Power transmission line path planning method and system based on natural language and knowledge graph

Abstract

The invention discloses a power transmission line path planning method and system based on a natural language and a knowledge graph, and the method comprises the steps: analyzing a natural language path planning target through a large language model, obtaining an intention classification result and a named entity, mapping the intention classification result and the named entity to a business rule knowledge graph for reasoning, and generating a task step sequence; receiving confirmation or modification of a user on each task step to obtain each target task step; searching a combination corresponding to each target task step from the GIS operation knowledge graph by utilizing a large language model, and disassembling the combination into a GIS operator call chain; and calling a corresponding GIS tool through an MCP protocol, executing spatial analysis operation in the GIS operator calling chain corresponding to each target task step according to the sequence of all the target task steps, and generating a path planning result of the power transmission line. According to the invention, full automation, process interpretability and man-machine interaction cooperation of power transmission line path planning are realized, and the power transmission line path planning efficiency is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power engineering planning and design, and particularly relates to a power transmission line path planning method and system based on natural language and a knowledge graph. BACKGROUND

[0002] At present, computer-aided path planning technology has been widely applied in the field of power transmission line route selection, and the method based on traditional A algorithm and its variants (such as weighted A and dynamic weighted A*) is the most mainstream technical solution. This method usually relies on professional engineers to perform a large number of manual operations in the GIS software environment: first, the real geographical environment is abstracted into a grid or network graph structure, and the value of each node is calculated based on a single or a small number of fixed factors such as slope and land use type to construct a static cost grid; then, A* algorithm is configured by writing scripts or calling software tools (such as the Path Distance tool of ArcGIS), and the search is expanded from the starting point to the ending point using the evaluation function f(n) = g(n) + h(n), the node with the smallest f(n) value is preferentially selected, and finally a path with the smallest value is output. The entire process relies on pre-solidified cost model and algorithm parameters, and the intelligentization and automation level is low.

[0003] However, the above-mentioned prior art has the following defects: first, the degree of intelligentization and automation is insufficient, and it highly depends on professional manual intervention in data preparation, cost grid generation, parameter configuration and result verification, etc. It is difficult to convert high-level task objectives into automatic execution processes, and it is difficult to achieve the intelligent experience of "natural language demand input, planning result automatic output". Second, it cannot understand complex natural language constraints and expert experience, and its cost model is solidified in the program script, which cannot dynamically identify instructions such as "avoid residential areas" and "try to be close to the highway" that contain spatial relationships and preference intentions. The constraint conditions need to be manually adjusted parameters or even modified codes by engineers, and the flexibility is poor. Third, the multi-constraint fusion capability is rigid, and the weights of terrain, ecology and social cost are usually pre-set and fixed, which cannot adapt to complex scenarios such as "considering cost in general area and strictly avoiding in sensitive area" that require dynamic adjustment of weights, and often requires multiple independent operations or complex post-processing to meet the requirements. Fourth, the decision-making process is a "black box" feature, only the final path is output without explainability, users cannot know which rules the path is based on, which areas are avoided, and how the weights are allocated. When the result is not satisfactory, the parameters can only be blindly adjusted based on experience, and the trial and error cost is high.

[0004] Therefore, it is urgent to provide a technical solution to solve the above problems. SUMMARY

[0005] To solve the above technical problems, the present application provides a power transmission line path planning method and system based on natural language and a knowledge graph.

[0006] In a first aspect, the present application provides a power transmission line path planning method based on natural language and knowledge graph, and the technical scheme of the method is as follows: receiving a natural language path planning target input by a user for a target power transmission line; analyzing the natural language path planning target by using a large language model to obtain an intention classification result and at least one named entity; wherein the at least one named entity includes at least one of a starting point, an ending point, an object to be avoided, and a preferred object; mapping the intention classification result and the at least one named entity to a business rule knowledge graph for reasoning to generate a task step sequence; displaying the task step sequence to the user through a visual interface and receiving confirmation or modification of each task step in the task step sequence by the user through the visual interface to obtain each target task step; retrieving, from a GIS operation knowledge graph, a combination corresponding to each target task step by using the large language model, the combination being composed of a plurality of GIS tools and corresponding execution sequences, and disassembling the combination corresponding to each target task step into a GIS operator call chain; calling corresponding GIS tools through an MCP protocol, and sequentially executing spatial analysis operations in the GIS operator call chain corresponding to each target task step in the order of all target task steps to generate a path planning result of the target power transmission line.

[0007] The power transmission line path planning method based on natural language and knowledge graph has the following beneficial effects: The method analyzes a natural language target by using a large language model and generates a task step sequence based on a business rule knowledge graph, automatically schedules tools for executing spatial analysis by combining a GIS operation knowledge graph and an MCP protocol, solves the problems of relying on manual work, being unable to understand complex constraints, and non-transparent decision-making process in traditional methods, realizes full automation, process explainability, and human-computer interaction collaboration of power transmission line path planning, and improves the efficiency of power transmission line path planning.

[0008] Based on the above-mentioned scheme, the power transmission line path planning method based on natural language and knowledge graph can be further improved as follows.

[0009] In an optional manner, the step of analyzing the natural language path planning target by using a large language model to obtain an intention classification result and at least one named entity includes: The large language model is used for intent classification processing on the natural language path planning target, and an intent classification result of the natural language path planning target in a path planning task of the target power transmission line is output. The large language model is used for named entity recognition processing on the natural language path planning target, and at least one named entity is recognized. The at least one named entity includes at least one of a starting point, an ending point, an object to be avoided, and a preferred object.

[0010] The above optional mode has the beneficial effect that the large language model performs intent classification and named entity recognition, accurately extracts key information such as a starting point, an ending point, an object to be avoided, and a preferred object, improves natural language understanding accuracy, provides a structured data basis for subsequent knowledge graph reasoning, and enhances the analysis capability of the path planning target.

