Power grid channel selection method, device, equipment, and medium based on knowledge graph

Abstract

The present invention proposes a method, device, equipment, and medium for selecting power grid channels based on knowledge graphs, which belongs to the field of knowledge graph artificial intelligence applications in the direction of smart grids. It includes: building a knowledge graph framework; integrating multi-source heterogeneous data required for power grid channel planning, building logical associations and semantic models, and providing data and knowledge support for planning; constructing a power grid channel planning model; combining knowledge graph technology with actual planning needs, building a planning model supported by knowledge graphs, and realizing intelligent planning of power grid channels. By integrating multi-source heterogeneous data obtained by multiple means such as geospatial information, remote sensing, and expert research, combined with a land suitability evaluation algorithm supported by knowledge graphs, automated evaluation of power grid channels in high-altitude areas is realized. This method not only improves the efficiency and accuracy of planning and design, but also effectively reduces the intensity and cost of manual work in high-altitude areas, providing an innovative solution for power grid channel planning.

Description

Technical Field

[0001] The present invention belongs to the field of knowledge graph artificial intelligence application technology in the direction of smart grid, and in particular, relates to a method, device, equipment, and medium for selecting a power grid channel based on a knowledge graph. Background Art

[0002] High-altitude areas are rich in hydropower resources and have great development potential. However, the geological conditions in these areas are extremely complex, the terrain is diverse, and there are a large number of environmentally sensitive areas, such as nature reserves, wetland parks, forest parks, and scenic spots. At the same time, problems such as heavy ice areas, strong wind areas, transportation road networks, and housing distribution have further increased the difficulty of power grid channel planning. Therefore, how to scientifically and rationally plan power grid channels in high-altitude areas and make full use of limited line corridor resources is of great significance.

[0003] Traditional power grid channel planning methods mainly rely on field surveys and measurements by staff, supplemented by computer-aided design to develop planning schemes. However, this method has obvious limitations in high-altitude areas with complex terrain and harsh environment. Specifically, this method greatly increases the investment of manpower, material resources and time. In addition, the extreme climatic conditions and complex terrain in high-altitude areas also greatly increase the workload and safety risks of field operations, restricting planning efficiency and accuracy.

[0004] With the rapid development of modern information technology, traditional measurement methods are gradually being replaced. Geographic Information System (GIS) has become an important technology in power grid channel planning due to its powerful spatial analysis capabilities, visual expression and mapping functions. Compared with traditional field surveys, GIS technology can not only improve the speed of data acquisition and processing, but also effectively reduce the number of field operations and reduce operational risks. Global Positioning System (GPS) and remote sensing technology (RS) further enhance the accuracy and extensiveness of data acquisition. Especially in areas with complex terrain and inconvenient transportation, these technologies can remotely obtain a wide range of geographic information and provide accurate spatial data support for the planning and design of power grid channels. At present, power grid channel planning mainly adopts the integration of traditional measurement methods with GIS technology, GPS technology and remote sensing technology, and utilizes the advantages of modern information technology to improve the efficiency and accuracy of planning and design to a certain extent. However, the existing technical means still have certain limitations. For example, although GIS, GPS and RS have advantages in data acquisition and analysis, they are still insufficient in data fusion management and collaborative computing, and there is a problem of inconvenience in integrating multi-source heterogeneous data. In addition, the process of grid channel planning often requires manual intervention, which is limited by the workers' technical level and thus affects the accuracy and precision of the final results.

[0005] In the process of selecting power grid channels, it is necessary to comprehensively consider a variety of factors, including the actual conditions of surrounding landforms, line facilities, meteorological conditions and construction difficulties, as well as the participation of multiple departments in collaborative design and real-time interaction. However, most of the existing channel selection methods rely on manual work, which has obvious limitations in high-altitude areas, mainly manifested in low accuracy and low efficiency, which increases the difficulty and cost of manual work. In addition, although the further integration of GIS technology, GPS technology, and RS technology can provide important data support for the planning and design of power lines, this type of method lacks consideration in data fusion management and collaborative computing. The selection of power grid channels involves the coordinated management of multiple data, and the data sources are diverse and scattered, and the processing is complex. How to efficiently integrate these multi-source heterogeneous data to form a systematic power grid channel design plan is still a problem that needs to be considered.

