Power transmission line artificial intelligence path planning route selection method and system based on multi-source heterogeneous data fusion

Abstract

The application provides a power transmission line artificial intelligence path planning route selection method and system based on multi-source heterogeneous data fusion, and relates to the technical field of power transmission line planning.The application constructs a multi-level cost map system by acquiring basic geographic data, obstacle marking data, meteorological data, environmental data and power transmission line starting point and ending point information, including a basic layer, an obstacle layer, a meteorological layer and an environmental layer, and marks the cost coefficients of the corresponding influence factors in each layer to form a cost function.The improved A* algorithm is combined with the cost function to perform path planning and generate an obstacle avoidance path and an economic path.Through multi-source heterogeneous data fusion and intelligent algorithms such as the Transformer neural network architecture, the application can efficiently and accurately complete power transmission line path planning, significantly improve the scientificity, safety and economy of path planning, meet the needs of smart grid construction and promote the intelligent development of power grids.

Description

Technical Field

[0001] The present invention relates to the technical field of transmission line planning, and in particular to a method and system for transmission line artificial intelligence path planning and line selection based on multi-source heterogeneous data fusion. Background Art

[0002] In recent years, with rapid economic development and increasing energy demand, the scale and complexity of power grid construction have continued to grow. Grid construction is encompassing more and more complex regions, transmission line corridors are becoming increasingly congested, and route planning involves an increasing number of obstacles. Traditional transmission line route planning and selection methods typically rely on manual expert planning and experience. This approach has numerous shortcomings: First, manual planning is time-consuming and cannot meet the rapid demands of large-scale power grid construction; second, manual planning is not cost-effective and accurate, and due to limited experience, it is easy to miss some important geographical, meteorological, or environmental factors, resulting in inadequate route planning and potentially increasing construction costs and operational risks. Furthermore, traditional methods no longer meet the development requirements of smart grids under the new circumstances and cannot effectively cope with the complex and changing geographical environment and diverse influencing factors.

[0003] Modern transmission line route planning faces challenges from multiple complex factors. First, the increasing complexity of basic geographic data and the diversity of topography require that route planning must account for natural obstacles such as mountains, rivers, and lakes, as well as human-made obstacles such as cities and villages. Second, meteorological conditions are crucial to the safe operation of transmission lines. Meteorological zones such as ice zones, wind zones, and microclimate zones place higher demands on route planning. Furthermore, environmental factors must be fully considered in route planning to avoid ignoring their impact on line construction and operation and maintenance. Furthermore, with the advancement of data technology, the integration of multi-source heterogeneous data has become possible. How to effectively utilize this data for intelligent route planning has become a pressing issue.

[0004] While some existing methods attempt to utilize technologies like geographic information systems (GIS) to assist with route planning, these methods often only process a single type of data and fail to comprehensively consider the impact of heterogeneous data from multiple sources. Furthermore, existing methods also have shortcomings in their route planning algorithms, failing to effectively balance the economic and safety aspects of routes. Therefore, developing an AI-powered route planning and selection method for transmission lines that can integrate heterogeneous data from multiple sources and utilize advanced algorithms for efficient and accurate route planning is crucial for improving the efficiency and quality of power grid construction. Summary of the Invention

[0005] The present invention aims to solve at least one of the technical problems existing in the prior art.

[0006] To this end, the first aspect of the present invention provides a transmission line artificial intelligence path planning and line selection method based on multi-source heterogeneous data fusion.

[0007] The second aspect of the present invention provides a transmission line artificial intelligence path planning and line selection system that integrates multi-source heterogeneous data.

[0008] The present invention proposes an artificial intelligence path planning and line selection method for power transmission lines based on multi-source heterogeneous data fusion, which includes:

[0009] Acquire multi-source heterogeneous data related to transmission line path planning, wherein the multi-source heterogeneous data includes basic geographic data, obstacle marking data, meteorological data, environmental data, and the starting point and end point of the target planned transmission line;

[0010] Identifying and grading obstacles according to the obstacle marking data;

[0011] Constructing a multi-level cost map system based on the multi-source heterogeneous data; the multi-level cost map system includes multiple layers of cost maps representing different influencing factors, the multiple layers of the cost map including a base layer, an obstacle layer, a meteorological layer, and an environmental layer; marking the cost coefficient of each node corresponding to the influencing factor in each layer of the cost map;

[0012] Based on the multi-level cost map system, an improved A* algorithm is used to plan the path of the target planned transmission line; wherein, a cost calculation function is constructed based on the cost coefficient of each node of the corresponding influencing factor marked in each layer of the cost map, and a combined cost function in the improved A* algorithm is defined according to the cost calculation function.

[0013] The method for power transmission line artificial intelligence path planning and selection based on multi-source heterogeneous data fusion according to the above technical solution of the present invention may also have the following additional technical features:

[0014] In the above technical solution, in the multi-level cost map system, the planning area including the starting point and end point of the target planned transmission line is divided into several grids, each grid is defined as a node, and the cost coefficient of the node is marked in the cost map of the corresponding level according to the basic geographic data, obstacle marking data, meteorological data, and environmental data in the area corresponding to the node.

[0015] In the above technical solution, in the base layer, the air line deviation cost of each node in the planning area is marked, and the marking method includes:

[0016] Calculate the vertical distance from the node to the aerial line; the direct line connecting the starting point and the end point of the target planned transmission line is defined as the aerial line, and the direction of the aerial line is the direction from the starting point to the end point;

[0017] Calculate the angle between the node and the direction of the air line;

[0018] The deviation cost of the node from the air line is calculated based on the vertical distance from the node to the air line and the angle between the node and the air line direction. The calculation method is:

[0019]

[0020] in, represents the air line deviation cost of node i; represents the vertical distance from node i to the aviation line; represents the distance weight coefficient; represents the angle weight coefficient; Represents the angle between node i and the direction of the air line.

