Fuel-driven compressor unit fault auxiliary decision-making method and system based on knowledge graph

Abstract

The invention discloses a fuel-driven compressor unit fault auxiliary decision-making method and system based on a knowledge graph, and relates to the technical field of pipe network digitization and intellectualization, and the method comprises the steps: determining the fault type of a fuel-driven compressor unit, and the fault phenomenon, fault phenomenon description, fault reason, consequence influence and disposal measures of each type of fault, and determining the fault type of the fuel-driven compressor unit according to the fault type, fault phenomenon description, fault reason, consequence influence and disposal measures; constructing a knowledge graph; and performing fault decision on a preset fault of the combustion-driven compressor unit by using the knowledge graph. The method can assist a compressor station technology and management personnel to make decisions, realizes rapid and accurate positioning of fault sources and accurate control of potential fault points, achieves the purposes of effective prevention before faults and timely maintenance after the faults, effectively improves the autonomous maintenance efficiency and capability of station personnel, reduces the maintenance cost, prolongs the fault-free operation time of a compressor unit, and improves the working efficiency of the compressor unit. And cost reduction, efficiency improvement and safety improvement are realized.

Description

Background Art

[0002] As a key equipment of the compressor station, the normal operation of the compressor unit has a great impact on the transportation capacity and safety of the entire production line. From the perspective of operation management, the shutdown of the compressor unit due to failure directly affects the downstream gas consumption and the economic benefits of the enterprise. From the perspective of safety management, the failure of the compressor unit may cause safety accidents such as surge, fire and explosion, causing irreparable losses to human life, the surrounding environment and the reputation of the enterprise. The compressor unit, especially the compressor unit driven by gas turbine, has a sophisticated system, complex structure, various causes of failure, and a complex correspondence between the fault type and the fault symptom, which brings many difficulties to fault detection and hidden danger management.

[0003] At present, there are three major problems in the safe operation and maintenance management of fuel-driven compressor stations. First, there are very limited references and books on fuel-driven compressor unit failures, and there is not much knowledge accumulated in the field. The reliable operation of fuel-driven compressor units is still under exploration. Second, there are not many people in each compressor station who truly understand and master the operation and maintenance technology of compressor units, and their understanding of compressor units is more based on field experience. Most troubleshooting and repair and maintenance still need to rely on field experts and manufacturer engineers, and there are technical barriers; third, the station staff often lack a deep understanding of the operating mode and interlocking logic of the compressor unit. It is difficult to make accurate and effective judgments in time when the compressor unit is operating abnormally or fails, and there are capacity barriers. Summary of the invention

[0004] The technical problem to be solved by the present invention is to address the deficiencies of the prior art and specifically provide a method and system for auxiliary decision-making of a fuel-driven compressor unit fault based on a knowledge graph, as follows:

[0005] 1) In the first aspect, the present invention provides a method for auxiliary decision-making of fuel-driven compressor unit faults based on knowledge graph, and the specific technical scheme is as follows:

[0006] Determine and construct a knowledge graph based on the fault categories of fuel-driven compressor units and the fault phenomena, fault phenomenon descriptions, fault causes, consequences and disposal measures of each type of fault;

[0007] The knowledge graph is used to make fault decisions for preset faults of fuel-driven compressor units.

[0008] The beneficial effects of the knowledge graph-based fuel-driven compressor unit fault auxiliary decision-making method provided by the present invention are as follows:

[0009] It can assist the decision-making of technical and management personnel of the compressor station, quickly and accurately locate the source of faults and accurately control potential fault points, so as to achieve the purpose of effective prevention before faults and timely repair after faults, effectively improve the efficiency and ability of station personnel in independent repair and maintenance, reduce repair and maintenance costs, increase the trouble-free operation time of the compressor unit, and achieve cost reduction, efficiency improvement and safety improvement.

[0010] On the basis of the above scheme, the knowledge graph-based fuel-driven compressor unit fault auxiliary decision-making method of the present invention can also be improved as follows.

[0011] Furthermore, the process of building a knowledge graph includes:

[0012] Determine the entities of each type of fault, the relationship between the entities of each type of fault and other entities, standardize the core concepts and technical terms, and organize them into a structured knowledge base. The entities of any type of fault include: the fault phenomenon, fault phenomenon description, fault cause, consequence impact and disposal measures of this type of fault;

[0013] Entities of each type of fault, associations between entities of each type of fault and other entities, standardized core concepts and technical terms are extracted from the structured knowledge base, and the knowledge graph is constructed in the structured knowledge base.

