Self-adaptive power grid channel selection data base construction method, device, equipment and medium

Abstract

The invention provides a self-adaptive power grid channel selection data base construction method and device, equipment and a medium, and belongs to the field of power system data management. Comprising the following steps: extracting and classifying power grid channel selection data ontology concepts; constructing a knowledge graph; and constructing a dynamic expert knowledge base and designing a data conversion reasoning mechanism. According to the expert knowledge base with the administrative division hierarchical relation as the core, the standard change can be automatically detected, and the structure and content of the expert knowledge base can be automatically updated according to the change. According to the multi-level reasoning mechanism, the corresponding reasoning rules and strategies can be applied according to the administrative level to which the data belongs, and the specific business data of the provincial level, the municipal level and the county level can be subjected to reasoning analysis. The automatic updating capability of the expert knowledge base ensures that the specific business data is consistent with the latest standard during application, and lagging and errors of manual updating are avoided.

Description

Technical Field

[0001] The present invention relates to the field of power system data management, and particularly to a method, device, equipment, and medium for constructing an adaptive power grid channel selection data base. More specifically, in view of the pain points of frequent and inconsistent changes in standards at different administrative levels (province, city, county), the present invention provides a technical solution combining a dynamic knowledge graph and multi-level reasoning, which can realize the automatic integration and update of power grid channel selection-related data during the process from provincial macro planning to city and county-level detailed implementation, and provide sustainable data support for the digital application of site selection and line selection. Background Art

[0002] With the continuous development of the power system, the complexity and diversity of power grid channel selection data are increasing day by day. Especially in the practice of site selection and line selection, since the provincial-level planning often needs to be further refined and implemented at the city and county levels, and there may be data standards that do not match or are lagging in update at the city and county levels, the problem of inconsistency in the data management and application process becomes more prominent. This dynamic change and non-uniformity of standards not only hinder data sharing, analysis, and utilization among multiple departments, but also directly lead to the difficulty of sustainable application of digital products for site selection and line selection, seriously affecting the efficiency and reliability of power grid channel laying.

[0003] Currently, the industry usually adopts the following several solutions to address the above problems: In order to standardize data management, power enterprises attempt to formulate unified data standards and require all departments and levels to follow the same description specifications. However, many difficulties have been encountered in the actual implementation process. Different departments and regions have their own business requirements and management habits, and the implementation of unified standards faces resistance. In addition, unified standards often lack flexibility and are difficult to adapt to local special needs and frequent updates of standards, resulting in poor effects in actual applications. Some enterprises have introduced data management platforms with certain flexibility that support multiple data formats and standards, attempting to be compatible with different description specifications. These systems have alleviated the problem of non-uniform data standards to a certain extent. However, since manual intervention is still required to handle standard changes, the degree of automation is limited. At the same time, the system functions are complex, with high requirements for the technical level of operators, increasing the training and maintenance costs.

[0004] There are also some solutions that use ontology models to describe the concepts and relationships of power grid channel selection data to achieve data integration and sharing. However, traditional ontology models are usually static and cannot adapt to frequent changes in standards, resulting in lagging model updates and affecting the accuracy of data processing. In addition, existing ontology reasoning methods usually perform unified processing on global data and lack refined reasoning for data at different administrative levels, resulting in insufficient accuracy of classification and grading.

[0005] The existing methods have the following problems:

[0006] (1) Insufficient consideration of the impact of standard dynamic changes on the classification and grading of power grid channel selection data: Existing technical methods usually adopt static data models to manage power grid channel selection data. However, with the rapid development of the power industry, the description standards for data by different competent departments and at different administrative levels (such as provinces, cities, and counties) may change frequently, including the introduction of new standards, the revision or repeal of old standards, etc. Existing technologies lack a mechanism to automatically detect standard changes and update the knowledge graph, resulting in the model being unable to reflect the latest standards in a timely manner, thereby affecting the accuracy and consistency of data classification and grading. Specifically, when the standard changes, existing systems often require manual intervention. This manual update method is inefficient and prone to introducing human errors, leading to inconsistencies and error accumulation in the data processing process. In addition, the static data model cannot adapt to the dynamic changes of standards and cannot meet the requirements of real-time and accuracy.

[0007] (2) Insufficient consideration of the impact of differences in data at different administrative levels on classification and grading: Existing technologies usually uniformly process power grid channel selection data and lack a refined reasoning mechanism for data at different administrative levels such as provincial, municipal, and county levels. There are significant differences in standards, specifications, and business requirements for data at different administrative levels. For example, provincial standards may focus on macro policies and overall planning, municipal standards may pay attention to regional characteristics and resource allocation, and county-level standards may emphasize specific implementation and detail optimization. Existing methods usually target only a specific unit, making it difficult to share data among different business departments and unable to meet the business needs of each level of department. Summary of the Invention

[0008] The purpose of the present invention is to overcome the above shortcomings of the existing technology and provide a method, device, equipment, and medium for constructing an adaptive power grid channel selection data base, which can dynamically adapt to standard changes and fully consider the differences in data at different administrative levels. The specific objectives are as follows: (1) Design a knowledge graph data base with the hierarchical relationship of administrative divisions as the core to store various spatio-temporal data of power grid channel selection. (2) Develop a multi-level reasoning mechanism that can, according to the administrative level to which the data belongs, apply corresponding reasoning rules and strategies to respectively conduct reasoning and analysis on provincial, municipal, and county-level data, and automatically update the structure and content of the graph.