[0011] In an optional mode, the step of mapping the intent classification result and the at least one named entity to a business rule knowledge graph for reasoning to generate a task step sequence includes: The intent classification result and the at least one named entity are matched with nodes in the business rule knowledge graph to obtain matched nodes. Based on a logical relationship represented by an edge defined in the business rule knowledge graph and a reasoning rule, reasoning is performed between the matched nodes to obtain a reasoning result. According to the reasoning result, at least one task step is extracted from the business rule knowledge graph, and a task step sequence is generated according to the execution order of each task step.

[0012] The above optional mode has the beneficial effect that the analysis result is matched with the business rule knowledge graph nodes, and logical reasoning is performed based on the edge relationship to automatically extract an ordered task step sequence, realizing formal expression and automatic application of complex business rules, and improving the accuracy and explainability of task decomposition.

[0013] In an optional mode, the task step sequence is displayed to the user through a visual interface, and confirmation or modification of each task step in the task step sequence by the user through the visual interface is received to obtain each target task step, including: Each task step in the task step sequence is displayed to the user through the visual interface. For each task step in the task step sequence, a confirmation instruction or a modification instruction issued by the user through the visual interface is received. When a modification instruction for any task step is received, the modification content of the task step is received through the visual interface. Based on the confirmation instruction or the modification instruction and the corresponding modification content for each task step, a target task step sequence with the same number of task steps as in the task step sequence is formed.

[0014] The beneficial effect of the above optional mode is that the task step sequence is further displayed through the visual interface and the user is supported to confirm or modify each step, realizing human-computer interaction collaboration, making the planning process transparent and controllable, allowing the user to adjust the automatically generated steps according to professional experience, and improving flexibility and user satisfaction.

[0015] In an optional mode, the large language model is used to retrieve a combination corresponding to any target task step from the GIS operation knowledge graph, the combination including a plurality of GIS tools and corresponding execution sequences, and the combination corresponding to the any target task step is disassembled into steps of a GIS operator call chain, including: The large language model is used to retrieve a plurality of GIS tools corresponding to the any target task step from the GIS operation knowledge graph; Based on the execution sequence relationship between the tools represented by the edges defined in the GIS operation knowledge graph, the execution sequence of the plurality of GIS tools corresponding to the any target task step is determined; According to the execution sequence corresponding to the any target task step, the plurality of GIS tools corresponding to the any target task step are disassembled into an ordered GIS operator call chain.

[0016] The beneficial effect of the above optional mode is that the plurality of tools corresponding to the target task step are further retrieved from the GIS operation knowledge graph and the execution sequence is determined, and the complex GIS operation is disassembled into an ordered operator call chain, realizing automatic arrangement of the spatial analysis process and improving GIS tool calling accuracy and execution efficiency.

[0017] In an optional mode, the step of calling the corresponding GIS tool through the MCP protocol, sequentially executing the spatial analysis operation in the GIS operator call chain corresponding to each target task step according to the order of all target task steps, and generating the path planning result of the target transmission line includes: According to the order of all target task steps, each target task step is sequentially processed; For the currently processed target task step, the corresponding GIS tool is called through the MCP protocol, at least one spatial analysis operation including slope analysis operation, crossing detection operation and buffer analysis operation in the GIS operator call chain corresponding to the target task step is executed in combination with the spatial analysis result of the previous target task step, the spatial analysis result of the currently processed target task step is generated and provided as input data to the spatial analysis operation corresponding to the next target task step. After all target task step processing is completed, the path planning result of the target power transmission line is generated by a path search algorithm based on the finally obtained spatial analysis result.

[0018] The beneficial effect of the above optional mode is that: further calling GIS tools in sequence through the MCP protocol to perform spatial operations such as slope analysis, crossing detection and buffer analysis, and combining the previous results for progressive calculation, to realize fine spatial analysis under multiple constraint conditions and generate path planning results that meet complex engineering requirements.

[0019] In an optional mode, the method further comprises: storing the path planning result to a database and performing visual rendering.

[0020] The beneficial effect of the above optional mode is that: further storing the path planning result to a database and performing visual rendering, to realize persistent management and graphical display of the planning results, facilitate subsequent query, analysis and decision support, and improve the usability and shareability of the results.

[0021] In a second aspect, the application provides a power transmission line path planning system based on natural language and knowledge graph, and the technical scheme of the system is as follows: A receiving module is configured to receive a natural language path planning target input by a user for a target power transmission line. An analysis module is configured to analyze the natural language path planning target by using a large language model to obtain an intent classification result and at least one named entity, wherein the at least one named entity includes at least one of a starting point, an ending point, an object to be avoided, and a preferred object. A reasoning module is configured to map the intent classification result and the at least one named entity to a business rule knowledge graph for reasoning to generate a task step sequence. An adjustment module is configured to display the task step sequence to the user through a visual interface, receive confirmation or modification of each task step in the task step sequence by the user through the visual interface, and obtain each target task step. A retrieval module is configured to use the large language model to retrieve a combination corresponding to each target task step from a GIS operation knowledge graph, the combination being composed of a plurality of GIS tools and corresponding execution sequences, and to disassemble the combination corresponding to each target task step into a GIS operator call chain. An execution module is configured to call corresponding GIS tools through an MCP protocol, perform spatial analysis operations in the GIS operator call chain corresponding to each target task step in sequence according to the order of all target task steps, and generate a path planning result of the target power transmission line.

[0022] The power transmission line path planning system based on natural language and a knowledge graph has the following beneficial effects: The system analyzes a natural language target through a large language model, generates a task step sequence based on a business rule knowledge graph, and automatically schedules a tool for executing spatial analysis in combination with a GIS operation knowledge graph and an MCP protocol, thereby solving the problems of a traditional mode, such as dependence on manual work, incapability of understanding complex constraints, and non-transparent decision-making process, and realizing full automation, process explainability, and human-computer interaction collaboration of power transmission line path planning, and improving the efficiency of power transmission line path planning.

