A wind turbine operation and maintenance management method based on a knowledge graph

Abstract

The application provides a wind turbine operation and maintenance management method based on a knowledge graph, comprising the following steps: performing text content and feature analysis and preprocessing, dividing entity types according to the similarities and differences of entity properties, giving constraint definitions for the relationships of entity types at various levels, and extracting entity words for wind farm operation and maintenance by using a rule matching algorithm based on an improved TextRank; defining the relationships between various entities according to field texts, and realizing the relationship linkage between different entities by using an entity relationship extraction algorithm based on rule matching; constructing a safety management procedure graph and a device operation and maintenance requirement graph established based on a protégé wind power equipment ontology instance topological relationship library; realizing the visualization of the knowledge graph for wind turbine operation and maintenance management, and forming an application scheme by using a query language and a graph embedding technology. Compared with the prior art, the application has the advantages of high accuracy, high efficiency and strong field professional application.

Description

Technical Field

[0001] This invention relates to the field of wind turbine operation and maintenance management technology, and in particular to a wind turbine operation and maintenance management method based on knowledge graphs. Background Technology

[0002] Against the backdrop of the development of new power systems, the installed capacity and complexity of wind turbines are constantly increasing. Meanwhile, due to varying skill levels among maintenance personnel, improper maintenance accounts for as much as 32.5% of accidents. As the maintenance lifespan of wind farms increases and the frequency of maintenance rises, relying solely on manual storage and correlation of maintenance standards becomes increasingly impractical. However, leveraging knowledge graph mining and analysis capabilities to connect large-scale and fragmented data can effectively correlate wind power maintenance standards information, thereby improving the level of intelligent operation and maintenance in wind farms.

[0003] However, in the construction of knowledge graphs for wind farm operation and maintenance management, current general methods still suffer from insufficient semantic understanding and inaccurate keyword positioning in text segmentation and relation extraction within the wind power operation and maintenance field. Furthermore, in the practical application of knowledge graphs for wind farm operation and maintenance management, research is needed on operation and maintenance fault classification, operation and maintenance solution question-and-answer, and recommendation strategies. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of the existing technology and provide an accurate, efficient, and highly applicable knowledge graph-based wind turbine operation and maintenance management method.

[0005] The objective of this invention can be achieved through the following technical solutions:

[0006] A knowledge graph-based method for wind turbine operation and maintenance management, the method comprising the following steps:

[0007] Based on the entity extraction algorithm for the operation and maintenance management of wind turbine generator sets, entity types are classified according to the similarities and differences of entity concepts and given constraint definitions. An improved TextRank-based rule matching algorithm is used to extract entity words for wind farm operation and maintenance.

[0008] Based on the entity relationship extraction algorithm for the operation and maintenance management of wind turbine generator sets, the relationship between various entities is defined according to the domain text, and the entity relationship is extracted using a rule matching algorithm.

[0009] To address the application requirements of wind turbine generator operation and maintenance management, a domain knowledge graph is constructed, including a safety management procedure graph and an equipment operation and maintenance requirement graph based on the Protégé wind power equipment ontology instance topology relation library.

[0010] Visualize the knowledge graph for wind turbine operation and maintenance management, and use query language to generate operation and maintenance management solutions.

[0011] Furthermore, the entity extraction specifically includes the following steps:

[0012] Obtain a standard text set for wind farm operation and maintenance management, preprocess the standard text set, use a word segmentation tool to segment and tag each sentence with part-of-speech tags and remove stop words, and introduce an external thesaurus in the wind power field as a custom dictionary into the word segmentation tool;

[0013] Entity categories are defined by classifying entity types based on domain text features and providing constraint definitions.

[0014] This study aims to obtain entity vocabulary in the field of wind power operation and maintenance, propose corresponding extraction schemes for different entity categories, and complete entity vocabulary extraction based on the improved TextRank rule matching algorithm.

[0015] Furthermore, the entity type classification is specifically as follows:

[0016] In general domains, common entity description information is used, including natural language definitions of place names, personal names, and organization names;

[0017] For the wind power operation and maintenance field, targeted entity description information is used, and entity categories are defined as: work items, specification objects, status phenomena, and safety requirements.

[0018] Furthermore, the rule matching algorithm based on the improved TextRank is specifically as follows:

[0019] The improved TextRank algorithm was used to extract keywords from the standardized text, and a rule matching algorithm was combined to filter and extract entities of work projects and standardized object categories.