[0009] The technical solution of the present invention is specifically as follows: A method for constructing an adaptive power grid channel selection data base includes the following steps:

[0010] Step 1: Extraction and classification of the ontology concepts of power grid channel selection data, including the following steps: Collect power grid channel standards and specifications, extract the principles of power grid channel selection, extract and classify ontology concepts, classify the ontology concepts according to administrative levels, and establish the attributes and relationships of ontology concepts;

[0011] Step 2, Knowledge Graph Construction;

[0012] Step 2.1, Knowledge Extraction;

[0013] Step 2.2, Construction and Hierarchical Organization of the Knowledge Graph;

[0014] Step 3, Construction of a Dynamic Expert Knowledge Base and Design of a Data Conversion Inference Mechanism; construct an automatically updated expert knowledge base, and provide a multi-level inference engine based on the expert knowledge base, and this inference engine realizes: level recognition, rule matching, condition judgment, and inference execution.

[0015] The present invention provides an adaptive power grid channel selection data base construction device, including the following modules:

[0016] Concept Extraction and Classification Module, used for the extraction and classification of power grid channel selection data ontology concepts, including: collecting power grid channel standards and specifications, extracting the principles of power grid channel selection, extracting and classifying ontology concepts, classifying ontology concepts according to administrative levels, and establishing ontology concept attributes and relationships;

[0017] Knowledge Graph Construction Module, used for knowledge graph construction, including: knowledge extraction; construction and hierarchical organization of the knowledge graph;

[0018] Dynamic Expert Knowledge Base Construction and Conversion Inference Module, used to construct an automatically updated expert knowledge base, and provide a multi-level inference engine based on the expert knowledge base, and this inference engine realizes: level recognition, rule matching, condition judgment, and inference execution.

[0019] The present invention also provides an electronic device, including: one or more processors; a memory, used to store one or more programs, wherein, when the one or more programs are executed by the one or more processors, the one or more processors are caused to implement the above method.

[0020] The present invention also provides a computer-readable storage medium, on which executable instructions are stored, and when the instructions are executed by a processor, the processor is caused to implement the above method.

[0021] The present invention has the following beneficial technical effects:

[0022] (1)Semantic reasoning-driven hierarchical knowledge graph construction to promote data sharing and collaboration: Compared with the prior art, the present invention constructs a knowledge graph according to the provincial, municipal, and county levels, and combines a semantic reasoning mechanism to systematically organize and manage standards and rules. This hierarchical knowledge graph structure with semantic understanding capabilities makes the relationship between superior and inferior standards clearer, facilitating the update and maintenance of rules. At the same time, the semantic reasoning ability enables the system to intelligently identify and apply the standards of specific departments or units, complete the reclassification and grading of data, and promote the shared and collaborative utilization of data. Through deep association and reasoning at the semantic level, the system can effectively integrate data at different administrative levels, solve the inconsistencies and conflicts in data sharing, significantly improve the efficiency and reliability of power grid channel laying, and ensure that the business needs of all levels of departments are fully met.

[0023] (2)Dynamically adapt to standard changes to improve the adaptability and real-time performance of the system: The static knowledge graph in the prior art cannot adapt to the frequent changes in the power grid channel selection standards, resulting in lagging model updates and affecting the accuracy of data processing. The present invention can automatically detect standard changes and automatically update the structure and content of the graph according to the changes. The system can respond to the latest standards in real time, maintain consistency with the latest specifications, and can also automatically reason and optimize data classification and grading, avoiding the lag of manual updates and possible errors introduced, significantly enhancing the system's adaptability to standard changes, and improving the efficiency and accuracy of data management.

[0024] (3)Reduce manual intervention to improve system intelligence and efficiency: The present invention significantly reduces the workload of manual data maintenance and the risk of human errors through automated knowledge extraction, knowledge graph construction, and dynamic update mechanisms. At the same time, the introduction of a multi-level expert knowledge reasoning engine makes the data classification and grading process more efficient and accurate. The semantic reasoning ability of the knowledge graph enables the system to perform intelligent analysis and decision-making based on the deep relationships between data, further improving the overall working efficiency and intelligence level of the site selection and line selection digital products. Description of the Drawings

[0025] Figure 1 It is a schematic diagram of concepts involved in power grid channel selection;

[0026] Figure 2 It is a process diagram of knowledge extraction;

[0027] Figure 3 It is a flow chart of dynamic update rules;

[0028] Figure 4 It is a flow chart of the reasoning process. Detailed Implementation Modes

[0029] In order to make the objectives, technical solutions and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other. To achieve the above objectives, the present invention adopts the following technical solutions.