[0023] In a third aspect, the technical scheme of the electronic device of the present application is as follows: The electronic device comprises a memory, a processor, and a program stored in the memory and running on the processor, and the processor implements the steps of the power transmission line path planning method based on natural language and a knowledge graph of the present application when executing the program.

[0024] In a fourth aspect, the technical scheme of the computer readable storage medium provided by the present application is as follows: The computer readable storage medium stores instructions, and when the computer readable storage medium reads the instructions, the computer readable storage medium executes the steps of the power transmission line path planning method based on natural language and a knowledge graph of the present application.

[0025] The above description is only a summary of the technical scheme of the present application, in order to more clearly understand the technical means of the present application, the specific embodiments of the present application can be implemented according to the content of the specification, and in order to make the above and other purposes, characteristics and advantages of the present application more obvious and easy to understand, the following specific embodiments of the present application are described. BRIEF DESCRIPTION OF DRAWINGS

[0026] The accompanying drawings are only used to show the embodiments and are not considered as limitations of the present application. Moreover, the same reference signs are used to represent the same parts throughout the drawings. In the drawings: Figure 1 It is a flowchart of an embodiment of the power transmission line path planning method based on natural language and a knowledge graph of the present application; Figure 2 It is a whole flowchart; Figure 3 It is a structure diagram of an embodiment of the power transmission line path planning system based on natural language and a knowledge graph of the present application; Figure 4 It is a structure diagram of an embodiment of the electronic device of the present application. DETAILED DESCRIPTION

[0027] Exemplary embodiments of the present application will be described in greater detail below with reference to the accompanying drawings. While exemplary embodiments of the present application are shown in the drawings, it is understood that the present application can be embodied in various forms and should not be limited by the embodiments set forth herein.

[0028] Figure 1 An embodiment of a flowchart of a power transmission line path planning method based on natural language and a knowledge graph provided by the present application is shown. The power transmission line path planning method based on natural language and a knowledge graph can be executed by electronic devices such as terminal devices or servers. The terminal device can be any fixed or mobile terminal such as a user equipment (UE), a mobile device, a user terminal, a terminal, a handheld device, a computing device, a vehicle-mounted device, a wearable device, etc. The server can be a single server or a server cluster composed of multiple servers. Any electronic device can implement the power transmission line path planning method based on natural language and a knowledge graph by calling computer-readable instructions stored in the memory through the processor. As shown in Figure 1 The steps include: S1, receiving a natural language path planning target input by a user for a target power transmission line.

[0029] The user refers to an operator using the power transmission line path planning method, usually a power engineering planning and design personnel. For example, a planning personnel A uses the method to plan the path of a certain power transmission line. The target power transmission line refers to a specific power transmission line that needs to plan a spatial path. For example, a newly built power transmission line from a 500 kV substation in City A to a 500 kV substation in County B. The natural language path planning target refers to the line planning requirements and constraints expressed by the user in daily language. For example, "from substation A to substation B, avoid residential areas and ecological protection areas, and try to lay the line along the expressway".

[0030] S2, using a large language model to analyze the natural language path planning target to obtain an intent classification result and at least one named entity; wherein the at least one named entity includes at least one of a starting point, an ending point, an avoidance object, and a preference object.

[0031] The large language model refers to a large-scale pre-training language understanding artificial intelligence model based on deep learning. For example, a Qwen series model is used to analyze and understand natural language input. The intent classification result refers to the classification identification obtained by the large language model after identifying the purpose of the user input natural language. For example, the user input is classified as the task type of "power transmission line path planning". The named entity refers to the proper noun and key element with specific meaning identified from the user natural language input. For example, the entities "A substation", "B substation", "residential area", "ecological protection area", "highway", etc. are identified from the input. The starting point refers to the starting location of the power transmission line path planning. For example, the location of the 500kV substation in A city. The end point refers to the termination location of the power transmission line path planning. For example, the location of the 500kV substation in B county. The avoidance object refers to the geographical area or element that needs to be avoided in the path planning process. For example, residential area, ecological protection area, military restricted area, etc. The preference object refers to the geographical element that is desired to be close to or considered preferentially in the path planning process. For example, highway, existing power corridor, traffic trunk, etc.

[0032] S3, mapping the intent classification result and the at least one named entity to a business rule knowledge graph for reasoning to generate a task step sequence.

[0033] The business rule knowledge graph refers to a structured knowledge base that stores power transmission line planning and design specifications, expert experience and business rules. For example, the knowledge base contains rules such as "ecological protection area needs to set 200m avoidance buffer" and "residential area needs to set 300m avoidance buffer". The task step sequence refers to an ordered list of operation steps generated based on the reasoning of the business rule knowledge graph. For example, ["generate basic geographic cost raster", "perform ecological protection area avoidance analysis", "conduct highway proximity optimization"].

[0034] S4, displaying the task step sequence to the user through a visual interface, and receiving the user's confirmation or modification of each task step in the task step sequence through the visual interface to obtain each target task step.

[0035] The visual interface refers to a user operation interface that provides graphical information display and interaction functions. For example, a Web page that displays map data and a list of task steps. The task step refers to a single operation link in the path planning process. For example, "perform slope analysis" or "conduct ecological protection area buffer analysis". The target task step refers to the finally determined task step after user confirmation or modification. For example, the buffer analysis step after the user adjusts the ecological protection area avoidance distance from 200m to 300m.

[0036] S5, retrieving, from the GIS operation knowledge graph, a combination corresponding to each target task step by using the large language model, the combination being composed of a plurality of GIS tools and corresponding execution sequences, and disassembling the combination corresponding to each target task step into a GIS operator call chain.