[0020] The rule matching algorithm designs extraction rules based on the structural features of different categories of entity concepts in the text. The extracted rules are as follows:

[0021] Entity=[Strstart]+…+[Strend]

[0022] In the formula, [Strstart] represents the start symbol; [Strend] represents the end symbol; the start symbol and the end symbol define the entity boundary, and the content in between is the entity concept. When the start symbol does not have a corresponding end symbol, punctuation marks such as periods and commas are used as natural end symbols.

[0023] The improved TextRank algorithm is as follows:

[0024] A domain dictionary is constructed as the core vocabulary, and the TextRank network graph model is initialized. The core vocabulary is introduced as an external variable. The TextRank values ​​are iteratively calculated and sorted, and the key words of the domain text are extracted according to the order.

[0025] Furthermore, the entity relationship specifically includes the following steps:

[0026] Based on the domain text features, we define the relationships between various entities, adopt a rule-based entity relationship extraction algorithm, design extraction rules to extract relationships, and realize the relationship links between different entities.

[0027] The rule-based entity relationship extraction algorithm is as follows:

[0028] The extracted entity nouns are classified to form entity dictionaries of different categories, which, together with the text to be identified, serve as input for the matching template.

[0029] Check if the number of entity nouns in the input text is greater than 2. If not, the input text is irrelevant.

[0030] Define the relationships between various entities, perform rule template matching, determine the corresponding relationships between entities based on their categories, and output the entity relationships.

[0031] Furthermore, the aforementioned safety management procedure diagram is as follows:

[0032] To address the requirement of associated storage in operation and maintenance management, a knowledge graph of security management procedures is constructed by combining the aforementioned entity extraction and relationship extraction algorithms.

[0033] The work items are used as the central nodes of the diagram, and the requirements for carrying out different types of work items are used as branches to express the relationship between various safety management procedures in the standard text.

[0034] The construction steps are as follows: a domain dictionary is formed through text preprocessing, which further assists in the extraction of entity vocabulary; entity relations are extracted using rule matching methods, the relationship links between entities are completed, triple information is formed, and the knowledge graph is constructed.

[0035] Furthermore, the specific requirements map for the construction equipment operation and maintenance is as follows:

[0036] To address the need for the integration and connection of the operating specifications of various equipment in a wind farm and the subordinate information of its components, a knowledge graph of wind farm equipment operation and maintenance requirements is constructed by combining the aforementioned entity extraction and relationship extraction algorithms.

[0037] By organizing and summarizing the components and equipment of wind farms, and storing them using Protégé, a graphical representation of the equipment's structural composition is created. Combined with the corresponding safety regulations, a graphical representation of the operation, maintenance, and repair requirements for wind farm equipment is constructed.

[0038] The construction steps are as follows: First, combine prior knowledge with Protégé to store instances of wind turbine generator equipment and components to form a wind farm equipment topology information database. Second, perform knowledge fusion with the extracted equipment and component entities to match the same entity nodes and establish the subordinate relationships between each equipment and component entity through the topology information database. Third, store the work items and operation specifications associated with each node through a safety management procedure graph. Fourth, form a wind farm equipment operation and maintenance requirement graph with equipment and components as the central nodes and subordinate relationships as branches.

[0039] Furthermore, the knowledge graph visualization for wind turbine generator operation and maintenance management, and the formation of operation and maintenance management solutions using query language, includes the following steps:

[0040] Based on the Neo4j graph database, the storage and interconnection of wind power operation and maintenance ternary information is completed to realize knowledge graph visualization;

[0041] This paper utilizes Neo4j's built-in Cypher query language to develop a knowledge graph-based intelligent information query and operation and maintenance support solution for efficient decision-making.

[0042] Furthermore, the visualization of the knowledge graph specifically includes:

[0043] Neo4j is used to store wind power operation and maintenance data with multiple interconnected relationships;

[0044] By adopting a labeled attribute graph model, knowledge entities are stored as nodes, and node labels represent the entity class to which they belong. By assigning multiple labels at the same time to represent knowledge entities of different types, the multi-dimensional interconnection of information in the field of wind power operation and maintenance can be realized.

[0045] Furthermore, the aforementioned intelligent information query and operation and maintenance-assisted efficient decision-making application solution specifically includes:

[0046] By inputting questions through the human-computer interaction interface, the question parsing module maps the input statement into a Cypher query statement of the Neo4j graph database. Based on the graph information retrieval, the answer generation module outputs the answer according to the retrieval results, realizing intelligent query of operation and maintenance information.