[0006] The introduction of knowledge graph technology provides a new solution to this challenge. Knowledge graphs can effectively organize and integrate multi-source heterogeneous data, break data silos, and improve data interoperability, especially in multi-departmental collaboration. Although knowledge graphs perform well in data fusion, how to apply geospatial data to specific scenarios such as power grid channel planning by designing effective knowledge reasoning mechanisms is still a problem that has not been fully solved in current research. Summary of the invention

[0007] The present invention proposes a method, device, equipment, and medium for selecting power grid channels based on knowledge graphs, and takes the selection of power grid channels in high-altitude areas as an application demonstration. By integrating multi-source heterogeneous data obtained by multiple means such as geospatial information, remote sensing, and expert surveys, combined with inference engine programs and land suitability evaluation algorithms, the automated evaluation of power grid channels in high-altitude areas is achieved. This method not only improves the efficiency and accuracy of planning and design, but also effectively reduces the intensity and cost of manual work in high-altitude areas, providing an innovative solution for power grid channel planning.

[0008] The present invention proposes a power grid channel selection method based on knowledge graph, and the method specifically comprises the following steps:

[0009] Step 1: Build a knowledge graph framework to integrate multi-source heterogeneous data required for power grid channel planning, build logical associations and semantic models, and provide data and knowledge support for power grid channel planning;

[0010] Step 1 specifically includes:

[0011] Step 1.1: Construct an ontology model to define concepts, attributes and their relationships related to high-altitude power grid channel planning and form a semantic expression framework for knowledge;

[0012] Step 1.2: Build a data base to collect and standardize multi-source heterogeneous data to form a basic data layer to support knowledge graph reasoning and calculation;

[0013] Step 1.3: Build an expert knowledge rule base to formalize expert experience and domain knowledge, form a rule base to support knowledge reasoning, and improve the specialization level of the power grid channel planning model;

[0014] Step 2: Construct a power grid channel planning model; it is used to combine knowledge graph technology with actual power grid channel planning needs, build a power grid channel planning model supported by knowledge graph, and realize intelligent planning of power grid channels.

[0015] The present invention also provides a power grid channel selection device based on knowledge graph, comprising the following modules:

[0016] Knowledge graph framework: used to integrate multi-source heterogeneous data required for power grid channel planning, build logical associations and semantic models, and provide data and knowledge support for planning; the knowledge graph framework specifically includes: ontology model: used to define concepts, attributes and their relationships related to power grid channel planning in high-altitude areas, and form a semantic expression framework for knowledge; data base: used to collect and standardize multi-source heterogeneous data to form a basic data layer to support knowledge graph reasoning and calculation; expert knowledge rule base: used to formalize expert experience and domain knowledge, form a rule base to support knowledge reasoning, and improve the professional level of power grid channel planning models;

[0017] Power grid channel planning model: used to combine the knowledge graph with actual planning needs, build a planning model supported by the knowledge graph, and realize intelligent planning of power grid channels.

[0018] The present invention also provides an electronic device, comprising: one or more processors; and 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 above-mentioned method.

[0019] The present invention also provides a computer-readable storage medium on which executable instructions are stored. When the instructions are executed by a processor, the processor implements the above method.

[0020] The present invention has the following beneficial effects:

[0021] 1. In the prior art, power grid channel planning involves the processing and integration of multi-source heterogeneous data. Due to the scattered data sources and inconsistent formats, data integration efficiency is low, and it is difficult to achieve efficient unified management, which has created certain obstacles to power grid channel planning. The present invention uses knowledge graph technology to effectively organize and manage multi-source heterogeneous data and build a data base, which to a certain extent solves the problem of inefficiency caused by data dispersion in traditional technologies, improves the operability of data, and provides data support for planning decisions of power grid channels.