[0037] The GIS operation knowledge graph refers to a structured knowledge base describing the functions, parameters, and calling relationships of geographic information system tools. For example, it includes knowledge bases of "slope analysis tool input and output parameters" and "buffer analysis tool execution conditions". GIS tools refer to software function modules for processing geographic spatial data. For example, the slope analysis tool of the ArcGIS platform and the buffer analysis tool of the QGIS platform. The execution sequence refers to the order in which multiple GIS tools are called. For example, first perform digital elevation model processing, then perform slope calculation, and finally perform slope direction analysis. The combination refers to a group of GIS tools and their execution order organized to complete a specific task step. For example, to complete the "terrain analysis" task step, the combination [slope analysis tool, elevation analysis tool, slope direction analysis tool] is used. The GIS operator call chain refers to an ordered sequence of GIS tool calls. For example, the tool call chain [elevation interpolation → slope calculation → slope direction analysis].

[0038] S6, by calling the corresponding GIS tools through the MCP protocol, sequentially executing the spatial analysis operations in the GIS operator call chain corresponding to each target task step in the order of all target task steps, to generate the path planning result of the target transmission line.

[0039] The MCP (Model Context Protocol) protocol refers to a general context protocol for large model tool calls. For example, it calls spatial analysis functions of different GIS platforms through a standardized interface. Spatial analysis operations refer to processing and analysis calculations on geographic spatial data. For example, slope calculation, buffer generation, overlay analysis, and other geographic processing operations. The path planning result refers to the final generated transmission line path scheme data. For example, it includes the line path and its related attribute information.

[0040] The technical solution of the embodiment analyzes natural language targets through a large language model and generates task step sequences based on business rule knowledge graphs. It automatically schedules tools to perform spatial analysis by combining GIS operation knowledge graphs and MCP protocols, solving the problems of relying on manual work, not understanding complex constraints, and opaque decision-making processes in traditional methods. It realizes full automation, process explainability, and human-computer interaction collaboration in transmission line path planning, and improves the efficiency of transmission line path planning.

[0041] In an optional manner, S2 specifically includes: The large language model performs intent classification processing on the natural language path planning target, and outputs the intent classification result of the natural language path planning target in the path planning task of the target power transmission line.

[0042] The intent classification processing refers to a process in which a large language model identifies the purpose of user input; for example, the model classifies user input into a "path planning" task category. The path planning task refers to a specific power transmission line path planning work item; for example, planning a 500-kilovolt power transmission line path from an alpha substation to a beta substation.

[0043] The large language model performs named entity recognition processing on the natural language path planning target to identify the at least one named entity; wherein the at least one named entity includes at least one of a starting point, an ending point, an avoidance object, and a preference object.

[0044] The named entity recognition refers to a process of identifying specific types of entities from natural language text; for example, identifying "alpha substation" as a starting point entity and "beta substation" as an ending point entity from user input.

[0045] In the above optional manner, the large language model performs intent classification and named entity recognition, accurately extracts key information such as starting point, ending point, avoidance object, and preference object, improves natural language understanding accuracy, provides a structured data foundation for subsequent knowledge graph reasoning, and enhances the analysis capability of the path planning target.

[0046] In an optional manner, S3 specifically includes: The intent classification result and the at least one named entity are matched with nodes in the business rule knowledge graph to obtain matching nodes.

[0047] The matching node refers to a node in the business rule knowledge graph that corresponds to the input entity; for example, a rule node in the knowledge graph that corresponds to the "ecological protection zone" entity.

[0048] Based on the logical relationships and reasoning rules represented by the edges defined in the business rule knowledge graph, reasoning is performed between the matching nodes to obtain a reasoning result.

[0049] The edge defined in the business rule knowledge graph refers to a relationship connecting different nodes in the knowledge graph, for example, a "trigger" relationship edge between the "ecological protection zone" node and the "avoidance rule" node. The logical relationship refers to a logical association between nodes in the knowledge graph, for example, a conditional relationship between the "ecological protection zone" entity and the "must avoid" rule. The inference rule refers to a rule for logical inference based on the knowledge graph; for example, if there is an "ecological protection zone" entity, the "buffer zone analysis" rule is triggered. The inference result refers to a conclusion obtained by reasoning based on the knowledge graph; for example, according to the input conditions, it is inferred that the steps of ["ecological protection zone avoidance analysis", "highway proximity analysis"] need to be performed.

[0050] According to the inference result, at least one task step is extracted from the business rule knowledge graph, and the task step sequence is generated according to the execution order of each task step.

[0051] In the above optional manner, the parsed result is further matched with the business rule knowledge graph node, and logical reasoning is performed based on the edge relationship to automatically extract an ordered task step sequence, realizing formalized expression and automatic application of complex business rules, and improving the accuracy and interpretability of task decomposition.

[0052] In an optional manner, S4 specifically includes: The visual interface displays each task step in the task step sequence to the user.

[0053] For each task step in the task step sequence, a confirmation instruction or a modification instruction issued by the user through the visual interface is received.

[0054] The confirmation instruction refers to an instruction that the user approves the task step; for example, the user clicks the "confirm" button or inputs "confirm" in the dialogue to approve a certain task step. The modification instruction refers to an instruction that the user adjusts the task step; for example, the user selects a certain task step and clicks the "modify" button or inputs "confirm" in the dialogue.

[0055] When a modification instruction for any task step is received, the modification content of the task step is received through the visual interface.

[0056] The modification content refers to specific adjustment information proposed by the user for the task step; for example, the user modifies the ecological protection zone avoidance distance from 200 meters to 300 meters.

[0057] Based on the confirmation instruction or the modification instruction for each task step and the corresponding modification content, a target task step sequence with the same number of task steps as the task step sequence in the task step sequence is formed.

