the rule base, analyze device data, generate device operation status predictions, optimization suggestions or fault warning information, integrate reasoning results and knowledge models, generate targeted operation and maintenance decision plans and push them to the responsible person, and update the knowledge graph and rule base based on execution feedback.

[0015] A computer device includes a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement a step of a device operation and maintenance decision support method.

[0016] A computer-readable storage medium stores a computer program, which implements the steps of an equipment operation and maintenance decision support method when executed by a processor.

[0017] Beneficial effects of the present invention: The equipment operation and maintenance decision support method provided by the present invention improves the accuracy and relevance of equipment data by dynamically managing and expanding equipment entity information, reduces the problems caused by data omission or redundancy, and lays a foundation for the integrity and accuracy of the knowledge graph; realizes dynamic adjustment and optimization of the association relationship between devices, so that the knowledge graph can reflect the equipment operation status and historical association in real time, and improves the flexibility and applicability of the graph; realizes the efficient conversion of structured data into semantic models, ensures the comprehensibility and availability of data, and provides high-quality input for subsequent reasoning and analysis; effectively solves the problem that unstructured data is difficult to use, and improves the richness and accuracy of the knowledge graph; through automatic verification and completion, the integrity and accuracy of the data are improved, the complexity of manual maintenance is reduced, and reliable data support is provided for the generation and application of the reasoning rule library; through the dynamic generation and automated reasoning of the reasoning rule library, the intelligence level of equipment operation and maintenance is improved, and accurate fault warning and operation optimization are achieved. The present invention achieves better results in terms of reliability, flexibility and applicability. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0019] Figure 1 This is an overall flow chart of the equipment operation and maintenance decision support method provided by the first embodiment of the present invention.

[0020] Figure 2 This is an overall flow chart of the equipment operation and maintenance decision support system provided in the third embodiment of the present invention. DETAILED DESCRIPTION

[0021] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0022] Example 1, reference Figure 1 , is an embodiment of the present invention, and provides a device operation and maintenance decision support method, including:

[0023] S1: Edit the graph to manage device data.

[0024] Furthermore, graph editing includes entity editing and relationship editing.

[0025] It should be noted that entity editing includes device entity data management based on multi-condition retrieval, searching according to entity name, definition, alias, attributes, and category, and adding, modifying, and deleting entity attributes in the search results; expansion of entity data and multimodal association management, adding images, documents, and multimodal resources to entities, and automatically associating related entities.

[0026] It should also be noted that the specific operations of a preferred solution for device entity data management include: the system receives search conditions entered by the user, matches qualified entities based on entity names or attribute fields; displays search results in a visual interface, allowing users to directly modify attribute values ​​or add associated attributes after selecting the target entity; when deleting an operation, the system performs data dependency verification, and if the target entity has an unfinished association relationship, the deletion operation is blocked and the user is prompted to process.

[0027] It should also be noted that a preferred solution for the expansion and multimodal association management of entity data includes the following specific operations: the user uploads images or document resources for the specified entity through the interface, and the system parses the resource content to identify potential related entities; the system establishes association relationships based on content analysis algorithms, such as matching key terms in the uploaded document to other entities; the final association results are displayed in the form of links, and support manual adjustment or confirmation by users to ensure data consistency.

[0028] It should also be noted that relationship editing includes, dynamic relationship creation and adjustment based on target entity pairs, creation of new relationships according to user-specified entity types or attributes, and real-time updating of relationship weights; querying and managing relationship data through multi-dimensional conditions, filtering by relationship type, associated entity or attribute conditions, and editing or deleting the filtering results.

[0029] It should also be noted that a preferred solution for real-time updating of relationship weights includes the following specific operations: after the user specifies the starting entity and the target entity, the system automatically recommends possible relationship types; after the user selects the relationship type, the system adds the relationship to the database and initializes the weight; the weight value can be adjusted through a slider or direct input, and the adjusted value is updated in real time to the knowledge database and reflected in the visualization interface.

[0030] It should also be noted that the specific operations of a preferred solution for querying and managing relational data include: the system receives query conditions specified by the user, such as relationship type or related entity, retrieves and displays matching relational data; for screening results, the user can choose to delete invalid relationships, and the system verifies data dependencies before deletion to prevent erroneous operations; during editing operations, the system allows users to change the attribute values ​​of the relationship, such as start and end time or strength coefficient, and synchronizes the updated results to the knowledge graph in real time.