[0030] Specifically, the present invention proposes a method for constructing an adaptive power grid channel selection data base, which specifically includes the following three steps:

[0031] Step 1: Extraction and classification of the ontology concepts of power grid channel selection data;

[0032] The knowledge graph is constructed with ontology concepts as the core. By establishing the attributes, hierarchical relationships and association rules of ontology concepts, a complete graph structure is formed. To construct the knowledge graph, it is first necessary to deeply understand the principles and relevant standards of power grid channel selection, extract the relevant ontology concepts and classify them. This step is the basis of the entire technical solution and directly affects the subsequent construction of the knowledge graph and the effectiveness of the multi-level reasoning mechanism. Step 1 specifically includes:

[0033] Step 1.1: Collect relevant standards and specifications;

[0034] National standards and industry specifications: GB 50545 "Code for Design of 110kV - 750kV Overhead Transmission Lines"; GB50665 "Code for Design of 1000kV Overhead Transmission Lines"; GB 50790 "Code for Design of ±800kV DC Overhead Transmission Lines"; GB 50183 "Code for Fire Protection Design of Petroleum and Natural Gas Engineering"; GB / T 19531 "Technical Requirements for Observation Environment of Seismic Stations"; DL / T 5040 "Code for Design of Protection Against Influence of Transmission Lines on Radio Stations"; DL / T 5536 "Code for Design of Protection Against Influence of DC Overhead Transmission Lines on Radio Stations"; DL / T 5582 "Code for Electrical Design of Overhead Transmission Lines"; HJ 681 "Monitoring Method for Electromagnetic Environment of AC Transmission and Transformation Projects (Trial)"; GB 50089 "Code for Safety Design of Engineering for Civil Explosives".

[0035] Laws and regulations: "Law of the People's Republic of China on the Protection of Oil and Gas Pipelines"; "Safety Code for Blasting" (GB6722); "Code for Design of Urban Gas Engineering" (GB 50028); "Code for External Corrosion Control of Steel Pipelines" (GB / T 21447); "Technical Standard for Gasoline, Diesel, LPG and Hydrogen Refueling Stations" (GB 50156); "Code for Safety Design of Fireworks and Firecracker Engineering" (GB 50161); "Technical Standard for Protection Against AC Interference of Buried Steel Pipelines" (GB / T 50698); "Code for Design of Wind Farms" (GB 51096).

[0036] Step 1.2: Extract the principles for selecting the power grid corridors;

[0037] Based on the collected standards and specifications, combined with the practical experience in the selection and planning of power grid transmission corridors, the principles for corridor selection are summarized. These principles reflect the key factors that need to be considered during the power grid line planning and construction process and are important bases for extracting ontology concepts. The principles include:

[0038] Safe and reliable, economical and reasonable: According to the requirements of power system planning, comprehensively consider factors such as construction, operation, traffic conditions, and line length, optimize the plan, and make the line path safe, reliable, economical, and reasonable.

[0039] Coordinate the relationship with other facilities: Comprehensively coordinate the contradictions between this line and the existing, under-construction, planned, and proposed power transmission lines, highways, railways, and other facilities along the line to avoid interference and conflicts.

[0040] Optimize traffic conditions: Try to be as close as possible to the existing national highways, provincial highways, county roads, and rural roads to improve the traffic conditions of the line and facilitate construction and maintenance.

[0041] Minimize environmental impact: Avoid sensitive areas such as forest areas, natural ecological protection areas, cultural relic protection areas, and world cultural heritages to protect the ecological environment and cultural heritages.

[0042] Consider geological and meteorological conditions:

[0043] Avoid bad geological zones: Such as geological disaster areas like landslides, mudslides, and ground settlement to ensure the safety and reliability of the line.

[0044] Avoid heavily polluted sections: Reduce the crossing of heavily polluted areas to improve the line reliability and reduce construction and maintenance costs.

[0045] Avoid severely ice-covered and large wind gap sections: Improve the ice and wind resistance of the line to ensure safe operation.

[0046] Consider human and social factors: Try to avoid crossing residential houses, reflect the people-oriented design concept, and reduce the impact on residents' lives.

[0047] Avoidance of military and important facilities: Avoid military facilities, large factories and mines, important communication facilities, etc., to avoid interfering with national security and important production activities.

[0048] Compliance with planning requirements: Coordinate with local planning requirements, reasonably select the line path, and avoid affecting the planned area.

[0049] Selection of line type: According to the path corridor situation, through technical and economic comparison, determine the erection method (such as double circuit on the same tower or two single circuits) to achieve the purpose of resource conservation and environmental friendliness.

[0050] Rational design of cross - overs: Reasonably select the location of important cross - over points to ensure the safety distance from other lines, roads, water areas, etc.

[0051] Step 1.3: Extract and classify ontology concepts;

[0052] Based on the above principles, in - depth reading and analysis of the collected standards and specifications were carried out, and ontology concepts closely related to power grid corridor selection were manually extracted. These ontology concepts cover multiple aspects such as geographical environment, natural conditions, and social factors, as Figure 1 shown. Different ontology concepts and their classification details in power grid corridor selection data. Ontology concepts are divided into two main categories: geographical and environmental concepts, and human and social concepts. These concepts are extracted based on in - depth analysis of various influencing factors in the power grid corridor selection process in Step 1.2 and are refined and classified in combination with administrative levels, thus providing a scientific and reasonable basis for power grid corridor selection.