[0014] Furthermore, determine the fault categories of the fuel-driven compressor unit and the fault phenomena, fault phenomenon description, fault causes, consequences and disposal measures of each type of fault, including:

[0015] Summarize and sort out the fault alarm information of the SCADA system and UCP system, and organize the technical documents of the fuel-driven compressor unit, the technical documents of the auxiliary equipment and facilities of the fuel-driven compressor unit, the existing expert knowledge and accumulated on-site disposal experience, determine the fault type of the fuel-driven compressor unit and the fault phenomenon, fault phenomenon description, fault cause, consequence and disposal measures of each type of fault.

[0016] Furthermore, it also includes:

[0017] Natural language processing technology is introduced to identify question intent, and a front-end is built to perform intelligent question-answering interaction and visual query through knowledge graphs.

[0018] The beneficial effects of adopting the previous technical solution are: introducing natural language processing technology and building a front-end, which makes it easier for station personnel to use and improves decision-making efficiency.

[0019] 2) In the second aspect, the present invention also provides a fuel-driven compressor unit fault auxiliary decision-making system based on knowledge graph, and the specific technical solution is as follows:

[0020] Includes knowledge graph creation module and fault decision module;

[0021] The knowledge graph creation module is used to: determine and construct a knowledge graph based on the fault categories of the fuel-driven compressor unit and the fault phenomena, fault phenomenon description, fault causes, consequences and disposal measures of each type of fault;

[0022] The fault decision module is used to make fault decisions on preset faults of fuel-driven compressor units using knowledge graphs.

[0023] On the basis of the above scheme, the knowledge graph-based fuel-driven compressor unit fault auxiliary decision-making system of the present invention can also be improved as follows.

[0024] Furthermore, the knowledge graph creation module is also specifically used for:

[0025] Determine the entities of each type of fault, the relationship between the entities of each type of fault and other entities, standardize the core concepts and technical terms, and organize them into a structured knowledge base. The entities of any type of fault include: the fault phenomenon, fault phenomenon description, fault cause, consequence impact and disposal measures of this type of fault;

[0026] Entities of each type of fault, associations between entities of each type of fault and other entities, standardized core concepts and technical terms are extracted from the structured knowledge base, and the knowledge graph is constructed in the structured knowledge base.

[0027] Furthermore, the knowledge graph creation module is also specifically used for:

[0028] Summarize and sort out the fault alarm information of the SCADA system and UCP system, and organize the technical documents of the fuel-driven compressor unit, the technical documents of the auxiliary equipment and facilities of the fuel-driven compressor unit, the existing expert knowledge and accumulated on-site disposal experience, determine the fault type of the fuel-driven compressor unit and the fault phenomenon, fault phenomenon description, fault cause, consequence and disposal measures of each type of fault.

[0029] Furthermore, an interaction module is included, and the interaction module is used to:

[0030] Natural language processing technology is introduced to identify question intent, and a front-end is built to perform intelligent question-answering interaction and visual query through knowledge graphs.

[0031] 3) In a third aspect, the present invention further provides an electronic device, comprising a processor, the processor being coupled to a memory, the memory storing at least one computer program, and the at least one computer program being loaded and executed by the processor, so that the electronic device can implement any of the above-mentioned knowledge graph-based fuel-driven compressor unit fault auxiliary decision-making methods.

[0032] 4) In a fourth aspect, the present invention further provides a computer-readable storage medium, in which at least one computer program is stored, and at least one computer program is loaded and executed by a processor so that a computer implements any of the above-mentioned knowledge graph-based fuel-driven compressor unit fault auxiliary decision-making methods.