[0031] It should also be noted that entity editing can efficiently find device information according to dimensions such as entity name, alias, and attributes through multi-condition retrieval and management of entity data, and realize data addition, deletion, query, and modification operations; further, by extending functions, multimodal resources (such as pictures and documents) are added to device entities, and automatic association between entities is realized to ensure the comprehensiveness and diversity of device data; the visualization and operability of device data are improved, and an accurate and complete data input source is provided for subsequent semantic modeling and reasoning; relationship editing can generate new relationships according to entity types or attributes through dynamic relationship creation and adjustment based on target entities, and update relationship weights in real time to ensure the dynamic nature of relationship data; by querying and filtering relationship data under multi-dimensional conditions, problems can be quickly located and redundant relationships can be edited or deleted in a timely manner; a dynamic semantic relationship network between devices is constructed, providing accurate relationship descriptions for semantic modeling of complex device systems.

[0032] S2: Build a semantic association model for structured and unstructured data.

[0033] Furthermore, a semantic association model for structured data and unstructured data is constructed, including: by parsing structured data and unstructured data, identifying the entities, attributes and relationships of devices, and converting structured data and unstructured data into knowledge expression formats; for structured data, parsing table or database information into entity and relationship triples through pattern matching algorithms; for unstructured data, using natural language processing methods to extract semantic information from text, images or videos, generating semantic associations between devices and storing them in a knowledge base; the system automatically detects duplicate definitions of synonymous entities or attributes during the analysis phase, and users make adjustments through editing tools. For missing data, suggested values ​​are automatically generated through context inference or completion algorithms, and the user is prompted to confirm that the data completion is complete.

[0034] It should be noted that structured data is currently the main source of knowledge, such as data such as tables, charts or information in databases. Structured data is used to define and model entities, attributes and relationships; they are converted into semantic links (triples) between entities, attributes and relationships to support the search and query of knowledge graphs; for structured data, they can usually be directly used and converted to form basic data sets, and then further expanded using knowledge graph completion technology; unstructured data is generally data without a clear and accurate format and structure, such as images, videos, voice, text, etc. For this type of data, semantic analysis, extraction and linking are performed on it to identify the entities, attributes and relationships contained therein, and convert them into the format and structure in the knowledge graph to support knowledge representation, search and query; for unstructured data, the main ways of acquiring knowledge include entity recognition, relationship extraction, attribute extraction, etc.

[0035] It should also be noted that structured data processing uses pattern matching algorithms to parse structured data such as tables and databases, maps equipment attributes and operating parameters into triplets of entities and relationships and stores them in the knowledge base; converts traditional tabular data into standardized knowledge graph data formats, improving the efficiency of data query and analysis; unstructured data processing uses natural language processing technology (such as text analysis, image recognition, etc.) to extract semantic information from equipment operation logs, videos and documents, generates semantic associations between devices and stores them in the knowledge base; makes implicit knowledge in unstructured data explicit, supplements and expands the content of the equipment knowledge graph; when checking and completing data consistency, the system automatically detects duplicate definitions of synonymous entities or attributes, and generates recommended values ​​for missing data through context inference algorithms; maintains the consistency and integrity of data in the knowledge graph, and ensures the reliability of the semantic model.

[0036] S3: Build an inference rule library to automatically infer equipment data and provide fault warnings and operation optimization suggestions.

[0037] Furthermore, an inference rule library is constructed, including formulating and extracting logical rules from equipment operation data, and dynamically sorting the rules based on their applicable scope and weight to form an inference rule library; extracting equipment attributes and associated parameters from newly entered data sample sets, entering qualified data into the knowledge base through validity verification, generating logical rules based on the newly added inference rule library data and mapping them to equipment entities and relationships, and constructing triple semantic relationships between devices. For invalid data that does not comply with the rules, the system automatically deletes and updates the operation log, prompting the reason for deleting the invalid data, filtering and querying the applicable scope of the rules, and dynamically sorting the rules based on their usage frequency and weight. By selecting the inference target and the current knowledge status, forward reasoning, reverse reasoning or machine learning strategies are selected according to the rules to generate adaptation rules.