[0053] Among the geographical and environmental concepts, there are multiple sub - categories including nature reserves, scenic spots, forest areas, basic farmlands, geological disaster areas, water source protection areas, heavily polluted areas, icing areas, strong wind areas, etc. These concepts mainly reflect the natural and geographical environmental conditions that need to be considered in the power grid corridor selection process to ensure the safety and feasibility of line laying and reduce environmental impacts.

[0054] Among the human and social concepts, factors such as cultural heritage protection areas, military facilities, large mining enterprises, important communication facilities, residential areas, and planned areas are covered. These concepts emphasize the impact of social activities, cultural resources, and important infrastructure on power grid corridor selection, ensuring that the interference of power grid laying on social life, cultural protection, and economic activities is minimized.

[0055] Through Figure 1The hierarchical relationships are shown, and these ontology concepts are clearly divided into different categories and refined according to the administrative regional levels of provinces, cities, and counties within the applicable scope, so as to be systematically organized and managed in the knowledge graph. This classification method ensures that during the reasoning and data processing processes, the system can automatically apply corresponding reasoning rules according to the data requirements at different levels to make scientific classification and grading decisions. The knowledge graph stores each ontology concept and its attributes (such as definitions, restrictive conditions, applicable scopes, and related regulations) in a structured manner, enabling the system to dynamically adapt to changes in standards, thereby improving the real-time performance and accuracy of power grid data management.

[0056] During the process of extracting ontology concepts, it is noted that the same ontology concept may have different expressions in different documents. For example, "basic farmland" and "permanent basic farmland" may refer to the same area in some cases. To ensure the consistency of ontology concepts, synonyms and near-synonyms are uniformly processed and a common and authoritative name is adopted.

[0057] Step 1.4: Classify according to administrative levels;

[0058] According to the applicable scope and source of the ontology concepts, the extracted ontology concepts are classified into three levels: provinces, cities, and counties:

[0059] Provincial-level concepts: Standards and specifications applicable to the whole province, which have a macro guiding role. For example, "provincial-level nature reserve", "provincial-level scenic spot".

[0060] Municipal-level concepts: Reflect the regulations and requirements of a specific municipal administrative region, embodying regional characteristics. For example, "municipal-level basic farmland protection area", "municipal-level geological disaster prone area".

[0061] County-level concepts: Involve specific requirements and implementation details at the county or district level, emphasizing details and operability. For example, "county-level water source protection area", "township ecological red line area".

[0062] Step 1.5: Establish ontology concept attributes and relationships;

[0063] Add relevant attribute information to each ontology concept, including definitions, restrictive conditions, applicable scopes, related regulations, etc. For example:

[0064] Nature reserve:

[0065] Definition: An area legally delimited to protect representative natural ecosystems, rare and endangered wild animal and plant species, and natural relics, etc.

[0066] Restrictive conditions: It is prohibited to newly build power grid corridors, and avoidance measures need to be taken.

[0067] Relevant regulations: "Regulations of the People's Republic of China on Nature Reserves".

[0068] Scope of application: Provincial, municipal, and county administrative regions.

[0069] Prime farmland:

[0070] Definition: Cultivated land demarcated in accordance with laws and regulations that shall not be occupied or converted for other uses.

[0071] Restriction conditions: Grid construction requires special approval and should avoid or take compensation measures as much as possible.

[0072] Relevant regulations: "Land Administration Law of the People's Republic of China", "Regulations on the Protection of Prime Farmland".

[0073] Scope of application: Provincial, municipal, and county administrative regions.

[0074] Through the above steps, the extraction and classification of the data ontology concepts for grid corridor selection are completed, unifying the concepts and terms, and providing accurate and detailed knowledge support for subsequent data processing and reasoning.

[0075] Step 2. Knowledge graph construction;

[0076] After completing the extraction and classification of the data ontology concepts for grid corridor selection, it is necessary to systematically organize these concepts and their relationships to construct a knowledge graph that stores specific classification criteria. This knowledge graph is based on the administrative divisions at the provincial, municipal, and county levels, and fully considers the different classification criteria for the same data category by different administrative levels and units.

[0077] After completing the extraction and classification of the data ontology concepts for grid corridor selection, it is necessary to further extract the knowledge related to the specified data categories from the collected standards, specifications, and policy documents, and construct a dynamic knowledge graph. Since the direct use of the template matching method is not suitable, a knowledge extraction method based on domain dictionaries and rules is adopted, combined with manual assistance, to more accurately extract the required information.

[0078] Step 2.1 Knowledge extraction;

[0079] Since the data for grid corridor selection has spatio-temporal attributes, the present invention standardizes and structures these data in GeoJSON format for easy storage and management in the knowledge graph.

[0080] Data collection: Collect various original business data, including DEM, road network, etc. The original business data is usually stored in tif or shapefile format.

[0081] Format conversion: Convert the collected spatio-temporal business data into a format that conforms to the GeoJSON specification. The Label of GeoJSON is the attribute of the original business data. GeoJSON is an open standard format based on JSON, used to represent simple geographical features, non-geographical attributes, and their interrelationships.