[0047] The equipment operation and maintenance requirements map enables precise location of equipment structure and topology information, while the safety management procedure map guides the safe conduct of operation and maintenance work, realizing the functions of process judgment, logic generation, and solution recommendation for wind farm operation and maintenance management.

[0048] Compared with the prior art, the present invention has the following beneficial effects:

[0049] 1) Propose a knowledge graph construction scheme that combines safety management procedures and equipment operation and maintenance requirements. The scheme describes the operation and maintenance requirements of wind farms from the perspectives of work project correlation and equipment subordinate structure correlation, thereby improving the ability of the graph to store and express the correlation of wind power operation and maintenance information.

[0050] 2) The present invention uses a rule matching algorithm based on an improved TextRank to complete entity extraction. By combining the rule matching algorithm with the introduction of a core vocabulary mechanism as external information, the target entity vocabulary within the domain can be extracted more accurately, effectively reducing the manual workload of entity extraction.

[0051] 3) Based on Neo4j graph data, a visualization and application scheme for knowledge graphs is proposed. The data is stored in a graph structure, which can directly and accurately reflect the internal structural correlation of operation and maintenance specifications. At the same time, based on its built-in Cypher language and graph embedding technology, knowledge can be queried, deeply mined and reasoned. Compared with other types of databases, the constructed knowledge graph has the advantages of high performance, fast access speed and lightweight.

[0052] 4) Applying knowledge graph technology to the research on intelligent operation and maintenance of wind farms, a knowledge graph and application method suitable for wind farm operation and maintenance specifications are proposed. This provides guidance and assistance for information such as wind farm safety management procedures, equipment operation and maintenance requirements, and fault diagnosis and repair, thereby improving the informatization and intelligent management level of wind farm operation and maintenance. Attached Figure Description

[0053] Figure 1 This is a flowchart of the method of the present invention;

[0054] Figure 2 This is an example diagram illustrating the relationship between security management procedures in the embodiments;

[0055] Figure 3 This is the process for constructing a security management procedure diagram in the embodiment;

[0056] Figure 4 This is an example diagram illustrating the relationship between equipment operation and maintenance requirements in the embodiments;

[0057] Figure 5 This is the process for constructing the equipment operation and maintenance requirements map in the embodiment;

[0058] Figure 6 This is an architecture diagram of the wind power operation and maintenance knowledge graph application scheme in the embodiment. Detailed Implementation

[0059] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments. These embodiments are based on the technical solution of the present invention and provide detailed implementation methods and specific operating procedures. However, the scope of protection of the present invention is not limited to the following embodiments.

[0060] Example 1

[0061] Step S1: Design an entity extraction algorithm for the operation and maintenance management of wind turbine generators. Entity types are categorized based on the similarities and differences in entity concepts, and constraint definitions are given. An improved TextRank-based rule matching algorithm is used to extract entity vocabulary for wind farm operation and maintenance. This includes the following sub-steps:

[0062] Step S11: Obtain the standard text set for wind farm operation and maintenance management, preprocess the standard text set, use the jieba word segmentation tool to segment and tag each sentence with part of speech, design and remove stop words, and at the same time, in order to improve the word segmentation effect of professional vocabulary, introduce an external thesaurus in the wind power field as a custom dictionary into the jieba word segmentation tool.

[0063] Step S12: Classify entity types, classify entity types according to domain text features and give constraint definitions;

[0064] The entity type classification, in the general domain, uses common entity description information such as natural language definitions of place names, personal names, and organization names;

[0065] Specifically, for the specific domain of this method, namely the wind power operation and maintenance domain, targeted entity description information is used to define entity categories including work items, specification objects, status phenomena, and safety requirements.

[0066] Step S13: Obtain entity words in the field of wind power operation and maintenance, provide corresponding extraction schemes for different entity categories, and propose a rule matching algorithm based on improved TextRank to complete entity word extraction;

[0067] The improved TextRank-based rule-matching entity extraction algorithm extracts entity categories of state phenomena and security requirements using a rule-matching algorithm, extracts keywords from the normative text using an improved TextRank algorithm, and combines the rule-matching algorithm to filter and complete the entity extraction of work projects and normative object categories.

[0068] Specifically, as shown in Table 1, extraction rules are designed based on the structural features of different categories of entity concepts in the text. The extraction rules are as follows:

[0069] Entity=[Strstart]+…+[Strend]

[0070] In the formula: [Strstart] represents the start symbol; [Strend] represents the end symbol. The start symbol and the end symbol define the entity boundary, and the content in between is the entity concept. When the start symbol does not have a corresponding end symbol (denoted as None), punctuation marks such as periods and commas are used as natural end symbols.