[0058] In the optional manner described above, the sequence of task steps is further displayed through the visualization interface and the user confirms or modifies each step, realizing human-computer interaction collaboration, making the planning process transparent and controllable, allowing the user to adjust the automatically generated steps according to professional experience, and improving flexibility and user satisfaction.

[0059] In an optional manner, a large language model is used to retrieve a combination of multiple GIS tools and corresponding execution sequences corresponding to any target task step from a GIS operation knowledge graph, and the combination corresponding to the any target task step is disassembled into steps of a GIS operator call chain, including: A large language model is used to retrieve multiple GIS tools corresponding to the any target task step from the GIS operation knowledge graph.

[0060] Based on the execution sequence relationship between tools represented by the edges defined in the GIS operation knowledge graph, the execution sequence of the multiple GIS tools corresponding to the any target task step is determined.

[0061] Among them, the edges defined in the GIS operation knowledge graph refer to the data flow and dependency relationship between GIS tools; for example, the data flow edge between the output of the slope analysis tool and the input of the elevation analysis tool. The execution sequence relationship refers to the execution dependency between GIS tools; for example, digital elevation model processing must be completed before subsequent slope calculation can be performed.

[0062] According to the execution sequence corresponding to the any target task step, the multiple GIS tools corresponding to the any target task step are disassembled into an ordered GIS operator call chain.

[0063] In the optional manner described above, the multiple tools corresponding to the target task step are further retrieved from the GIS operation knowledge graph and the execution sequence is determined, the complex GIS operation is disassembled into an ordered operator call chain, the automatic arrangement of the spatial analysis process is realized, and the GIS tool calling accuracy and execution efficiency are improved.

[0064] In an optional manner, S6 specifically includes: According to the order of all target task steps, each target task step is processed in turn.

[0065] For the target task step currently being processed, at least one of the slope analysis operation, the crossing detection operation and the buffer analysis operation included in the GIS operator call chain corresponding to the target task step is executed by invoking the corresponding GIS tool through the MCP protocol, in combination with the spatial analysis result of the previous target task step, to generate a spatial analysis result of the target task step currently being processed, which is provided as input data to the spatial analysis operation corresponding to the next target task step.

[0066] The target task step currently being processed refers to the task step being executed, for example, the "ecological protection zone 300-meter buffer analysis" step is currently being executed. The spatial analysis result refers to the data result generated after the spatial analysis operation is executed, for example, the slope grid data generated after the slope analysis is executed. The slope analysis operation refers to the spatial analysis of calculating the terrain slope value, for example, the slope distribution is calculated based on the digital elevation model data. The crossing detection operation refers to the spatial analysis of detecting the intersection of the line and the geographic feature, for example, the intersection points of the line and the river, road. The buffer analysis operation refers to the spatial analysis of creating a buffer area of a geographic feature, for example, a 300-meter-wide buffer area is generated based on the boundary of the ecological protection zone.

[0067] It should be noted that the spatial analysis operation includes but is not limited to the slope analysis operation, the crossing detection operation and the buffer analysis operation, which are not limited herein.

[0068] After all the target task steps are processed, the path planning result of the target transmission line is generated by a path search algorithm based on the finally obtained spatial analysis result.

[0069] The finally obtained spatial analysis result refers to the comprehensive data obtained after all the spatial analysis operations are completed, for example, the comprehensive cost grid integrating the slope, avoidance area, proximity and other factors. The path search algorithm refers to an algorithm for finding the optimal path based on the spatial analysis result, for example, the A* algorithm is run on the cost grid to find the path with the minimum cost.

[0070] In the above optional manner, the MCP protocol is further used to sequentially invoke GIS tools to execute spatial operations such as slope analysis, crossing detection and buffer analysis, and progressive calculation is performed in combination with the previous result, to realize fine spatial analysis under multiple constraint conditions and generate a path planning result meeting complex engineering requirements.

[0071] In an optional manner, the method further comprises: The path planning result is stored in a database and visualized.

[0072] In this context, "database" refers to a database management system that stores planning data and results; for example, using a PostgreSQL database to store spatial data and planning results. "Visualization rendering" refers to the process of presenting the route planning results graphically; for example, displaying the planned route on a map interface using specific symbols and colors.

[0073] In the above-mentioned optional methods, the path planning results are further stored in a database and visualized to achieve persistent management and graphical display of the planning results, which facilitates subsequent querying, analysis and decision support, and improves the usability and sharing of the results.

[0074] In this embodiment, it should be noted that, as Figure 2 As shown: 1) The user performs a plotting operation on the map interface, drawing two points; the user inputs the natural language path planning objective: "Plan a 110kV pre-feasibility study transmission line from A to B". After receiving this natural language path planning objective, the system uses a large language model to parse the input content, obtain the intent classification result, and identify named entities.

[0075] 2) Map the obtained intent classification results and named entities to the business rule knowledge graph for reasoning to generate a task step sequence. Display the task step sequence to the user through a visual interface, and receive the user's confirmation or modification of each task step to form the target task step sequence.

[0076] 3) Using a large language model, retrieve the combination of multiple GIS tools and their corresponding execution order from the GIS operation knowledge graph for each target task step, and decompose this combination into a GIS operator call chain. Invoke the corresponding GIS tools via the MCP protocol, and execute the spatial analysis operations in the GIS operator call chain corresponding to each target task step in the order of all target task steps.

[0077] 4) During execution, for the current target task step, the spatial analysis results from the previous target task step are combined with the corresponding spatial analysis operations to generate the spatial analysis results for the current step, which are then used as input data for the next target task step. After all target task steps are processed, the path planning results for the target transmission line are generated based on the final obtained spatial analysis results using a path search algorithm. Finally, the path planning results are stored in the database and visualized.