[0038] It should be noted that providing fault warnings and operation optimization suggestions includes obtaining entity data and relationship data that have been added, deleted, checked and modified from the equipment map editing, generating a complete equipment operation status model, and storing structured data and unstructured data in the knowledge base based on the constructed semantic association model. The equipment operation parameters and historical records are extracted to form a semantic link between the equipment status and the fault mode. The inference rule base is used to automatically derive the equipment's operation optimization suggestions or fault warning information based on the equipment operation status model and semantic links, and generate targeted operation and maintenance decision plans. The decision plans are distributed to relevant responsible persons through an automated push mechanism, and the results of the plan execution are fed back to the knowledge base to update the map content and the rule base weights.

[0039] It should also be noted that the generation and management of inference rules extracts logical rules from equipment operation data, and dynamically sorts them based on the scope of application and weight of the rules to form an inference rule base; for newly added data samples, the data is entered into the knowledge base through validity verification, and new logical rules are generated and triple semantic relationships between devices are constructed; through the dynamic generation and optimization of rules, an efficient rule reasoning infrastructure is constructed, which improves the coverage and adaptability of the rules; the automated analysis of the inference engine selects the inference target and the current knowledge status, and the inference engine combines the rule base to evaluate the equipment operation status using forward reasoning, reverse reasoning or machine learning strategies to generate optimization suggestions or fault warning information; the data analysis and reasoning process is automated to achieve seamless connection from equipment data to operation and maintenance decisions.

[0040] Example 2 is an embodiment of the present invention, which provides an equipment operation and maintenance decision support method. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation experiments.

[0041] First, the historical operation data of a mechanical equipment system is collected, including tabular operation records, maintenance logs, abnormal event reports, and related video and image data; an operation and maintenance support system with knowledge graph editing, semantic modeling, and reasoning engine functions is built; the knowledge graph contains equipment entities and their relationships, and stores the equipment's operating status, associated parameters, and maintenance history records; the operation data is preprocessed, structured data is imported into the database, and natural language processing technology is used to parse unstructured data to generate semantic associations; the entity data management module is used to manage the equipment entity data; the entity information of a key equipment is located through multi-condition retrieval, and its attributes (such as temperature sensor model, operating time); upload relevant operating videos and maintenance documents for the device, the system automatically parses and establishes semantic links with other related devices; in the relational data management module, new association relationships are automatically generated according to the target entity type, such as the relationship between "device A" and "temperature control unit", the weight is set to 0.8, and adjusted to 0.9 through the slider to reflect its importance; some redundant or invalid relationships between devices are deleted, such as connections with deactivated devices; use pattern matching algorithms to parse structured data (such as operating parameter tables) and generate entity and relationship triplets, such as [device A]-[operating time]-[6000 hours].

[0042] Natural language processing technology is used to extract semantic information from unstructured data (such as maintenance logs). For example, "bearing temperature abnormality" and "maintenance operation record" are identified from maintenance records, and association relationships are generated and stored in the knowledge base. The system automatically detects the repeated synonymous entities "equipment A" and "main equipment A", merges them into a unified entity through context inference, and completes the missing attribute "rated power", prompting the user to confirm and complete the data completion. Logical rules are extracted from operating data, such as "if the temperature is >85℃ and the vibration frequency is abnormal, a bearing failure may occur". The rules in the rule base are sorted according to weight and scope of application, and the weight of the newly added rules is set to 0.7. The inference engine automatically analyzes the current equipment operation status, combines semantic links and inference rules to generate prediction results, and prompts "equipment A may have a bearing overheating failure after 48 hours of operation", and gives suggestions: "adjust the operating load or check the cooling system". The decision plan is pushed to the equipment maintenance personnel through the system. After the maintenance is completed, the feedback results are recorded in the knowledge base, and the map content and rule base are updated. Referring to Table 1, the experimental data are recorded and analyzed.