[0033] It should be noted that the beneficial effects achieved by the technical solutions of the second to fourth aspects of the present invention and the corresponding possible implementation methods can be found in the above-mentioned technical effects of the first aspect and its corresponding possible implementation methods, and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Other features, objects and advantages of the present invention will become more apparent from the detailed description of non-limiting embodiments made with reference to the following drawings:

[0035] Figure 1 It is a flow chart of a method for auxiliary decision-making of a fuel-driven compressor unit fault based on a knowledge graph according to an embodiment of the present invention;

[0036] Figure 2 It is a structural schematic diagram of a fuel-driven compressor unit fault auxiliary decision-making system based on a knowledge graph according to an embodiment of the present invention;

[0037] Figure 3 It is a schematic diagram of constructing a knowledge graph-based auxiliary decision-making system for a fuel-driven compressor unit fault according to an embodiment of the present invention;

[0038] Figure 4 It is the entity relationship diagram of fault category;

[0039] Figure 5 It is a schematic diagram of the architecture of the knowledge graph-based auxiliary decision-making system for fuel-driven compressor unit faults of the present invention;

[0040] Figure 6 It is the interface of the knowledge graph-based fuel-driven compressor unit fault auxiliary decision-making system of the present invention;

[0041] Figure 7 Visualize the functional flow chart for the knowledge graph front end;

[0042] Figure 8 A diagram showing the knowledge graph visualization results;

[0043] Fig. 9 Retrieve functional flow charts for entities and relationships;

[0044] Fig.10 Display graph for entity search results;

[0045] Fig.11 Display graphs for relational search results;

[0046] Fig.12 This is the flow chart of the fault intelligent question and answer function;

[0047] Fig.13 This is a diagram showing the results of the intelligent question and answer query;

[0048] Fig.14 The figure is a schematic diagram of the structure of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0049] In order to make the objectives, technical solutions and advantages of the present invention more clear, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.

[0050] like Figure 1 As shown, a fuel-driven compressor unit fault auxiliary decision-making method based on a knowledge graph in an embodiment of the present invention includes the following steps:

[0051] S1. Determine and construct a knowledge graph based on the fault categories of fuel-driven compressor units and the fault phenomena, fault phenomenon descriptions, fault causes, consequences and disposal measures of each type of fault;

[0052] S2. Use the knowledge graph to make fault decisions for preset faults of the fuel-driven compressor unit.

[0053] Among them, the preset fault can be the current fault of the fuel-driven compressor unit, or it can be the fault specified by the technical and management personnel of the compressor station. Fault decision-making is carried out for the preset fault of the fuel-driven compressor unit, mainly identifying the cause of the fault and giving corresponding disposal measures, thereby assisting the technical and management personnel of the compressor station to make subsequent decisions.

[0054] Optionally, in S1, the process of constructing the knowledge graph includes:

[0055] S10. Determine the entities of each type of fault, the relationship between the entities of each type of fault and other entities, standardize core concepts and technical terms, and organize them into a structured knowledge base, where the entities of any type of fault include: the fault phenomenon of this type of fault, the description of the fault phenomenon, the cause of the fault, the consequences and disposal measures.

[0056] S11. Extracting entities of each type of fault, the relationship between entities of each type of fault and other entities, standard core concepts and technical terms from the structured knowledge base, and constructing the knowledge graph in the structured knowledge base.

[0057] The purpose of fault diagnosis of fuel-driven compressor units is to obtain an explanation from fault phenomenon to phenomenon description through pattern matching, and then find out the cause of the fault and the possible consequences, and finally propose treatment measures. Therefore, it is necessary to establish the relationship between the fault phenomenon and the fault phenomenon description, fault cause, consequence impact and treatment measures. The fault phenomenon is the core entity, and there are two types of relationships between it and the four entities of fault phenomenon description, fault cause, consequence impact and treatment measures, which means that other entities can be queried through known core entities, and core entities can also be reversely inferred through other known entities. By establishing the triple form of ("phenomenon", "relationship", "phenomenon description / cause / consequence / treatment measures"), the knowledge graph technology is introduced to perform entity fusion and construct the knowledge graph.

[0058] The structured knowledge base may use a Neo4j graph database. In this embodiment, Python is used for underlying programming to implement the following operations:

[0059] Use the Neo4j graph database to store the entities of each type of fault, the relationship between the entities of each type of fault and other entities, and standardize core concepts and technical terms; use Py2neo to create and store nodes and relationships between nodes, and realize the connection and operation of the Neo4j graph database.

[0060] It should be noted that:

[0061] The reason for choosing Neo4j graph database is that Neo4j graph database is the most commonly used graph database at present, with the advantages of high performance, strong practicality, and lightweight. It can also be easily accessed through Python language, realizing convenient query of knowledge graphs and convenient conversion of named entities.