[0082] Data verification: Ensure that the converted GeoJSON data conforms to the specifications, including the correct coordinate system, valid geometric types (such as Point, LineString, Polygon), etc. Among them, Point represents a point; LineString represents a line; Polygon represents a closed polygon area.

[0083] Step 2.2: Construct a knowledge graph and form a hierarchical organization;

[0084] Import the GeoJSON files obtained from the above knowledge extraction into the knowledge graph to form a multi-level and extensible knowledge network. The construction process of the knowledge graph includes the following steps:

[0085] Since different administrative levels and units have different classification criteria for the same data category, refined processing is carried out in the knowledge graph:

[0086] 1. Under each ontology concept node, add the specific classification criteria of the corresponding administrative level and unit as attributes.

[0087] 2. Establish the hierarchical relationship between ontology concepts, and clarify the inheritance and extension relationships between superior and subordinate concepts.

[0088] To effectively store and manage the knowledge graph, GraphDB is used for storage. The graph database can efficiently store the data structures of nodes and edges, support complex relationship queries and fast retrievals.

[0089] Step 3: Construct a dynamic expert knowledge base and design a data conversion inference mechanism;

[0090] Construct an automatically updated expert knowledge base, and based on the expert knowledge base, provide a multi-level inference engine that realizes: level recognition, rule matching, condition judgment, and inference execution;

[0091] 3.1 Construction of the expert knowledge base

[0092] First, based on the established ontology concepts, conduct in-depth extraction and construction of the expert knowledge base from the collected standards, specifications, and policy documents. The purpose of extraction is to convert the knowledge elements in the text into a structured data form, namely triples (subject, predicate, object). Figure 2Shows the knowledge extraction process of the expert knowledge base. The purpose of knowledge extraction is to convert the knowledge elements in the collected standards, specifications, and policy documents into a structured data form, namely triples (subject, predicate, object). Specifically, it includes the following steps: Adopt a knowledge extraction method based on domain dictionaries and rules, combined with manual assistance, to more accurately extract the required information. The knowledge extraction process is as Figure 2 shown. It includes the following steps:

[0093] Establish a domain dictionary: Organize the previously extracted ontology concepts and related terms into a domain dictionary to support term recognition and concept matching in the knowledge extraction process.

[0094] Design extraction rules: For each data category, design corresponding rules to extract the relevant knowledge. The data category refers to different types of power grid channel selection data defined in the knowledge graph, such as geographical environment factors, human social factors, etc. The extraction rules are matched through the domain dictionary to ensure that relevant information for each category of data can be accurately extracted.

[0095] Key sentence identification: By identifying the key verbs and noun phrases in the text, find the sentences describing regulations and restrictions. These key phrases are usually associated with specific ontology concepts and help identify the content that has an important impact on data classification and standards.

[0096] Triple extraction: Through template matching with the designed extraction rules, obtain the corresponding triples (subject, predicate, object).

[0097] Subject: Usually a specific concept or entity in power grid channel selection, such as "nature reserve".

[0098] Predicate: Describes the relationship between the subject and the object, such as "should avoid".

[0099] Object: The object affected by the subject, such as "power grid line". These triples are used to construct the relationships in the expert knowledge base, enabling knowledge to be stored and reasoned in a structured form.

[0100] Manual review: Due to the complexity of regulatory texts, automatic extraction may have omissions or errors, so manual verification and supplementation are required. Through manual assistance, ensure that the extracted triples are accurate, thereby improving the quality of the knowledge graph and the reliability of reasoning.

[0101] Step 3.2, Design of dynamic update rules for the expert knowledge base

[0102] To ensure that the expert knowledge base is always consistent with the latest standards and specifications, dynamic update rules for the expert knowledge base are designed, as Figure 3as shown Figure 3 shows the dynamic update process of the expert knowledge base. This process is divided into three main steps, namely standard change monitoring, knowledge update process, and version control.

[0103] 1. Standard change monitoring: In the construction of the expert knowledge base, the system first monitors the changes in relevant standards and specifications in real time. These standards include national, industry, and local regulations, ensuring that the system can quickly capture the latest policy changes and technical requirements. Through natural language processing (NLP) technology, the system automatically parses the new text, extracts the changed parts, and marks them as content to be updated.

[0104] 2. Knowledge update process: After the standard changes are identified, the system updates the relevant nodes in the knowledge graph according to the extracted changed content. Each ontology concept node in the expert knowledge base is associated with specific attributes and rules, such as applicable administrative levels and unit classification standards. During the update process, the system dynamically adjusts the structure of the expert knowledge base based on these changes, adding new concepts or modifying existing concepts to ensure that the expert knowledge base always remains up-to-date.

[0105] 3. Version control: To ensure the traceability of knowledge updates, the system performs version control on each update of the expert knowledge base. The version control module records the time of each update, the changed content, and the review records of relevant personnel, so that in case of errors or disputes, it can quickly trace back to the previous version. This mechanism effectively reduces potential errors in manual updates and provides a reliable historical basis for subsequent reviews and optimizations.