[0043] Table 1 Experimental data record table

[0044]

[0045] In the initial state, the operating data of device A is scattered and incomplete. The system retrieves and updates the device information through the entity editing function, and establishes new associations through multimodal resource expansion. Based on dynamic management and real-time updating of data, the accuracy and relevance of the device map are significantly improved, laying the foundation for subsequent semantic modeling. The table data shows that after the data is completed, the semantic model of device A in the knowledge map is gradually improved. For example, in the "further data supplementation" stage, the system generates missing attribute data through the context inference algorithm and prompts the user to confirm the completion. Through the construction of semantic links, the deep association between operating parameters and fault modes is realized, which improves the accuracy of fault prediction. The newly added logical rules in the rule base play a key role after the temperature rises. The system automatically infers the time when the fault may occur (48 hours) and generates optimization suggestions (such as adjusting the operating load). After the optimization is executed, the device status returns to normal and the future fault prediction is cleared. This shows that the dynamic combination of the inference rule base and the inference engine improves the intelligence and efficiency of operation and maintenance.

[0046] Example 3, reference Figure 2 , which is an embodiment of the present invention, provides an equipment operation and maintenance decision support system, including a graph editing module 100, a semantic association module 200, and a decision feedback module 300.

[0047] Among them, S4: the graph editing module 100 includes an entity data management module 101 and a relationship data management module 102. The entity data management module 101 is used to support the addition, deletion, query and modification operations of device entity data, retrieve entities according to multiple conditions, and expand the multimodal associated resources of entities. The relationship data management module 102 is used to support dynamic creation and adjustment of relationships between devices, filter and manage relationship data according to multi-dimensional conditions, and update relationship weights and logic in real time.

[0048] It should also be noted that the entity and relationship data generated by the entity data management module 101 and the relationship data management module 102 will be transmitted to the semantic association module 200 in real time as data input to provide support for subsequent semantic modeling.

[0049] S5: The semantic association module 200 includes a structured data parsing module 201 and an unstructured data parsing module 202. The structured data parsing module 201 is used to parse the device attributes, parameters and related information in a table or database through a pattern matching algorithm to generate a triple format of entities and relationships. The unstructured data parsing module 202 is used to extract semantic information from unstructured data using natural language processing technology to generate semantic associations between devices.

[0050] It should also be noted that the semantic association module 200 receives entity and relationship data from the graph editing module 100, performs deep semantic modeling on it, and passes the result to the rule reasoning module 301 of the decision feedback module 300 as basic data for reasoning.

[0051] S6: The decision feedback module 300 includes a rule reasoning module 301 and a decision generation module 302. The rule reasoning module 301 is used to extract logical rules from the data and generate a reasoning rule base, dynamically sort the rules based on their scope of application and weight, and support real-time updating and optimization of the rule base. The decision generation module 302 is used to select reasoning strategies based on the rule base, analyze equipment data, generate equipment operating status predictions, optimization suggestions or fault warning information, integrate reasoning results and knowledge models, generate targeted operation and maintenance decision plans and push them to the responsible person, and update the knowledge graph and rule base based on execution feedback.

[0052] It should also be noted that the rule reasoning module 301 relies on the semantic model and data support constructed by the semantic association module 200, and the decision generation module 302 combines the reasoning rule library and the graph data to generate a decision plan, and at the same time passes the decision results and feedback to the graph editing module 100 to achieve data updating and closed-loop management.

[0053] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods of each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0054] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0055] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering, or processing in another suitable manner as necessary, and then stored in a computer memory.

[0056] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc. It should be noted that the above embodiments are only used to illustrate the technical solution of the present invention and are not limited. Although the present invention is described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention, which should be included in the scope of the claims of the present invention.

[0057] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. Equipment operation and maintenance decision support method, characterized in that: include: Perform graph editing and manage equipment data; Build semantic association models for structured and unstructured data; Build an inference rule library to automatically reason about equipment data and provide fault warnings and operation optimization suggestions.

2. The equipment operation and maintenance decision support method according to claim 1, characterized in that: The map editing includes: Entity editing and relationship editing.

3. The equipment operation and maintenance decision support method according to claim 2, characterized in that: The entity editing includes, Equipment entity data management based on multi-condition retrieval, search by entity name, definition, alias, attribute, category, and add, modify, and delete entity attributes of search results; Expand entity data and manage multimodal associations, add images, documents, and multimodal resources to entities, and automatically associate related entities.