[0062] Optionally, in S1, the fault category of the fuel-driven compressor unit and the fault phenomenon, fault phenomenon description, fault cause, consequence impact and disposal measures of each type of fault are determined, including:

[0063] S12. Summarize and sort out the fault alarm information of the SCADA system and the UCP system, and organize the technical documents of the fuel-driven compressor unit, the technical documents of the auxiliary equipment and facilities of the fuel-driven compressor unit, the existing expert knowledge and the accumulated on-site disposal experience, determine the fault category of the fuel-driven compressor unit and the fault phenomenon, fault phenomenon description, fault cause, consequence impact and disposal measures of each type of fault, specifically:

[0064] The daily troubleshooting and disposal of the compressor station is partly based on the fault alarm information issued by the SCADA system and the fault alarm information issued by the UCP system, and some fault information is discovered by daily inspections and regular maintenance and troubleshooting. Fault alarm information and fault information contain complex affiliations and causal relationships. A fault alarm information and fault information may contain multiple causes, and may correspond to different fault handling methods (handling measures). In view of complex fault scenarios and knowledge dimensions, inexperienced on-duty personnel are often helpless and can only inform people with fault handling capabilities to handle them, thereby delaying the time for fault handling and increasing the risk of sudden accidents; and experienced on-duty personnel can only deal with faults that have occurred before, and are powerless for faults that have not occurred. In the present invention, a comprehensive summary of the two types of situations, fault alarm information and fault information, is integrated to effectively solve the actual needs on site.

[0065] Optionally, in the above technical solution, it also includes:

[0066] S3. Introduce natural language processing technology to identify question intent and build a front-end to perform intelligent question-answering interaction and visual query through knowledge graphs.

[0067] After the knowledge graph is built, the query of the knowledge graph can only be realized in the Neo4j graph database, and the user-specified questions cannot be matched and searched. In this embodiment, the natural language processing technology introduced is specifically the Aho-Corasick multi-mode matching algorithm, and the user intention recognition is implemented based on the Aho-Corasick multi-mode matching algorithm. The Aho-Corasick multi-mode matching algorithm is an algorithm for string search, which can efficiently find multiple pattern strings / strings in a text. The algorithm principle is to first build a Trie tree for storing all strings. When the character matching on a node fails, the algorithm will jump to another node through an invalid pointer to continue matching, instead of starting from scratch, and practice efficient search and matching of strings, avoiding repeated judgments, and realizing accurate classification of questions and answer matching by solving the similarity between the input sentence phrase and the entity feature word.

[0068] The front end constructed can be a web front end or other front ends to facilitate intelligent interaction of questions and answers and visual query.

[0069] In another embodiment, comprising:

[0070] S101. Summarize and sort out the alarm information of the SCADA system and UCP (compressor unit control panel) system, organize the technical documents of the fuel-driven compressor unit and its auxiliary equipment and facilities, the existing expert knowledge and accumulated on-site handling experience and other relevant information, and clarify the category of fault information, the cause of occurrence, the consequence and impact, and the fault handling method;

[0071] S102. Clarify the entity relationship categories that need to be established, standardize core concepts and terms, and organize them into an offline structured knowledge base;

[0072] S103, defining and extracting entity types, attributes, and relationships between entities based on the constructed offline structured knowledge base;

[0073] S104. Create a knowledge graph to realize intelligent retrieval and visualization of fault information;

[0074] S105. Introduce natural language processing technology to realize question intent recognition and entity recognition, build a lightweight web front-end, and realize intelligent question-answering interaction and visual query.

[0075] The present invention comprehensively sorts out the fault types, causes and handling methods of fuel-driven compressor units, forms a fault diagnosis knowledge map, introduces natural language processing technology, and develops a fuel-driven compressor unit fault decision-making auxiliary system to assist compressor station technology and management personnel in decision-making, thereby achieving rapid and accurate positioning of fault sources and precise control of potential fault points, achieving the purpose of effective prevention before failure and timely repair after failure, effectively improving the efficiency and ability of station personnel in independent repair and maintenance, reducing repair and maintenance costs, and increasing the trouble-free operation time of compressor units, achieving cost reduction, efficiency improvement and safety improvement.

[0076] In the above embodiments, although the steps are numbered S1, S2, etc., these are only specific embodiments given by the present invention. Those skilled in the art may adjust the execution order of S1, S2, etc. according to actual conditions, which is also within the protection scope of the present invention. It can be understood that in some embodiments, some or all of the above embodiments may be included.