[0106] Through Figure 3 the process shown, it can be clearly seen the entire process of the expert knowledge base from monitoring standard changes to updates and version management. The relationships between the concepts in the figure are progressive and interdependent. Standard change monitoring is the starting point for triggering updates, while the knowledge update process is the structural adjustment of the expert knowledge base. Finally, version control ensures the robustness of the system and the orderliness of updates.

[0107] Through the construction of the expert knowledge base and the design of dynamic update rules, the corresponding specifications and standards are successfully organized systematically and structurally. The expert knowledge base is based on the administrative divisions at the provincial, municipal, and county levels, fully considering different classification standards for the same data category by different administrative levels and units, providing a solid foundation for the subsequent multi-level reasoning mechanism. The introduction of the dynamic update mechanism ensures that the expert knowledge base can reflect the latest standard and specification changes in real time, significantly improving the efficiency and accuracy of data management.

[0108] For the constructed expert knowledge base, different standards of different departments and enterprises at each administrative level have been stored. When actually used, only the standards of the department or unit that wants to share data need to be obtained from the expert knowledge base, and the data can be reclassified according to the actual attributes of the data.

[0109] After constructing an expert knowledge base based on the administrative division levels of provinces, cities, and counties, a multi-level inference mechanism based on the expert knowledge base has been developed to make full use of the advantages of the knowledge graph and achieve precise classification and grading of power grid channel selection data. This mechanism can automatically apply corresponding inference rules according to the actual attributes of the data and the administrative region to which it belongs, and perform intelligent processing on the data.

[0110] First, extract the standards, specifications of administrative regions and departments at all levels, and relevant inference rules from the expert knowledge base. These rules describe in detail how data should be classified or processed under specific conditions. For example:

[0111] These rules are systematically sorted out to form a multi-level inference rule base, providing a solid foundation for the inference mechanism.

[0112] Next, a multi-level inference engine is designed and implemented. The multi-level inference engine is an inference execution module based on the knowledge graph and inference rules. Inference rules refer to the logical rules that describe the conditions for data classification and grading. The inference engine realizes intelligent inference by calling these rules. This inference engine has the following functions:

[0113] Level identification: Automatically determine the provincial, municipal, and county levels to which the data belongs according to the geographical location of the data.

[0114] Rule matching: Select rules related to the level and attributes of the data from the inference rule base.

[0115] Condition judgment: Judge which rules are applicable according to the actual attributes of the data (such as geographical coordinates, characteristic attribute values, etc.).

[0116] Inference execution: Apply the applicable rules to classify and grade the data.

[0117] This inference engine specifically includes the following modules:

[0118] Data preprocessing module: Function: Preprocess the input data, extract necessary attribute information, such as geographical location, thematic attribute values, etc. Implementation: Use geographic information system (GIS) technology to map the geographical coordinates of the data to administrative regions and specific locations.

[0119] Level identification module: Function: Determine the provincial, municipal, and county levels to which the data belongs according to the geographical location of the data. Implementation: Use administrative division boundary data for spatial overlay analysis.

[0120] Rule matching module: Function: Select inference rules related to the level and attributes of the data from the rule base. Implementation: Load the corresponding rule set according to the level recognition result.

[0121] Condition judgment module: Function: Judge which rules are applicable according to the actual attributes of the data. Implementation: Traverse the rule set to judge whether the condition part of the rule is satisfied.

[0122] Inference execution module: Function: Apply the applicable rules to classify and grade the data and generate inference results. Implementation: Execute the action part of the rule to update the classification mark and related information of the data.

[0123] Result output module: Function: Output the inference results, generate reports or update the database. Implementation: Output the inference results in a structured form to support subsequent applications.

[0124] In practical applications, when planning a specific power grid channel, the system will automatically obtain the coordinate range of the data in the knowledge graph and obtain relevant standards and inference rules from the expert knowledge base to classify and grade the data. The specific inference process is as Figure 4 shown. Figure 4 It shows the specific process of data classification and grading inference for power grid channel selection based on the knowledge graph. This inference process includes six main steps, namely data input, level recognition, rule matching, condition judgment, inference execution and result output.

[0125] Data input: Obtain all data within the administrative division range involved by the starting points of the power grid channel (as the lower left corner and the upper right corner respectively). The data includes geographical locations and other relevant attributes. The system extracts necessary attribute information through the data preprocessing module, such as geographical coordinates and thematic attribute values.

[0126] Level recognition: The system determines the provincial, municipal and county levels to which the data belongs according to the geographical location of the data. By using the administrative division boundary data for spatial overlay analysis, the system can accurately identify the administrative level of the data, so as to apply appropriate rules in subsequent processing.

[0127] Rule matching: The system selects inference rules related to the level and attributes of the data from the expert knowledge base and the rule base. According to the result of level recognition, load the corresponding rule set to ensure the most appropriate classification and grading standards are used in the inference process.

[0128] Condition judgment: The system judges which rules are applicable according to the attributes of the data. By traversing the rule set, the system can judge whether the condition part of each rule is satisfied, so as to select the applicable rules.

[0129] Inference execution: For applicable rules, the system executes the action part of the rules to classify and grade the data. The inference execution module is responsible for updating the classification tags and related information of the data to ensure the accuracy and consistency of the inference results.