[0078] Figure 3 This diagram illustrates the structure of an embodiment of a power transmission line path planning system 200 based on natural language and knowledge graphs provided by the present invention. Figure 3 As shown, the power transmission line path planning system 200 based on natural language and knowledge graphs includes: The receiving module 201 is configured to receive a natural language path planning target input by a user for a target power transmission line; The parsing module 202 is configured to parse the natural language path planning target by using a large language model to obtain an intent classification result and at least one named entity, wherein the at least one named entity includes at least one of a starting point, an ending point, an object to be avoided, and a preferred object. The reasoning module 203 is configured to map the intent classification result and the at least one named entity to a business rule knowledge graph for reasoning to generate a task step sequence. The adjusting module 204 is configured to display the task step sequence to the user through a visual interface, receive confirmation or modification of each task step in the task step sequence by the user through the visual interface, and obtain each target task step. The retrieval module 205 is configured to use the large language model to retrieve a combination corresponding to each target task step from a GIS operation knowledge graph, the combination being composed of a plurality of GIS tools and corresponding execution sequences, and to decompose the combination corresponding to each target task step into a GIS operator call chain. The execution module 206 is configured to call corresponding GIS tools through an MCP protocol, and sequentially execute spatial analysis operations in the GIS operator call chain corresponding to each target task step in the order of all target task steps to generate a path planning result of the target power transmission line.

[0079] In an optional manner, the parsing module 202 is specifically configured to: perform intent classification processing on the natural language path planning target by using the large language model, and output the intent classification result of the natural language path planning target in the path planning task of the target power transmission line. perform named entity recognition processing on the natural language path planning target by using the large language model to identify the at least one named entity, wherein the at least one named entity includes at least one of a starting point, an ending point, an object to be avoided, and a preferred object.

[0080] In an optional manner, the reasoning module 203 is specifically configured to: match the intent classification result and the at least one named entity with nodes in the business rule knowledge graph to obtain matching nodes. based on logical relationships and reasoning rules represented by edges defined in the business rule knowledge graph, reason between the matching nodes to obtain a reasoning result. According to the inference result, at least one task step is extracted from the business rule knowledge graph, and the task step sequence is generated according to the execution order of each task step.

[0081] In an optional manner, the adjustment module 204 is specifically configured to: exhibit each task step in the task step sequence to the user through the visualization interface; for each task step in the task step sequence, receive a confirmation instruction or a modification instruction issued by the user through the visualization interface; when receiving a modification instruction for any task step, receive the modification content of the task step through the visualization interface; based on the confirmation instruction or the modification instruction for each task step and the corresponding modification content, form a target task step sequence with the same number of task steps as the task step sequence.

[0082] In an optional manner, the retrieval module 205 is specifically configured to: use the large language model to retrieve a plurality of GIS tools corresponding to any target task step from the GIS operation knowledge graph; based on the execution order relationship between the tools represented by the edges defined in the GIS operation knowledge graph, determine the execution order of the plurality of GIS tools corresponding to the any target task step; according to the execution order corresponding to the any target task step, the plurality of GIS tools corresponding to the any target task step are disassembled into an ordered GIS operator call chain.

[0083] In an optional manner, the execution module 206 is specifically configured to: in the order of all target task steps, each target task step is processed in turn; for the currently processed target task step, the corresponding GIS tool is called through the MCP protocol, at least one spatial analysis operation including slope analysis operation, crossing detection operation and buffer analysis operation in the GIS operator call chain corresponding to the target task step is executed combined with the spatial analysis result of the previous target task step, the spatial analysis result of the currently processed target task step is generated and provided as input data to the spatial analysis operation corresponding to the next target task step; after all target task steps are processed, based on the finally obtained spatial analysis result, the path planning result of the target transmission line is generated through a path search algorithm.

[0084] In an optional manner, further comprising: a rendering module; the rendering module is configured to: The path planning result is stored in a database and visualized.

[0085] It should be noted that the beneficial effects of the power transmission line path planning system 200 based on natural language and knowledge graph provided by the above embodiments are the same as those of the power transmission line path planning method based on natural language and knowledge graph, and will not be repeated here. In addition, when the system provided by the above embodiments implements its functions, only the division of the above functional modules is exemplified, and in actual application, the above functions can be completed by different functional modules according to needs, that is, the system is divided into different functional modules according to actual conditions to complete all or part of the above described functions. In addition, the system and method embodiments provided by the above embodiments belong to the same concept, and the specific implementation process is described in the method embodiments, which will not be repeated here.

[0086] Among them, the power transmission line path planning system 200 based on natural language and knowledge graph of the application can be a computer program (including program code) running in a computer device, for example, the power transmission line path planning system 200 based on natural language and knowledge graph of the application is an application software, which can be used to execute the corresponding steps in the power transmission line path planning method based on natural language and knowledge graph of the application.

[0087] In some embodiments, the power transmission line path planning system 200 based on natural language and knowledge graph of the application can be realized in a combination of software and hardware, for example, the power transmission line path planning system 200 based on natural language and knowledge graph of the application can be a processor in the form of a hardware decoding processor, which is programmed to execute the power transmission line path planning method based on natural language and knowledge graph of the application. For example, the processor in the form of a hardware decoding processor can use one or more application specific integrated circuits (ASICs), DSPs, programmable logic devices (PLDs), complex programmable logic devices (CPLDs), field programmable gate arrays (FPGAs) or other electronic components.

[0088] Among them, the modules involved in the embodiments of the application can be realized by software or hardware. Among them, the name of the module does not constitute a limitation of the module itself in some cases.

[0089] An electronic device according to an embodiment of the present application includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements any of the above power transmission line path planning methods based on natural language and a knowledge graph when executing the computer program, that is, the electronic device according to an embodiment of the present application can include, but is not limited to, a processor and a memory; the memory is configured to store a computer program; and the processor is configured to execute the power transmission line path planning method based on natural language and a knowledge graph shown in any of the embodiments of the present application by invoking the computer program.