[0077] like Figure 2 As shown, a knowledge graph-based fuel-driven compressor unit fault auxiliary decision system 200 according to an embodiment of the present invention includes a knowledge graph creation module 201 and a fault decision module 202;

[0078] The knowledge graph creation module 201 is used to: determine and construct a knowledge graph based on the fault category of the fuel-driven compressor unit and the fault phenomenon, fault phenomenon description, fault cause, consequence impact and disposal measures of each type of fault;

[0079] The fault decision module 202 is used to make a fault decision on a preset fault of a fuel-driven compressor unit by using a knowledge graph.

[0080] Optionally, in the above technical solution, the knowledge graph creation module 201 is further specifically used for:

[0081] Determine the entities of each type of fault, the relationship between the entities of each type of fault and other entities, standardize the core concepts and technical terms, and organize them into a structured knowledge base. The entities of any type of fault include: the fault phenomenon, fault phenomenon description, fault cause, consequence impact and disposal measures of this type of fault;

[0082] The entities of each type of fault, the relationship between the entities of each type of fault and other entities, the standardized core concepts and technical terms are extracted from the structured knowledge base to construct a knowledge graph.

[0083] Optionally, in the above technical solution, the knowledge graph creation module 201 is further specifically used for:

[0084] Summarize and sort out the fault alarm information of the SCADA system and UCP system, and organize the technical documents of the fuel-driven compressor unit, the technical documents of the auxiliary equipment and facilities of the fuel-driven compressor unit, the existing expert knowledge and accumulated on-site disposal experience, determine the fault type of the fuel-driven compressor unit and the fault phenomenon, fault phenomenon description, fault cause, consequence and disposal measures of each type of fault.

[0085] Optionally, in the above technical solution, an interaction module is further included, and the interaction module is used to:

[0086] Natural language processing technology is introduced to identify question intent, and a front-end is built to perform intelligent question-answering interaction and visual query through knowledge graphs.

[0087] like Figure 3 As shown in the figure, it mainly includes the establishment of an offline structured fault library, entity and relationship feature extraction, creation of a knowledge graph, and development of a web front end. The establishment of a structured fault library is the prerequisite for building a knowledge graph. The daily troubleshooting and disposal of the compressor station is partly based on the alarm information of the SCADA system and the UCP system, and some fault information is discovered by daily inspections and regular maintenance and troubleshooting. Alarm information and fault information contain complex affiliations and causal relationships. An alarm information and fault information may contain multiple causes, and correspond to different fault handling methods. In view of complex fault scenarios and knowledge dimensions, inexperienced on-duty personnel are often helpless and can only inform people with fault handling capabilities to handle them, thereby delaying the time for fault handling and increasing the risk of sudden accidents; and experienced on-duty personnel can only deal with faults that have occurred before, and are powerless for faults that have not occurred. Therefore, the fault library integrates a comprehensive summary of the two types of situations, alarm information and fault information, to effectively solve the actual needs of the site.

[0088] In this embodiment, the entity relationship of the fault category of the knowledge graph of the fuel-driven compressor unit is as follows: Figure 4As shown in the figure, the purpose of fault diagnosis of fuel-driven compressor units is to obtain an explanation from fault phenomenon to phenomenon description through pattern matching, and then find out the cause of the fault and the possible consequences, and finally propose treatment measures. Therefore, it is necessary to establish the relationship between the fault phenomenon and the phenomenon description, the fault cause, the consequence impact and the treatment measures. The fault phenomenon is the core entity, and there are two types of relationships between it and the four entities of phenomenon description, fault cause, consequence impact and treatment measures, including inclusion and subordination, which means that the core entity can query other entities when it is known, and the core entity can be reversely inferred from other known entities. By establishing the triple form of ("phenomenon", "relationship", "phenomenon description / cause / consequence / treatment measures"), the knowledge graph technology is introduced to perform entity fusion.

[0089] In this embodiment, the knowledge graph of fuel-driven compressor unit faults is constructed by using Python for underlying programming, Neo4j graph database for knowledge storage, Py2neo for node and relationship creation, and connection and operation of Neo4j graph database. The reason for selecting Neo4j graph database is that Neo4j is the most commonly used graph database, with the advantages of high performance, strong practicality, and light weight, and can be easily accessed through Python language, realizing convenient conversion of query graphs and named entities.