[0022] 2. The knowledge graph reasoning technology in the prior art is difficult to cope with complex and changeable spatiotemporal scenarios, which limits the application effect of knowledge graphs in business scenarios. Based on the knowledge graph rule reasoning technology, the present invention proposes an optimization algorithm for land suitability scoring. The algorithm can collaboratively process a variety of spatial data in business scenarios, and automatically evaluate site suitability based on the geographical characteristics and planning requirements of the data, thereby improving the spatial analysis capabilities of knowledge graph technology.

[0023] 3. The existing knowledge graph application scenario design is usually relatively fixed, and it is difficult to flexibly respond to changing business needs and environmental changes. The present invention supports users to flexibly adjust and expand the expert knowledge rule base according to different business needs through modular design and customized knowledge graph rule base, thereby ensuring the application effect of knowledge graph in various application scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 This is a diagram illustrating the power grid channel selection method based on the knowledge graph of the present invention;

[0025] Figure 2 It is the conceptual structure diagram of the ontology model;

[0026] Figure 3 It is a grid coding structure diagram;

[0027] Figure 4 This is the inference rule structure diagram. DETAILED DESCRIPTION

[0028] In order to make the purpose, technical scheme and advantages of the present invention more clearly understood, the present invention is 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 intended to limit the present invention. In addition, the technical features involved in each embodiment 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-mentioned purpose, the present invention adopts the following technical scheme.

[0029] The present invention provides a power grid channel selection method based on knowledge graph, such as Figure 1 As shown, the specific steps include:

[0030] Step 1: Build the knowledge graph framework; including:

[0031] Step 1.1: Construct the ontology model;

[0032] The ontology model of the power grid channel planning of the present invention includes three parts: domain ontology, space ontology and reasoning rule ontology, which together constitute the conceptual constraints of knowledge in the knowledge graph. The conceptual structure of the ontology model is shown in Figure 2 The present invention uses the Protégé ontology modeling tool to construct an ontology model.

[0033] The domain ontology covers entity types, attributes, and relationships between entities related to power grid channel planning. The entities in this domain ontology include: research area, object model, field model, road, building, ice zone, wind zone, nature reserve, and road is divided into national highway, provincial highway, and county highway. The relationships between entities include: roads include national highways, roads include provincial highways, and roads include county highways; buildings belong to object models, roads belong to object models, and nature reserves belong to object models; ice zones belong to field models, and wind zones belong to field models.

[0034] Spatial ontology defines and expresses the spatial characteristics of geographic entities in the knowledge graph. It uses the Geographic Semantic Query Language (GeoSPARQL) to provide standardized representation and retrieval capabilities for spatial information. Spatial ontology uses features to store attribute information of geographic entities, uses points, lines, and surfaces to describe the geometric characteristics of various entities, and uses geometric operations to calculate spatial relationships between entities.

[0035] The reasoning ontology defines the concept of automated reasoning rules in the knowledge graph. The concept of reasoning rules combines the expert knowledge of the power grid channel planning field business to guide the reasoning engine program to realize the intelligent planning of power grid channels. The root element of the reasoning rule is the rule object, and the rule object entity is composed of event objects and action objects. The event object defines the conditions under which the reasoning rule is triggered; the action object defines the action performed after the reasoning rule is triggered, and the action is specifically implemented by the action function.

[0036] Step 1.2: Build the data base;

[0037] The data base is composed of instantiated source data. The instantiation process is to map specific data to entities and relationships in the knowledge graph through knowledge extraction and fusion.