[0090] An electronic device is provided in an optional embodiment, as shown in Figure 4 Figure 4 The electronic device 4000 shown in the optional embodiment includes a processor 4001 and a memory 4003. The processor 4001 and the memory 4003 are connected, for example, through a bus 4002. Optionally, the electronic device 4000 can further include a transceiver 4004, which can be used for data interaction, such as data transmission and / or data reception, between the electronic device and other electronic devices. It should be noted that the transceiver 4004 is not limited to one in actual application, and the structure of the electronic device 4000 does not constitute a limitation on the embodiments of the present application.

[0091] The processor 4001 can be a CPU (Central Processing Unit, central processor), a general-purpose processor, a DSP (Digital Signal Processor, data signal processor), an ASIC (Application Specific Integrated Circuit, application specific integrated circuit), an FPGA (Field Programmable Gate Array, field programmable gate array), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute various exemplary logical blocks, modules and circuits described in combination with the present disclosure. The processor 4001 can also be a combination of computing functions, such as one or more microprocessor combinations, combinations of DSP and microprocessor, etc.

[0092] The bus 4002 can include a path for transmitting information between the above components. The bus 4002 can be a PCI (Peripheral Component Interconnect, peripheral component interconnect) bus or an EISA (Extended Industry Standard Architecture, extended industry standard architecture) bus, etc. The bus 4002 can be divided into an address bus, a data bus, a control bus, etc. For the sake of expression, Figure 4 ​The bus 4002 is represented by only one thick line, but this does not mean that there is only one bus or one type of bus.

[0093] The memory 4003 may be ROM (Read Only Memory) or other types of static storage devices capable of storing static information and instructions, RAM (Random Access Memory) or other types of dynamic storage devices capable of storing information and instructions, or EEPROM (Electrically Erasable Programmable Read Only Memory), CD-ROM (Compact Disc Read Only Memory) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but not limited thereto.

[0094] The memory 4003 stores application code (computer program) for executing the present invention, and its execution is controlled by the processor 4001. The processor 4001 executes the application code stored in the memory 4003 to implement the content shown in the foregoing method embodiments.

[0095] Among them, electronic devices can also be terminal devices. A terminal device can be any terminal device that can install applications and access web pages through applications, including at least one of smartphones, tablets, laptops, desktop computers, smart speakers, smartwatches, smart TVs, and smart in-vehicle devices.

[0096] It should be noted that, Figure 4 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of use of the embodiments of the present invention.

[0097] An embodiment of the present invention provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements any of the above-mentioned transmission line path planning methods based on natural language and knowledge graphs.

[0098] Optionally, the computer readable storage medium can be a Read-Only Memory (ROM), a Random Access Memory (RAM), a Compact Disc Read-Only Memory (CD-ROM), a magnetic tape, a floppy disk, and an optical data storage device, etc.

[0099] In an example embodiment, a computer program product or computer program is also provided, which includes computer instructions stored in a computer readable storage medium. A processor of an electronic device reads the computer instructions from the computer readable storage medium, and the processor executes the computer instructions, so that the electronic device performs the above-mentioned power transmission line path planning method based on natural language and a knowledge graph.

[0100] Computer program code for carrying out operations of the present application can be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C++ or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code can execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection can be made to an external computer (for example, through the Internet using an Internet Service Provider).

[0101] It should be understood that the flowchart and block diagrams in the drawings illustrate the architecture, functionality, and operation of possible implementations of various embodiments of the present application. In this regard, each block in the flowchart and block diagrams can represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical functions. It should also be noted that in some alternative implementations, the functions noted in the blocks can occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently or the blocks can sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and / or flowchart illustrations, and combinations thereof, can be implemented by a dedicated hardware-based system that performs the specified functions or operations, or combinations of hardware and software.

[0102] The computer readable storage medium provided by the embodiments of the present application can be, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or apparatus, or any combination of the above. More specific examples of the computer readable storage medium can include, but are not limited to, an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, the computer readable storage medium can be any tangible medium that contains or stores a program that can be used by or in connection with an instruction execution system, apparatus or device.

[0103] The computer readable storage medium described above carries one or more programs, which, when executed by the electronic device, cause the electronic device to perform the method shown in the above embodiments.

[0104] The above description is merely the preferred embodiments of the present application and the explanation of the technical principles used. It should be understood by those skilled in the art that the disclosed scope of the present application is not limited to the technical solutions formed by the specific combinations of the above technical features, and should also cover other technical solutions formed by any combinations of the above technical features or their equivalent features without departing from the disclosed concept. For example, the technical solutions formed by the mutual replacement of the above features and the technical features disclosed in the present application (but not limited to) with similar functions.

[0105] It should be noted that the terms "first", "second" and the like in the specification and claims of the present application are used to distinguish similar objects, and represent no specific order or sequential order. The order of use of similar objects can be interchanged under appropriate circumstances, so that the embodiments of the present application described herein can be implemented in an order other than that illustrated or described.

[0106] Those skilled in the art will appreciate that the present application can be implemented as a system, method or computer program product. Therefore, the present application can be embodied in the form of an entirely hardware embodiment, an entirely software embodiment (including firmware, resident software, microcode, etc.), or a combination of hardware and software. Herein, the present application is generally referred to as "circuitry", "module" or "system". Furthermore, in some embodiments, the present application can also be implemented as a computer program product in one or more computer readable media having computer readable program code.

[0107] Although the embodiments of the present application have been shown and described above, it is understood that the above-described embodiments are exemplary and are not to be construed as limiting the present application, and that variations, modifications, substitutions and changes can be made by those skilled in the art without departing from the scope of the present application.