[0090] After the knowledge graph is built, it can only be queried in the Neo4j graph database, and it is not possible to match and search for user questions. In this embodiment, user intent recognition is implemented based on the Aho-Corasick multi-pattern matching algorithm. The Aho-Corasick algorithm is an algorithm for string search, which can efficiently find multiple pattern strings / strings in a text. The algorithm principle is to first build a Trie tree to store all strings. When the character match on a node fails, the algorithm will jump to another node through an invalid pointer to continue matching, instead of starting from scratch, to efficiently search and match the internship string, avoid repeated judgments, and achieve accurate classification of questions and answer matching by solving the similarity between the input sentence phrase and the entity feature word.

[0091] like Figure 5 As shown in the figure, the system architecture mainly consists of five parts, namely the basic technology layer, data layer, model layer, business layer and display layer. Specifically:

[0092] 1) Basic technology layer: Provides basic technology for the system, lays the foundation for the development of upper-level systems, develops underlying code based on Python, stores knowledge based on the Neo4j graph database, and implements Python connection and operation of the Neo4j graph database based on the Python third-party package Py2neo.

[0093] 2) Data layer: It provides the data support foundation for the system, including data acquisition, ontology construction, knowledge extraction, and knowledge storage. It obtains knowledge entities and constructs knowledge triples through named entity recognition, and then stores the triples in the Neo4j graph database, and implements node and relationship matching queries through Python+Cypher.

[0094] 3) Model layer: It includes question intention recognition model and question named entity recognition model. Question intention recognition is used to interpret user questions to obtain question intent. Question named entity recognition is used to extract knowledge entities related to fuel-driven compressor unit failure in user questions to provide answers.

[0095] 4) Business layer: Implement the system question-answering function based on the basic functions of the knowledge graph and user demand analysis, obtain the entities and intentions identified by the model layer to form question triples, perform queries and retrieval in the knowledge graph based on Cypher, and return the answers required by users.

[0096] 5) Display layer: HTML and CSS technologies are used to develop the Web front-end display interface, and the Django framework is used to quickly build it, so as to realize entity relationship query, question-answer interaction and knowledge graph visualization. The Django framework is a Python-based Web development framework that can quickly develop secure and maintainable web pages. It has the advantages of strong flexibility, easy expansion, security and reliability.

[0097] In this embodiment, the interface of the fuel-driven compressor unit fault auxiliary decision-making system based on the knowledge graph is as follows: Figure 6 As shown in the figure, the four main functions of the system are listed on the far left of the interface, namely entity retrieval, relationship retrieval, automatic question and answer, and knowledge graph display. Click the corresponding function and enter the content to be searched in the text box on the right to achieve question matching and intelligent question and answer. Some common questions are floating in a circle on the right side of the interface. Users can quickly answer questions by clicking the corresponding floating text.

[0098] The overall visualization of the knowledge graph is used to display the entity categories and overall relationships in the current fault library on the system interface. The internal operation process of the system is as follows: Figure 7 To ensure the smooth operation of the system, this system sets the number of display nodes to 100. If you need to obtain a visual display of all nodes and relationships, you can log in to the backend to view the Neo4j graph database. The front-end display of the knowledge graph is as follows: Figure 8 shown.

[0099] In this embodiment, the flowchart of entity and relationship retrieval is as follows: Fig. 9As shown in the figure, the user inputs the relevant fault entity or relationship, and the program automatically queries through Cypher statements, searches for relevant entity nodes and their relationship information from the Neo4j graph database, and extracts entities and relationships. If the query is successful, the subgraph is returned, and if the query fails, no subgraph is returned. For example, Fig.10 As shown in the figure, enter "intake filter pressure difference high alarm" in the entity search box, and the result visualization shows the entity node and all node information related to the entity node. Fig.11 As shown in the figure, enter "reason" in the relationship search box, and the result visualization shows all nodes with the "reason" relationship. Similarly, you can perform a quick search by clicking the floating text on the right side of the interface.