[0038] The data source of the present invention covers a variety of spatial data related to power grid channel planning, including buildings, roads, nature reserves, ice areas and wind areas. By analyzing these spatial data, the geographic entities and attributes related to the power grid channel planning business are extracted, and the standardized expression of the spatial information of the geographic entities is realized based on the spatial ontology. Through the guidance of the ontology model, it is ensured that the geographic entities and their attributes from different sources can be accurately associated, thereby achieving the consistency and operability of the data. In this process, the present invention uses the Web Mercator projection grid partitioning method to generate a grid code for the geographic space where the geographic entity is located, so as to realize the spatial indexing of the geographic entity. The method recursively partitions the earth's surface according to different levels by projecting the geographic coordinates into a plane grid system, and maps the geographic entities to grids of different scales based on the spatial position relationship. The grid code consists of x, y, and z, where x represents the horizontal coordinate of the Web Mercator projection grid partitioning space, y represents the vertical coordinate, and z is the partitioning level. At the initial level (0), the entire earth is projected on only one grid, that is, the side length of this grid represents 40075016.69 meters. The subdivision is realized by the quadtree, that is, after zooming in one level, the original grid is split into 4 grids, and the actual length represented by the grid side length becomes half of the original, see Figure 3 The following are the actual distances represented by the grid side lengths corresponding to different subdivision levels: 152.87m when z=18; 76.43m when z=19; 38.21m when z=20; ...; 0.60m when z=26. These grids construct a unified spatial index for the geographic entities in the data base, providing a positioning basis for subsequent spatial retrieval and calculation.

[0039] In the present invention, the main objects of data instantiation include roads, buildings, ice areas, windy areas and nature reserves. The spatial location information and attributes of each geographic entity are generated into triples through standardized semantic representation, mapped to the entities and relationships of the ontology, so as to construct the data base in the knowledge graph. For example, the building entity can be instantiated into the following triple: "building_06c94da8-0c0d type building", indicating that the entity type numbered "building_06c94da8-0c0d" is "building";

[0040] “Building_06c94da8-0c0d hasExactGeometry http: / / www.opengis.net / def / cr s / OGC / 1.3 / CRS84 Polygon ((120.602529644966 31.3406950235367,…,120.602529644966 31.3406950235367))” indicates that the exact geometry information (Exact Geometry) of the building entity numbered “Building_06c94da8-0c0d” is a polygon represented by the geo:wktLiteral data format in the GeoSPARQL spatial ontology. The coordinate system is OGC CRS84, and its geometry is defined by a set of latitude and longitude coordinates. In addition, each geographic entity corresponds to a series of grid codes, for example: "Building_06c94da8-0c0dhasTileCode z12_x3420_y1672", which means that the entity Building_06c94da8-0c0d is located on the grid coded as "z19_x3420_y1672", and the code represents the Mercator projection grid subdivision space with a horizontal coordinate of 3420, a vertical coordinate of 1672, and a subdivision level of 19.

[0041] Step 1.3: Build an expert knowledge rule base;

[0042] The construction of the expert knowledge rule base is intended to systematically store and manage the complex business logic and expert experience in the process of power grid channel planning. In order to ensure the flexibility and scalability of the rule base, the present invention uses a method independent of the inference engine code program, that is, the expert knowledge abstracted into rules is represented by a mind map in the .POS file format. In the process of rule base construction, various factors in power grid channel planning are extracted under the guidance of domain experts, and their weights and evaluation criteria are determined. For example, for data expressed in field models such as ice areas and wind areas, the degree of ice and wind speed of their attribute values ​​are considered, and a score value is assigned, that is, the higher the degree of ice and wind speed, the lower the suitability score of the corresponding area. For data expressed in object models such as roads, buildings, and nature reserves, the suitability score criteria are set according to the distance. The closer the distance, the higher or lower the suitability score. The evaluation factors and weights that have been regularized will be converted into rules that can be recognized and executed by the inference mechanism.