[0071] Table 1. Entity Extraction Rules (Partial)

[0072]

[0073] Specifically, the improved TextRank method first needs to construct a domain dictionary as core vocabulary and initialize the TextRank network graph model. Furthermore, the core vocabulary is introduced as an external variable to increase the weight of the domain vocabulary. Finally, the TextRank values ​​are iteratively calculated and sorted, and the key words of the domain text are extracted according to the order.

[0074] Step S2: Design an entity relationship extraction algorithm for the wind turbine generator operation and maintenance management field. Define the relationships between various entities based on the domain text, and use a rule-based matching algorithm to extract the relationships. This includes the following sub-steps:

[0075] Step S21: Define the relationships between various entities based on the domain text features;

[0076] Step S22: Use a rule-based entity relationship extraction algorithm to design extraction rules to extract relationships and realize relationship links between different entities;

[0077] The rule-based entity relationship extraction algorithm requires classifying the extracted entities to form entity dictionaries of different categories, which, together with the text to be identified, serve as input for the matching template.

[0078] Specifically, the system checks whether the number of entity nouns in the input text is greater than 2. If not, the input text is considered irrelevant.

[0079] As shown in Table 2, rule template matching is performed to define the relationships between various entities, the corresponding relationships between entities are determined according to the entity categories, and the entity relationships are output.

[0080] Table 2. Entity Relationship Extraction Rules (Partial)

[0081] Head Entity A Tail Entity B Relationship Category Work Projects Work Projects A contains B / has no relation Work Projects Standard Object A is unrelated to B. Standard Object Standard Object A contains B / has no relation Standard Object State phenomena A faces B / No relation Standard Object Safety requirements A should conform to B / No relation State phenomena Safety requirements A should conform to B / No relation Safety requirements Technical parameters Technical parameter A is unrelated to B.

[0082] Step S3: Construct a domain knowledge graph for the application requirements of wind turbine generator operation and maintenance management, and propose a construction process for the safety management procedure graph and a construction process for the equipment operation and maintenance requirement graph based on the Protégé wind power equipment ontology instance topology relation library. Specifically, this includes the following sub-steps:

[0083] Step S31: Design a knowledge graph of security management procedures to meet the needs of operation and maintenance management requirements related to wind farm production, operation, human resources, materials, security measures, etc., and propose a construction scheme based on the aforementioned knowledge extraction algorithm;

[0084] like Figure 2 As shown, the safety management procedure knowledge graph takes work items as the central node of the graph and the requirements for carrying out different types of work items as branches, thereby expressing the relationship between various safety management procedures in the standard text;

[0085] like Figure 3 As shown, the construction steps include: forming a domain dictionary through text preprocessing; further assisting in entity word extraction using the domain dictionary; further, extracting entity relationships based on rule matching; completing the relationship links between entities; and forming triple information to achieve graph construction.

[0086] Step S32: Design a knowledge graph of wind farm equipment operation and maintenance requirements for the integration and connection of the work specifications for the installation, acceptance, operation, inspection and maintenance of various equipment in the wind farm and the subordinate information of its components, and propose a construction scheme in combination with the aforementioned knowledge extraction algorithm;

[0087] like Figure 4 As shown, the equipment operation and maintenance requirements knowledge graph organizes and summarizes the components of wind farm operating equipment and stores them using Protégé to form a graph information describing the equipment structure. It is then further combined with the corresponding safety rules of the equipment to build a wind farm equipment operation, maintenance and repair requirements graph. This can help operators have a comprehensive understanding of equipment operation and maintenance work projects and improve work efficiency.

[0088] like Figure 5 As shown, the construction steps include: combining prior knowledge with Protégé to complete the storage of wind turbine generator equipment and component instances to form a wind farm equipment topology information database; performing knowledge fusion with the extracted equipment component entities; matching the same entity nodes; and establishing the subordinate relationship between each equipment component entity through the topology information database; further, storing the work items and operation specifications associated with each node through a safety management procedure graph; and finally forming a wind farm equipment operation and maintenance requirement graph with equipment components as the central node and subordinate relationships as branches.