[0100] In this embodiment, the flowchart of the fault intelligent question and answer implementation is as follows: Fig.12 As shown in the figure, the user inputs a natural language question, and the system automatically classifies the question intent and extracts keywords, and performs entity and relationship query matching. If the query is successful, the answer to the question is returned, and if the query fails, the fixed words are returned. For example, Fig.13 As shown, enter "How to deal with metal chip warnings in lubricating oil pipelines" in the automatic question and answer search box, and the result shows "Handling measures for metal chip warnings in lubricating oil pipelines: immediately apply for shutdown, dismantle the metal chip detector for inspection, if metal chips are really present, check the transmission box, engine bearing components, and clean impurities in the lubricating oil pipeline", and visualize the entity relationship.

[0101] It should be noted that the beneficial effects of the knowledge graph-based fuel-driven compressor unit fault auxiliary decision system 200 provided in the above embodiment are the same as the beneficial effects of the knowledge graph-based fuel-driven compressor unit fault auxiliary decision method, which will not be repeated here. In addition, when the system provided in the above embodiment realizes its functions, it only uses the division of the above-mentioned functional modules as an example. In actual applications, the above-mentioned functions can be assigned to different functional modules as needed, that is, the system can be divided into different functional modules according to actual conditions to complete all or part of the functions described above. In addition, the system and method embodiments provided in the above embodiment belong to the same concept, and the specific implementation process is detailed in the method embodiment, which will not be repeated here.

[0102] like Fig.14 As shown, an electronic device 300 according to an embodiment of the present invention includes a processor 320, the processor 320 is coupled to a memory 310, and the memory 310 stores at least one computer program 330, and the at least one computer program 330 is loaded and executed by the processor 320, so that the electronic device 300 implements any of the above-mentioned fuel-driven compressor unit fault auxiliary decision-making methods based on knowledge graphs, specifically:

[0103] The electronic device 300 may have relatively large differences due to different configurations or performances, and may include one or more processors 320 (Central Processing Units, CPU) and one or more memories 310, wherein the one or more memories 310 store at least one computer program 330, and the at least one computer program 330 is loaded and executed by the one or more processors 320, so that the electronic device 300 implements any of the fuel-driven compressor unit fault auxiliary decision-making methods based on knowledge graphs provided in the above embodiments. Of course, the electronic device 300 may also have components such as a wired or wireless network interface, a keyboard, and an input and output interface for input and output. The electronic device 300 may also include other components for realizing device functions, which will not be repeated here. The electronic device may specifically be a computer, etc.

[0104] A computer-readable storage medium according to an embodiment of the present invention stores at least one computer program, and the at least one computer program is loaded and executed by a processor so that the computer implements any of the above-mentioned knowledge graph-based fuel-driven compressor unit fault auxiliary decision-making methods.

[0105] Optionally, the computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a compact disc (CD-ROM), a magnetic tape, a floppy disk, an optical data storage device, and the like.

[0106] In an exemplary embodiment, a computer program product or a computer program is also provided, the computer program product or the computer program comprising computer instructions, the computer instructions being stored in a computer-readable storage medium. A processor of an electronic device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the electronic device executes any of the above-mentioned fuel-driven compressor unit fault auxiliary decision-making methods based on knowledge graphs.

[0107] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application are used to distinguish similar objects and represent the definition of a specific order or sequence. The order of use of similar objects can be interchanged where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than the order shown or described.

[0108] Those skilled in the art know that the present invention can be implemented as a system, method or computer program product. Therefore, the present invention can be specifically implemented in the following forms, namely: it can be complete hardware, it can be complete software (including firmware, resident software, microcode, etc.), or it can be a combination of hardware and software, which is generally referred to as "circuit", "module" or "system" herein. In addition, in some embodiments, the present invention can also be implemented in the form of a computer program product in one or more computer-readable media, and the computer-readable medium contains computer-readable program code.

[0109] Any combination of one or more computer-readable media can be used. The computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media (a non-exhaustive list) include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present invention, a computer-readable storage medium can be any tangible medium containing or storing a program, which can be used by an instruction execution system, device or device or used in combination with it.

[0110] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations of the present invention. A person skilled in the art may change, modify, replace and vary the above embodiments within the scope of the present invention.

Claims

1. A knowledge graph-based auxiliary decision-making method for fuel-driven compressor unit faults, characterized in that: include: Determine and construct a knowledge graph based on the fault categories of fuel-driven compressor units and the fault phenomena, fault phenomenon descriptions, fault causes, consequences and disposal measures of each type of fault; The knowledge graph is used to make fault decisions for preset faults of the fuel-driven compressor unit.