[0043] The expert knowledge rule base is connected to the data base of the knowledge graph and the inference machine code program through the reasoning ontology. Each rule is stored as an independent semantic entity and is associated with its triggering condition event object and the action object that executes the response. The reasoning rule logic of power grid channel planning is: when the study area is input into the reasoning program, it is judged whether the data base contains buildings, roads, nature reserves, wind zones, and ice zone data located in the area. If so, these data are obtained and the rule is triggered. After the rule is triggered, the corresponding action is executed to obtain the wind zone, ice zone data, buildings, roads, and nature reserve data, and the corresponding scores and weight configurations are obtained, and the "land suitability score action function" is executed. The rule structure of this reasoning can be seen in Figure 4 In the figure, the “xx score and weight configuration” is replaced by real data and score in the application. Taking buildings as an example, the format of building score and weight configuration can be: {“Label”:“Building”, “Score”:“0-800:0;800-1800:4;1800-3000:7;3000-99999:10;”, “Weight”:“0.39”}

[0044] Meaning: When the distance between a location in the study area and the "building" entity is within 0-800m, the suitability of the location is assigned 0 points, within 800-1800m, the suitability of the location is assigned 4 points, within 1800-3000m, the suitability of the location is assigned 7 points, and when it is greater than 3000m, the suitability of the location is assigned 10 points. After calculating the suitability scores of all types of data, read the scores and corresponding weights, and multiply and add them to obtain the final suitability score, which is the result of the power grid channel planning.

[0045] By building an expert knowledge rule base, it supports the automated execution of complex reasoning operations in power grid channel planning, reduces manual intervention, and improves planning efficiency and accuracy. When business needs change, by re-reading different types of geographic data in the data base and designing dedicated business rules, the suitability scoring model can be reused in multiple scenarios. Since the code of the rule base and the reasoning mechanism is separated, it is easier to update the business of the rule base, which enables the knowledge graph to adapt to changing business needs and new technologies.

[0046] Step 2 constructs a power grid channel planning model;

[0047] The core idea of ​​the power grid channel planning model is to assign a suitability score to each location in the study area to evaluate its suitability as a candidate for a power grid channel. The higher the score, the more suitable the location is as a candidate for a power grid channel. For the grids that construct the spatial index, at a certain subdivision level, the actual ground size represented by each grid is consistent, so it is only necessary to calculate the distance based on the relative position of the grid and the center point of each grid as the reference. Considering the scale range of the study area, the project accuracy requirements, and the computational efficiency, a more suitable subdivision level is selected for scoring.

[0048] Construct a suitability score dictionary: {"grid code": {"geographic entity type": {"suitability score": double, "weight": double}}}. Traverse each grid at a specific level and construct dictionary items based on the grid code. Determine whether the grid is associated with data. If no geographic data is associated, skip it. If the grid is associated with field model data such as "wind zone", update the suitability score dictionary based on the average wind intensity and score and weight configuration of the grid's location. If the grid is associated with object model data such as "building", mark the grid as unsuitable as a power grid channel, and update the corresponding item of the grid in the suitability score dictionary: {"grid code": {"building": {"suitability score": 0, "weight": 0.39}}. For the surrounding grids, calculate the distance based on the relationship between the spatial horizontal and vertical coordinates of the grid code and the grid side length. The calculation formula is as follows, where is the distance between two grids, is the ground distance represented by the side length of the grid, The x and y numbers of the two grids.

[0049] ,

[0050] After the distance is calculated, the suitability score dictionary is updated according to the score and weight configuration. If a geographic entity type in a grid is scored multiple times, the minimum suitability score is retained.

[0051] After traversing all grids, each grid has its corresponding score, including the score of the geographic entity type and the corresponding weight. The weighted average of these scores is calculated as the final suitability score of the grid and stored in the newly constructed grid suitability dictionary {"grid code": "suitability score value"}. Finally, read the code of the grid, obtain the coordinate values ​​of its four corners, and use the suitability score value as an attribute. According to expert knowledge, the score value is divided into types such as "prohibited", "not recommended", "general", "recommended", and "best", and the power grid channel planning results are output in GeoJSON format.

[0052] This method is a general algorithm that controls the influencing factors involved in the evaluation through defined expert knowledge rules and assigns evaluation factors and weights to each factor to achieve land suitability analysis.