[0089] Step S4: Implement knowledge graph visualization for wind turbine generator operation and maintenance management based on Neo4j. The application solution is formed using Neo4j's built-in Cypher query language, which includes the following sub-steps:

[0090] Step S41: Based on the Neo4j graph database, complete the storage and interconnection of wind power operation and maintenance ternary information to realize knowledge graph visualization;

[0091] The visualization solution uses Neo4j to store wind power operation and maintenance data with multiple interconnected relationships. It can achieve efficient and localized storage of massive graph data information and has excellent scalability. It can ensure fast traversal of any graph algorithm and avoid excessive resource consumption caused by using global indexes.

[0092] Meanwhile, the storage strategy for entity naming and relationships in the wind power operation and maintenance field in Neo4j is designed. It adopts a labeled attribute graph model, stores knowledge entities as nodes, and uses node labels to represent the entity class to which they belong. Due to the non-unique nature of labels, multiple labels can be assigned at the same time to represent cross-type knowledge entities, thereby realizing the multi-dimensional interconnection of information in the wind power operation and maintenance field.

[0093] Step S42: Utilize Neo4j's built-in Cypher query language to form a knowledge graph-based intelligent information query and operation and maintenance assistance efficient decision-making application solution;

[0094] The intelligent information query solution allows maintenance personnel to input questions through a human-computer interaction interface. The question parsing module maps the input statement to a Cypher query statement in the Neo4j graph database. Based on the graph information retrieval, the answer generation module outputs the answer according to the retrieval results to achieve intelligent query of maintenance information.

[0095] like Figure 6 As shown, the operation and maintenance auxiliary decision-making application embeds a knowledge graph into the intelligent decision-making system. The equipment operation and maintenance requirement graph enables precise location of equipment structural topology information, while the safety management procedure graph guides the safe conduct of operation and maintenance work. This allows for functions such as process determination, logic generation, and solution recommendation in wind farm operation and maintenance management.

[0096] The preferred embodiments of the present invention have been described in detail above. It should be understood that those skilled in the art can make numerous modifications and variations based on the concept of the present invention without creative effort. Therefore, all technical solutions that can be obtained by those skilled in the art based on the concept of the present invention through logical analysis, reasoning, or limited experimentation on the basis of existing technology should be within the scope of protection defined by the claims.

Claims

1. A wind turbine operation and maintenance management method based on knowledge graphs, characterized in that, The method includes the following steps: Based on the entity extraction algorithm for the operation and maintenance management of wind turbine generator sets, entity types are classified according to the similarities and differences of entity concepts and given constraint definitions. An improved TextRank-based rule matching algorithm is used to extract entity words for wind farm operation and maintenance. Based on the entity relationship extraction algorithm for the operation and maintenance management of wind turbine generators, this algorithm extracts entity relationships by defining various relationships between entities according to the domain text, and specifically includes the following steps: Based on domain text features, various relationships between entities are defined. A rule-based entity relationship extraction algorithm is adopted, and extraction rules are designed to extract relationships and realize the linking of relationships between different entities. The rule-based entity relationship extraction algorithm is specifically as follows: The extracted entity nouns are classified to form entity dictionaries of different categories, which, together with the text to be identified, serve as input for the matching template. Check if the number of entity nouns in the input text is greater than 2. If not, the input text is irrelevant. Define the relationships between various entities, perform rule template matching, determine the corresponding relationships between entities based on their categories, and output the entity relationships. To address the application needs of wind turbine generator operation and maintenance management, a domain knowledge graph is constructed, including a safety management procedure graph and an equipment operation and maintenance requirement graph based on the Protégé wind power equipment ontology instance topology relation library. The construction of the security management procedure diagram is as follows: To address the requirement of associated storage in operation and maintenance management, a knowledge graph of security management procedures is constructed by combining the aforementioned entity extraction and relationship extraction algorithms. The work items are used as the central nodes of the diagram, and the requirements for carrying out different types of work items are used as branches to express the relationship between various safety management procedures in the standard text. The construction steps are as follows: a domain dictionary is formed through text preprocessing, which further assists in the extraction of entity vocabulary; entity relations are extracted using rule matching methods, the relationship links between entities are completed, triple information is formed, and the knowledge graph is constructed. The equipment operation and maintenance requirements map is constructed as follows: To address the need for the integration and connection of the operating specifications of various equipment in a wind farm and the subordinate information of its components, a knowledge graph of wind farm equipment operation and maintenance requirements is constructed by combining the aforementioned entity extraction and relationship extraction algorithms. By organizing and summarizing the components and equipment of wind farms, and storing them using Protégé, a graphical representation of the equipment's structural composition is created. Combined with the corresponding safety regulations, a graphical representation of the operation, maintenance, and repair requirements for wind farm equipment is constructed. The construction steps are as follows: First, combine prior knowledge with Protégé to store instances of wind turbine generator equipment and components to form a wind farm equipment topology information database. Second, perform knowledge fusion with the extracted equipment and component entities to match the same entity nodes and establish the subordinate relationships between the equipment and component entities using the topology information database. Third, store the work items and operating procedures associated with each node through a safety management procedure graph. Fourth, form a wind farm equipment operation and maintenance requirement graph with equipment and components as the central nodes and subordinate relationships as branches. Visualize the knowledge graph for wind turbine operation and maintenance management, and use query language to generate operation and maintenance management solutions.