Claims

1. A transmission line path planning method based on natural language and knowledge graph, characterized in that, include: Receive the user's natural language path planning target for the target transmission line; The natural language path planning objective is analyzed using a large language model to obtain an intent classification result and at least one named entity; wherein, the at least one named entity includes at least one of: starting point, ending point, avoidance object, and preference object; The intent classification results and the at least one named entity are mapped to a business rule knowledge graph for reasoning to generate a task step sequence. The task step sequence is displayed to the user through a visual interface, and the user's confirmation or modification of each task step in the task step sequence is received through the visual interface to obtain each target task step. Using the large language model, the combination of multiple GIS tools and their corresponding execution order corresponding to each target task step is retrieved from the GIS operation knowledge graph, and the combination corresponding to each target task step is decomposed into a GIS operator call chain. By calling the corresponding GIS tools through the MCP protocol, the spatial analysis operations in the GIS operator call chain corresponding to each target task step are executed sequentially according to the order of all target task steps, thereby generating the path planning results of the target transmission line.

2. The transmission line path planning method based on natural language and knowledge graphs according to claim 1, characterized in that, The step of parsing the natural language path planning objective using a large language model to obtain the intent classification result and at least one named entity includes: The natural language path planning target is subjected to intent classification processing by the large language model, and the intent classification result of the natural language path planning target in the path planning task of the target transmission line is output. The natural language path planning target is processed by the large language model to identify at least one named entity; wherein, the at least one named entity includes at least one of the following: starting point, ending point, avoidance object, and preference object.

3. The transmission line path planning method based on natural language and knowledge graphs according to claim 1, characterized in that, The step of mapping the intent classification result and the at least one named entity to a business rule knowledge graph for reasoning and generating a task step sequence includes: The intent classification result and the at least one named entity are matched with nodes in the business rule knowledge graph to obtain matching nodes; Based on the logical relationships and reasoning rules represented by the edges defined in the business rule knowledge graph, reasoning is performed between the matching nodes to obtain the reasoning result; Based on the reasoning result, at least one task step is extracted from the business rule knowledge graph, and the task step sequence is generated according to the execution order of each task step.

4. The transmission line path planning method based on natural language and knowledge graphs according to claim 1, characterized in that, The steps of displaying the task step sequence to the user through a visual interface and receiving the user's confirmation or modification of each task step in the task step sequence through the visual interface to obtain each target task step include: The user is shown each task step in the task step sequence through the visual interface. For each task step in the task step sequence, receive a confirmation command or modification command issued by the user through the visual interface; When a modification instruction for any task step is received, the modification content for that task step is received through the visual interface; Based on the confirmation or modification instructions for each task step and the corresponding modification content, a target task step sequence with the same number of task steps as the task step sequence is formed.

5. The transmission line path planning method based on natural language and knowledge graphs according to claim 4, characterized in that, Using the large language model, the combination of multiple GIS tools and their corresponding execution order corresponding to any target task step is retrieved from the GIS operation knowledge graph. This combination is then broken down into steps of a GIS operator call chain, including: Using the large language model, multiple GIS tools corresponding to any target task step are retrieved from the GIS operation knowledge graph; Based on the execution order relationship between tools represented by the edges defined in the GIS operation knowledge graph, the execution order of multiple GIS tools corresponding to any target task step is determined; Based on the execution order of any target task step, the multiple GIS tools corresponding to any target task step are decomposed into an ordered GIS operator call chain.

6. The transmission line path planning method based on natural language and knowledge graphs according to claim 5, characterized in that, The step of calling the corresponding GIS tool via the MCP protocol, and sequentially executing the spatial analysis operations in the GIS operator call chain corresponding to each target task step according to the order of all target task steps, to generate the path planning result of the target transmission line, includes: Process each target task step sequentially according to the order of all target task steps; For the current target task step, the corresponding GIS tool is called through the MCP protocol. Combined with the spatial analysis results of the previous target task step, at least one spatial analysis operation is executed in the GIS operator call chain corresponding to the target task step, including slope analysis, cross detection, and buffer analysis. The spatial analysis results of the current target task step are generated and provided as input data to the spatial analysis operation corresponding to the next target task step. After all the target task steps are completed, the path planning result of the target transmission line is generated based on the final obtained spatial analysis results through a path search algorithm.

7. The transmission line path planning method based on natural language and knowledge graphs according to any one of claims 1 to 6, characterized in that, Also includes: The path planning results are stored in a database and then visualized.

8. A transmission line path planning system based on natural language and knowledge graph, characterized in that, include: The receiving module is used to receive the natural language path planning target input by the user for the target transmission line; The parsing module is used to parse the natural language path planning target using a large language model to obtain the intent classification result and at least one named entity; wherein, the at least one named entity includes at least one of: start point, end point, avoidance object, and preference object; The reasoning module is used to map the intent classification results and the at least one named entity to a business rule knowledge graph for reasoning, and generate a sequence of task steps; The adjustment module is used to display the task step sequence to the user through a visual interface, and to receive the user's confirmation or modification of each task step in the task step sequence through the visual interface, so as to obtain each target task step. The retrieval module is used to retrieve, from the GIS operation knowledge graph, the combination of multiple GIS tools and their corresponding execution order corresponding to each target task step using the large language model, and decompose the combination corresponding to each target task step into a GIS operator call chain. The execution module is used to call the corresponding GIS tools through the MCP protocol, and sequentially execute the spatial analysis operations in the GIS operator call chain corresponding to each target task step in the order of all target task steps, so as to generate the path planning results of the target transmission line.

9. An electronic device, characterized in that, The electronic device includes a processor coupled to a memory storing at least one computer program, which is loaded and executed by the processor to enable the electronic device to implement the power transmission line path planning method based on natural language and knowledge graph as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores at least one computer program, which, when executed by a processor, implements the transmission line path planning method based on natural language and knowledge graph as described in any one of claims 1 to 7.

Images