[0130] Result output: After the inference is completed, the system outputs the classified and graded results. The results can be presented in the form of a report or updated to the database for subsequent use and query. The system ensures that the inference results are output in a structured form and used as the input of the route selection algorithm to directly assist in the planning of transmission lines.

[0131] Through Figure 4 The shown inference process enables an intuitive understanding of the complete process of power grid channel selection data from input to final output. There is a clear logical relationship between each step. Data input triggers level identification, and rule matching and condition judgment ensure the accuracy of classification and grading. The inference execution and result output modules guarantee the efficiency and reliability of the inference process. By developing a data conversion inference mechanism, the intelligent classification and grading of power grid channel selection data are achieved. This mechanism makes full use of the data in the knowledge graph and the knowledge in the expert knowledge base, combines multi-level inference rules, can automatically adapt to changes in standards, and accurately processes data of different administrative levels and different business departments. The introduction of the inference mechanism improves the efficiency of data management and application.

Claims

1. A method for constructing an adaptive power grid channel selection data base, characterized in that: The following steps are involved: Step 1, extraction and classification of the grid channel selection data ontology concepts, including the following steps: collecting grid channel standards and specifications, extracting the principles of grid channel selection, extracting and classifying ontology concepts, classifying ontology concepts according to administrative levels, and establishing ontology concept attributes and relationships; Step 2: Knowledge graph construction, including: Step 2.1, knowledge extraction; Step 2.2: Build a knowledge graph and form a hierarchical organization. Step 3: Dynamic expert knowledge base construction and data conversion reasoning mechanism design; construct an automatically updated expert knowledge base, and provide a multi-level reasoning engine based on the expert knowledge base, which implements: level identification, rule matching, condition judgment, and reasoning execution.

2. The method for constructing an adaptive power grid channel selection data base according to claim 1, characterized in that: Step 1 is as follows: Step 1.1, collecting grid channel standards and specifications; the grid channel standards and specifications include national standards, industry specifications and laws and regulations; Step 1.2: Based on the above grid channel standards and specifications, extract the principles for grid channel selection; the principles reflect the factors considered in the planning and construction of grid lines; Step 1.3: Extract and classify ontology concepts based on the factors in step 1.2; ontology concepts cover geographical environment, natural conditions, and social factors; Step 1.4: Classify the ontology concepts according to administrative levels; according to the applicable scope and source of the ontology concepts, classify the extracted ontology concepts into three levels: province, city, and county; Step 1.5, establish the ontology concept attributes and relationships; add relevant attribute information for each ontology concept, including definition, restrictions, scope of application, and related regulations.

3. The method for constructing an adaptive power grid channel selection data base according to claim 2, characterized in that: Step 1.2 specifically includes: Based on the collected grid channel standards and specifications, combined with the practical experience of grid transmission channel selection and planning, the principles of channel selection are summarized, including: According to the power system planning requirements, the plan is optimized by comprehensively considering the construction, operation, traffic conditions and line length factors; Comprehensively coordinate the contradictions between the lines and the transmission lines, roads, railways and other facilities that have been built, are under construction, are planned to be built and are planned along the lines to avoid interference and conflicts; Close to existing national, provincial, county and rural roads; Avoid forest areas, natural ecological environment protection areas, cultural relics protection areas, and world cultural heritage areas to protect the ecological environment and cultural heritage; Avoid adverse geological zones, including areas of landslide, debris flow, and ground subsidence geological disasters; Avoid polluted sections and reduce crossing of heavily polluted areas; Avoid ice-covered and windy areas; Avoid crossing over residential buildings to reduce the impact on residents’ lives; Avoid military facilities, large factories and mines, and important communication facilities; Coordinate with local planning requirements and reasonably select line routes to avoid impact on the planned area; Line mode selection: Determine the installation mode according to the path corridor conditions, including double circuits on the same tower or two single circuits; Reasonably select the location of the crossing point to ensure a safe distance from other lines, roads and waters.

4. The method for constructing an adaptive power grid channel selection data base according to claim 3, characterized in that: Step 1.3 specifically includes: Ontological concepts are divided into two categories: geographical and environmental concepts, and humanistic and social concepts; Geographical and environmental concepts include multiple subcategories such as nature reserves, scenic spots, forest areas, basic farmland, geological disaster areas, water source protection areas, heavily polluted areas, ice-covered areas, and windy areas; The humanities and society concepts include cultural heritage protection areas, military facilities, mining enterprises, communication facilities, residential areas and planning areas.

5. The method for constructing an adaptive power grid channel selection data base according to claim 4, characterized in that: Step 1.4 specifically includes: According to the scope of application and source of the concepts, the extracted ontology concepts are classified according to the three administrative levels of province, city and county: Provincial concept: standards and specifications applicable to provincial administrative regions; City-level concept: reflects the regulations and requirements of the city-level administrative region and embodies regional characteristics; County-level concept: requirements and implementation details related to county-level administrative areas.

6. The method for constructing an adaptive power grid channel selection data base according to claim 5, characterized in that: Step 1.5 is as follows: Add relevant attribute information to each ontology concept, including definition, restrictions, scope of application, and associated regulations.