2. According to claim 1, a knowledge graph-based auxiliary decision-making method for fuel-driven compressor unit faults is characterized in that: The process of building a knowledge graph includes: Determine the entities of each type of fault, the relationship between the entities of each type of fault and other entities, standardize the core concepts and technical terms, and organize them into a structured knowledge base. The entities of any type of fault include: the fault phenomenon, fault phenomenon description, fault cause, consequence impact and disposal measures of this type of fault; Entities of each type of fault, associations between entities of each type of fault and other entities, standardized core concepts and technical terms are extracted from the structured knowledge base, and the knowledge graph is constructed in the structured knowledge base.

3. The method for auxiliary decision-making of fuel-driven compressor unit fault based on knowledge graph according to claim 1 is characterized in that: Determine the fault categories of fuel-driven compressor units and the fault phenomena, fault phenomenon descriptions, fault causes, consequences and disposal measures for each type of fault, including: Summarize and sort out the fault alarm information of the SCADA system and UCP system, and organize the technical documents of the fuel-driven compressor unit, the technical documents of the auxiliary equipment and facilities of the fuel-driven compressor unit, the existing expert knowledge and accumulated on-site disposal experience, determine the fault type of the fuel-driven compressor unit and the fault phenomenon, fault phenomenon description, fault cause, consequence and disposal measures of each type of fault.

4. A fuel-driven compressor unit fault auxiliary decision-making method based on knowledge graph according to any one of claims 1 to 3, characterized in that: Also includes: Natural language processing technology is introduced to identify question intent, and a front-end is built to perform intelligent question-answering interaction and visual query through knowledge graphs.

5. A knowledge graph-based fuel-driven compressor unit fault auxiliary decision system, characterized in that: Includes knowledge graph creation module and fault decision module; The knowledge graph creation module is used to: determine and construct a knowledge graph based on the fault category of the fuel-driven compressor unit and the fault phenomenon, fault phenomenon description, fault cause, consequence impact and disposal measures of each type of fault; The fault decision module is used to make a fault decision on a preset fault of a fuel-driven compressor unit by using the knowledge graph.

6. The knowledge graph-based fuel-driven compressor unit fault auxiliary decision system according to claim 5 is characterized in that: The knowledge graph creation module is also specifically used for: Determine the entities of each type of fault, the relationship between the entities of each type of fault and other entities, standardize the core concepts and technical terms, and organize them into a structured knowledge base. The entities of any type of fault include: the fault phenomenon, fault phenomenon description, fault cause, consequence impact and disposal measures of this type of fault; Entities of each type of fault, associations between entities of each type of fault and other entities, standardized core concepts and technical terms are extracted from the structured knowledge base, and the knowledge graph is constructed in the structured knowledge base.

7. The knowledge graph-based fuel-driven compressor unit fault auxiliary decision system according to claim 5 is characterized in that: The knowledge graph creation module is also specifically used for: Summarize and sort out the fault alarm information of the SCADA system and UCP system, and organize the technical documents of the fuel-driven compressor unit, the technical documents of the auxiliary equipment and facilities of the fuel-driven compressor unit, the existing expert knowledge and accumulated on-site disposal experience, determine the fault type of the fuel-driven compressor unit and the fault phenomenon, fault phenomenon description, fault cause, consequence and disposal measures of each type of fault.

8. A knowledge graph-based fuel-driven compressor unit fault auxiliary decision system according to any one of claims 5 to 7, characterized in that: It also includes an interaction module, wherein the interaction module is used to: Natural language processing technology is introduced to identify question intent, and a front-end is built to perform intelligent question-answering interaction and visual query through knowledge graphs.

9. An electronic device, characterized in that: The electronic device includes a processor, the processor is coupled to a memory, at least one computer program is stored in the memory, and the at least one computer program is loaded and executed by the processor so that the electronic device implements a knowledge graph-based fuel-driven compressor unit fault auxiliary decision-making method as described in any one of claims 1 to 4.

10. A computer-readable storage medium, characterized in that: At least one computer program is stored in the computer-readable storage medium, and the at least one computer program is loaded and executed by the processor so that the computer implements a knowledge graph-based auxiliary decision-making method for fuel-driven compressor unit faults as described in any one of claims 1 to 4.

Images