Claims

1. A power grid channel selection method based on knowledge graph, characterized in that: The method specifically comprises the following steps: Step 1: Build a knowledge graph framework to integrate multi-source heterogeneous data required for power grid channel planning, build logical associations and semantic models, and provide data and knowledge support for power grid channel planning; Step 1 specifically includes: Step 1.1: Construct an ontology model to define concepts, attributes and their relationships related to high-altitude power grid channel planning and form a semantic expression framework for knowledge; Step 1.2: Build a data base to collect and standardize multi-source heterogeneous data to form a basic data layer to support knowledge graph reasoning and calculation; Step 1.3: Build an expert knowledge rule base to formalize expert experience and domain knowledge, form a rule base to support knowledge reasoning, and improve the specialization level of the power grid channel planning model; Step 2: Construct a power grid channel planning model; it is used to combine knowledge graph technology with actual power grid channel planning needs, build a power grid channel planning model supported by knowledge graph, and realize intelligent planning of power grid channels; Step 2 is as follows: The grid corridor planning model assigns a suitability score to each location in the study area to assess its suitability as a grid corridor candidate; Construct a suitability score dictionary: {"grid code": {"geographic entity type": {"suitability score": double, "weight": double}}}; traverse each grid at a specific level, construct a dictionary item based on the grid code, and determine whether the grid is associated with data. If no geographic data is associated, skip it. If the grid is associated with field model data such as wind zone, update the suitability score dictionary based on the average wind intensity and score and weight configuration of the grid location. For the surrounding grids, calculate the distance based on the relationship between the spatial horizontal and vertical coordinates of the grid code and the grid side length. The calculation formula is as follows: ， in, is the distance between two grids, is the ground distance represented by the side length of the grid, are the x and y numbers of the two grids; i and j are index numbers. After calculating the distance, the suitability score dictionary is updated according to the score and weight configuration. After traversing all the grids, each grid has its corresponding score, including the score and corresponding weight of the geographic entity type. The weighted average of these scores is calculated as the final suitability score of the grid and stored in the newly constructed grid suitability dictionary {"grid code": "suitability score value"}. Finally, read the code of the grid, obtain the coordinate values ​​of its four corners, and use the suitability score value as an attribute. According to expert knowledge, the score value is divided into prohibited, not recommended, general, recommended, and optimal types, and the power grid channel planning results are output in GeoJSON format.

2. According to a knowledge graph-based power grid channel selection method according to claim 1, it is characterized in that: Step 1.1 is as follows: The ontology model of power grid channel planning includes three parts: domain ontology, spatial ontology and reasoning rule ontology, which together constitute the conceptual constraints of knowledge in the knowledge graph. The domain ontology covers entity types and their attributes related to power grid channel planning. The entities in this domain ontology include: study area, roads, buildings, ice areas, wind areas, and nature reserves. Roads are divided into national roads, provincial roads, and county roads. The relationships between entities include: roads contain national roads, roads contain provincial roads, and roads contain county roads. Buildings belong to object models, roads belong to object models, and nature reserves belong to object models. Ice areas belong to field models, and wind areas belong to field models. Spatial ontology defines and expresses the spatial characteristics of geographic entities in the knowledge graph. It uses geographic semantic query specifications to provide standardized representation and retrieval query capabilities for spatial information. Spatial ontology uses elements to store attribute information of geographic entities, uses points, lines, and surfaces to describe the geometric characteristics of various geographic entities, and realizes the calculation of spatial relationships between geographic entities through geometric operations. The inference rule ontology defines the concept of automated inference rules in the knowledge graph; the root element of the inference rule is the rule object, and the rule object entity is composed of event objects and action objects. The event object defines the conditions under which the inference rule is triggered; the action object defines the action performed after the inference rule is triggered, and the action is specifically implemented by the action function.

3. The method for selecting a power grid channel based on a knowledge graph according to claim 2, characterized in that: Step 1.2 is as follows: The data base is composed of instantiated source data. The instantiation process is to map specific data to entities and relationships in the knowledge graph through knowledge extraction and fusion. The source data covers a variety of spatial data related to power grid channel planning, including buildings, roads, nature reserves, ice areas and wind areas. By analyzing these spatial data, business-related geographic entities and attributes are extracted, and standardized expression of spatial information is achieved based on spatial ontology. Through the guidance of the ontology model, geographic entities and their attributes from different sources can be accurately associated, thereby achieving data consistency and operability.