7. The method for constructing an adaptive power grid channel selection data base according to claim 6, characterized in that: Step 2.1 includes: The data is standardized and structured using GeoJSON format; including: Data collection: collect various original business data, including DEM and road network. The original business data is stored in tif or shapefile format; Format conversion: convert the collected original spatiotemporal business data into a format that complies with the GeoJSON specification. The Label of GeoJSON is the attribute of the original business data. Data validation: Ensure that the converted GeoJSON data complies with the specifications, including the correct coordinate system and valid geometry type.

8. The method for constructing an adaptive power grid channel selection data base according to claim 7, characterized in that: Step 2.2 includes: Under each ontology concept node, add the specific classification standards of the corresponding administrative level and unit as attributes; Establish a hierarchical relationship between ontology concepts, clarify the inheritance and extension relationship between superior and subordinate concepts; and use a graph database to store knowledge graphs.

9. The method for constructing an adaptive power grid channel selection data base according to claim 8, characterized in that: Step 3 includes: Step 3.1: Constructing the expert knowledge base; including: Based on the established ontology concepts, in-depth knowledge extraction is performed on the collected standards, specifications and policy documents; knowledge extraction is used to convert the knowledge elements in the text into a structured data form, namely a triple, including subject, predicate and object; The knowledge extraction method based on domain dictionary and rules is adopted, which includes the following steps: Establish a domain dictionary: organize the previously extracted ontology concepts and ontology-related terms into a domain dictionary; Design extraction rules: For each data category, design corresponding rules to extract knowledge related to the data category; Key sentence identification: Find sentences describing regulations and restrictions by identifying key verbs and noun phrases; Triple extraction: The corresponding triples are obtained by performing template matching through the designed extraction rules; Manual review: manual verification and supplement; Step 3.2: Design of dynamic update rules for expert knowledge base, including: standard change monitoring, knowledge update process and version control; Standard change monitoring includes: first, in the construction of the expert knowledge base, real-time monitoring of changes in power grid channel standards and specifications; these standards include national, industry and local regulations, and new texts are automatically parsed through natural language processing, and the changed parts are extracted and marked as content to be updated; The knowledge update process includes: after the standard changes are identified, the relevant nodes in the knowledge graph are updated according to the extracted changes. Each ontology concept node in the expert knowledge base is associated with specific attributes and rules; during the update process, the structure of the expert knowledge base is dynamically adjusted according to these changes, new concepts are added or existing concepts are revised to ensure that the expert index library is always up to date; Version control includes: In order to ensure the traceability of knowledge updates, version control is performed on each update of the expert knowledge base. The version control module records the time, change content and audit records of each update.

10. The method for constructing an adaptive power grid channel selection data base according to claim 9, characterized in that: Step 3 also includes: providing a multi-level reasoning engine based on an expert knowledge base, the reasoning engine having the following functions: Level identification: Automatically determine the province, city, or county level to which the data belongs based on its geographic location; Rule matching: Select rules related to the level and attributes of the data from the inference rule library; Conditional judgment: Determine applicable rules based on the actual attributes of the data, including geographic coordinates and feature attribute values; Inference execution: Apply applicable rules to classify and grade data.

11. A method for constructing an adaptive power grid channel selection data base according to claim 10, characterized in that: The inference engine specifically includes the following modules: Data preprocessing module: used to preprocess the input data, extract attribute information including geographic location and thematic attribute values; map the geographic coordinates of the data to administrative regions and specific locations; Level identification module: used to determine the province, city, or county level to which the data belongs based on its geographical location; and to conduct spatial overlay analysis using administrative division boundary data; Rule matching module: used to select inference rules related to the level and attributes of the data from the rule base; load the corresponding rule set according to the level identification result; Conditional judgment module: used to judge the applicable rules according to the actual attributes of the data; traverse the rule set to judge whether the condition part of the rule is met; Reasoning execution module: used to apply applicable rules, classify and grade data, and generate reasoning results; execute the action part of the rules, and update the classification tags and related information of the data; Result output module: used to output reasoning results, generate reports or update databases; output reasoning results in a structured form.

12. An adaptive power grid channel selection data base construction device, characterized in that: Includes the following modules: The concept extraction and classification module is used to extract and classify the concepts of the grid channel selection data ontology, including: collecting grid channel standards and specifications, extracting the principles of grid channel selection, extracting and classifying ontology concepts, classifying ontology concepts according to administrative levels, and establishing ontology concept attributes and relationships; The knowledge graph construction module is used for knowledge graph construction, including: knowledge extraction; knowledge graph construction and hierarchical organization; The dynamic expert knowledge base construction and conversion reasoning module is used to build an expert knowledge base that can be automatically updated, and provide a multi-level reasoning engine based on the expert knowledge base. The reasoning engine implements: level identification, rule matching, condition judgment, and reasoning execution.

13. An electronic device, characterized in that: include: one or more processors; A memory for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the method for constructing an adaptive power grid channel selection data base as described in any one of claims 1 to 11.

14. A computer-readable storage medium, characterized in that: Executable instructions are stored thereon, and when the instructions are executed by a processor, the processor implements a method for constructing an adaptive power grid channel selection data base as described in any one of claims 1 to 11.

Images