4. The equipment operation and maintenance decision support method according to claim 2, characterized in that: The relationship editing includes, Dynamic relationship creation and adjustment based on target entity pairs, creating new relationships based on user-specified entity types or attributes, and updating relationship weights in real time; Query and manage relational data through multi-dimensional conditions, filter by relationship type, related entity or attribute conditions, and edit or delete the filtered results.

5. The equipment operation and maintenance decision support method according to claim 1, characterized in that: The construction of a semantic association model of structured data and unstructured data includes: By parsing structured and unstructured data, identify the entities, attributes and relationships of devices, and convert structured and unstructured data into knowledge expression formats; For structured data, the table or database information is parsed into entity and relationship triples through pattern matching algorithms; For unstructured data, natural language processing methods are used to extract semantic information from text, images or videos, generate semantic associations between devices and store them in the knowledge base; During the analysis phase, the system automatically detects duplicate definitions of synonymous entities or attributes, and users make adjustments through editing tools. For missing data, the system automatically generates suggested values ​​through context inference or completion algorithms, and prompts users to confirm that data completion is complete.

6. The equipment operation and maintenance decision support method according to claim 1, characterized in that: The construction of the inference rule base includes: Formulate and extract logical rules from equipment operation data, and dynamically sort them based on the scope of application and weight of the rules to form an inference rule library; Extract device attributes and associated parameters from the newly entered data sample set, enter qualified data into the knowledge base through validity verification, generate logical rules based on the newly added reasoning rule library data and map them to device entities and relationships, build triple semantic relationships between devices, and automatically delete and update the operation log for invalid data that does not comply with the rules, prompt the reason for deleting the invalid data, filter and query the scope of application of the rules, and dynamically sort the rules based on their frequency of use and weight. By selecting the reasoning target and the current knowledge status, select forward reasoning, reverse reasoning or machine learning strategies according to the rules to generate adaptation rules.

7. The equipment operation and maintenance decision support method according to claim 1, 2, 5 or 6, characterized in that: The provision of fault warnings and operation optimization suggestions includes: Acquire entity data and relationship data that have been added, deleted, checked and modified from the equipment graph editing, generate a complete equipment operation status model, and based on the constructed semantic association model, store structured data and unstructured data in the knowledge base in a unified manner, extract equipment operation parameters and historical records, and form a semantic link between equipment status and failure mode. Utilize the inference rule base, based on the equipment operation status model and semantic link, automatically derive equipment operation optimization suggestions or fault warning information, and generate targeted operation and maintenance decision plans. The decision plans are distributed to relevant responsible persons through an automated push mechanism, and the results of the plan execution are fed back to the knowledge base to update the graph content and rule base weights.

8. Equipment operation and maintenance decision support system, characterized by: It includes a graph editing module (100), a semantic association module (200), and a decision feedback module (300); The graph editing module (100) includes an entity data management module (101) and a relationship data management module (102). The entity data management module (101) is used to support the addition, deletion, query and modification operations of device entity data, retrieve entities according to multiple conditions, and expand the multi-modal associated resources of entities. The relationship data management module (102) is used to support dynamic creation and adjustment of relationships between devices, filter and manage relationship data according to multi-dimensional conditions, and update relationship weights and logic in real time. The semantic association module (200) comprises a structured data analysis module (201) and an unstructured data analysis module (202), wherein the structured data analysis module (201) is used to analyze device attributes, parameters and associated information in a table or database through a pattern matching algorithm to generate a triple format of entities and relationships, and the unstructured data analysis module (202) is used to extract semantic information from unstructured data using natural language processing technology to generate semantic associations between devices; The decision feedback module (300) comprises a rule reasoning module (301) and a decision generation module (302). The rule reasoning module (301) is used to extract logical rules from data and generate a reasoning rule base, dynamically sort the rules based on their applicable scope and weight, and support real-time updating and optimization of the rule base. The decision generation module (302) is used to select a reasoning strategy based on the rule base, analyze the equipment data, generate equipment operation status prediction, optimization suggestions or fault warning information, integrate reasoning results and knowledge models, generate targeted operation and maintenance decision plans and push them to responsible persons, and update the knowledge graph and rule base based on execution feedback.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the equipment operation and maintenance decision support method described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the equipment operation and maintenance decision support method described in any one of claims 1 to 7 are implemented.

Images