2. The wind turbine operation and maintenance management method based on knowledge graphs according to claim 1, characterized in that, The entity extraction specifically includes the following steps: Obtain a standard text set for wind farm operation and maintenance management, preprocess the standard text set, use a word segmentation tool to segment and tag each sentence with part-of-speech tags and remove stop words, and introduce an external thesaurus in the wind power field as a custom dictionary into the word segmentation tool; Entity categories are defined by classifying entity types based on domain text features and providing constraint definitions. This study aims to obtain entity vocabulary in the field of wind power operation and maintenance, propose corresponding extraction schemes for different entity categories, and complete entity vocabulary extraction based on the improved TextRank rule matching algorithm.

3. The wind turbine operation and maintenance management method based on knowledge graphs according to claim 2, characterized in that, The entity type classification is as follows: In general domains, common entity description information is used, including natural language definitions of place names, personal names, and organization names; For the wind power operation and maintenance field, targeted entity description information is used, and entity categories are defined as: work items, specification objects, status phenomena, and safety requirements.

4. The wind turbine operation and maintenance management method based on knowledge graphs according to claim 2, characterized in that, The rule matching algorithm based on the improved TextRank is as follows: The improved TextRank algorithm was used to extract keywords from the standardized text, and a rule matching algorithm was combined to filter and extract entities of work projects and standardized object categories. The rule matching algorithm designs extraction rules based on the structural features of different categories of entity concepts in the text. The extracted rules are as follows: Entity = [Strstart] + … + [Strend] In the formula, [Strstart] represents the start symbol; [Strend] represents the end symbol; the start symbol and the end symbol define the entity boundary, and the content in between is the entity concept. When the start symbol does not have a corresponding end symbol, the period or comma punctuation mark is used as the natural end symbol. The improved TextRank algorithm is as follows: A domain dictionary is constructed as the core vocabulary, and the TextRank network graph model is initialized. The core vocabulary is introduced as an external variable. The TextRank values ​​are iteratively calculated and sorted, and the key words of the domain text are extracted according to the order.

5. The wind turbine operation and maintenance management method based on knowledge graphs according to claim 1, characterized in that, The visualization of the knowledge graph for wind turbine operation and maintenance management, and the generation of operation and maintenance management solutions using query languages, includes the following steps: Based on the Neo4j graph database, the storage and interconnection of wind power operation and maintenance ternary information is completed to realize knowledge graph visualization; This paper utilizes Neo4j's built-in Cypher query language to develop a knowledge graph-based intelligent information query and operation and maintenance support solution for efficient decision-making.

6. The wind turbine operation and maintenance management method based on knowledge graphs according to claim 5, characterized in that, The knowledge graph visualization is specifically as follows: Neo4j is used to store wind power operation and maintenance data with multiple interconnected relationships; By adopting a labeled attribute graph model, knowledge entities are stored as nodes, and node labels represent the entity class to which they belong. By assigning multiple labels at the same time to represent knowledge entities of different types, the multi-dimensional interconnection of information in the field of wind power operation and maintenance can be realized.

7. The wind turbine operation and maintenance management method based on knowledge graphs according to claim 5, characterized in that, The intelligent information query and operation and maintenance assistance efficient decision-making application solution is as follows: By inputting questions through the human-computer interaction interface, the question parsing module maps the input statement into a Cypher query statement of the Neo4j graph database. Based on the graph information retrieval, the answer generation module outputs the answer according to the retrieval results, realizing intelligent query of operation and maintenance information. The equipment operation and maintenance requirements map enables precise location of equipment structure and topology information, while the safety management procedure map guides the safe conduct of operation and maintenance work, enabling process judgment, logic generation, and solution recommendation functions for wind farm operation and maintenance management.

Images