4. The method for selecting a power grid channel based on a knowledge graph according to claim 3, characterized in that: Step 1.2 also includes: The Web Mercator projection grid partitioning method is used to generate grid codes for the geographic space where the geographic entities are located to realize the spatial indexing of geographic entities. The Web Mercator projection grid partitioning method recursively partitions the earth's surface at different levels by projecting geographic coordinates into a plane grid system, and maps geographic entities to grids of different scales based on spatial position relationships. The grid code consists of x, y, and z, where x represents the horizontal coordinate of the Web Mercator projection grid partitioning space, y represents the vertical coordinate, and z is the partitioning level.

5. The method for selecting a power grid channel based on a knowledge graph according to claim 3, characterized in that: The instantiated objects of the source data include roads, buildings, ice areas, windy areas and nature reserves; the spatial location information and attributes of each geographic entity are generated into triples through standardized semantic representation, mapped to the entities and relationships of the ontology, thereby constructing the data base in the knowledge graph.

6. A method for selecting a power grid channel based on a knowledge graph according to claim 4, characterized in that: Step 1.3 is as follows: The construction of the expert knowledge rule base is used to systematically store and manage the business logic and expert experience in the process of power grid channel planning; the expert knowledge abstracted into rules is represented by a mind map in POS file format using the inference engine program. In the process of building the expert knowledge rule base, various factors in power grid channel planning are extracted under the guidance of domain experts, and their weights and evaluation criteria are determined. For the data expressed in the field model of ice and wind areas, the attribute values ​​of ice degree and wind speed are considered and assigned a score value, that is, the higher the ice degree and wind speed, the lower the suitability score of the corresponding area; for the data expressed in the object model of roads, buildings, and nature reserves, the suitability score standard is set according to the distance, and the closer the distance, the lower the suitability score; The expert knowledge rule base is connected to the data base and inference engine program of the knowledge graph through the reasoning ontology. Each rule is stored as an independent semantic entity and is associated with its trigger condition event object and the action object that executes the response. The reasoning rule logic of the power grid channel planning is: when the study area is input, it is determined whether the data base contains buildings, roads, nature reserves, wind zones, and ice zone data located in the area. If so, these data are obtained and the rule is triggered. After the rule is triggered, the corresponding action is executed to obtain the wind zone, ice zone data, buildings, roads, and nature reserve data, and the corresponding scores and weight configurations are obtained. The action function land suitability score is executed. After calculating the suitability scores of all types of data, the scores and corresponding weights are read, and the final suitability score, that is, the result of the power grid channel selection, is obtained by multiplication and addition.

7. A selection device for a power grid channel selection method based on a knowledge graph according to any one of claims 1 to 6, characterized in that: Includes the following modules: Knowledge graph framework; It is used to integrate the multi-source heterogeneous data required for power grid channel planning, build logical associations and semantic models, and provide data and knowledge support for planning; the knowledge graph framework specifically includes: ontology model; used to define concepts, attributes and their relationships related to power grid channel planning in high-altitude areas, and form a semantic expression framework for knowledge; data base; used to collect and standardize multi-source heterogeneous data to form a basic data layer to support knowledge graph reasoning and calculation; expert knowledge rule base; used to formalize expert experience and domain knowledge, form a rule base to support knowledge reasoning, and improve the professional level of power grid channel planning models; Power grid channel planning model: used to combine knowledge graph technology with actual planning needs, build a planning model supported by knowledge graph, and realize intelligent planning of power grid channels.

8. 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 a knowledge graph-based power grid channel selection method as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that: Executable instructions are stored thereon, and when the instructions are executed by the processor, the processor implements a power grid channel selection method based on a knowledge graph as described in any one of claims 1 to 6.

Images