[0021] In the above technical solution, in the obstacle layer, the obstacle cost of the node is assigned according to the obstacle level within the node range;

[0022] Among them, for nodes located in an impassable obstacle, the obstacle cost is set to infinity; for nodes located in an obstacle-free area, the obstacle cost is set to 0.

[0023] In the above technical solution, in the meteorological layer, the meteorological cost of the node is assigned according to the meteorological area type within the node range; the meteorological area type includes ice area, wind area and micro-meteorological area;

[0024] Alternatively, in the meteorological layer, the meteorological cost of the node is assigned according to the historical average wind speed and the historical average ice thickness within the node range.

[0025] In the above technical solution, in the environmental layer, the environmental cost of the node is assigned according to the landform type within the node range.

[0026] In the above technical solution, the cost calculation function is constructed based on the cost coefficient of each node corresponding to the impact factor marked in each layer of the cost map, including:

[0027]

[0028] in, represents the cost calculation function of node i; Indicates the cost weight of air route deviation; represents the air line deviation cost of node i; represents the obstacle cost weight; represents the obstacle cost of node i; represents the meteorological cost weight; represents the meteorological cost of node i; represents the weight of environmental cost; represents the environmental cost of node i.

[0029] In the above technical solution, the cost calculation function is constructed based on the cost coefficient of each node corresponding to the impact factor marked in each layer of the cost map, including:

[0030]

[0031] in, represents the cost calculation function of node i; Indicates the cost weight of air route deviation; represents the air line deviation cost of node i; represents the tortuosity weight; The tortuosity coefficient of node i is used to indicate the tortuosity of the current path; represents the obstacle cost weight; represents the obstacle cost of node i; represents the meteorological cost weight; represents the meteorological cost of node i; represents the weight of environmental cost; represents the environmental cost of node i.

[0032] In the above technical solution, the identification and classification of obstacles includes:

[0033] Use data annotation tools to mark various obstacles;

[0034] Use the Transformer deep learning architecture to intelligently perceive and identify obstacles.

[0035] In the above technical solution, the path planning of the target planned transmission line using the A* algorithm based on the multi-level cost map system includes:

[0036] According to the set constraints, the improved A* algorithm is used to plan the path of the target transmission line to obtain the obstacle avoidance path and the economic path;

[0037] Among them, in the improved A* algorithm, the node whose cost of avoiding all obstacles is greater than the set obstacle cost is configured to obtain the obstacle avoidance path;

[0038] Among the several path results output by the improved A* algorithm, the path with the minimum cumulative cost is selected as the economic path.

[0039] The present invention provides a multi-source heterogeneous data fusion transmission line artificial intelligence path planning and line selection system, which is applied to the multi-source heterogeneous data fusion transmission line artificial intelligence path planning and line selection method as described in any of the above technical solutions. The system includes:

[0040] Data input module, used to obtain multi-source heterogeneous data related to transmission line path planning;

[0041] A data processing module is used to perform data processing, including pre-training models, obstacle recognition processing, point cloud processing, and terrain parameter calculation;

[0042] Costmap construction module, used to build a multi-level costmap system;

[0043] Intelligent path planning module, used for intelligent path planning using improved A* algorithm;

[0044] The results output module is used to output the results and display them visually.

[0045] In summary, due to the adoption of the above technical features, the beneficial effects of the present invention are:

[0046] First, this invention integrates multi-source heterogeneous data for transmission line routing, comprehensively considering multiple factors such as basic geographic data, obstacle identification data, meteorological data, and environmental data. Compared with traditional planning methods that rely on manual experience, this method significantly improves the comprehensiveness and accuracy of routing. It effectively avoids omissions and errors caused by human factors, ensuring more scientific and reasonable routing for transmission lines, thereby reducing construction risks and subsequent operation and maintenance costs.

[0047] The present invention adopts a multi-level cost map system to construct a path planning model, mapping different influencing factors to different cost map layers, such as the base layer, obstacle layer, meteorological layer and environmental layer, and marking the cost coefficient of the corresponding influencing factor at each node in each layer of the cost map. This layered processing method enables the influence of each factor to be clearly quantified and distinguished, which facilitates flexible adjustment and optimization according to the importance of different factors during the path planning process. At the same time, the improved A* algorithm is used for path planning based on the multi-level cost map system, which can efficiently search for the optimal path, greatly improve the efficiency of path planning, shorten the planning time, and meet the needs of the rapid development of modern power grid construction.

[0048] This invention utilizes data annotation tools to label various obstacles and incorporates deep learning architectures such as Transformers to intelligently perceive and identify obstacles. This not only improves the accuracy and reliability of obstacle identification but also provides more precise obstacle information for subsequent path planning, further enhancing the safety and cost-effectiveness of path planning.

[0049] Furthermore, during the path planning process, the present invention not only determines an obstacle-avoiding path based on set constraints, but also selects the path with the lowest cumulative cost as the economic path from among the multiple paths output by the improved A* algorithm. This path planning approach, which balances obstacle avoidance and economic efficiency, can minimize construction costs while ensuring the safe operation of power transmission lines, achieving an organic balance between economic and safety benefits.

[0050] Finally, the implementation of the present invention can effectively promote the development of smart grids, provide an efficient, intelligent and scientific path planning and line selection method for grid construction, help improve the overall performance and operating efficiency of the grid, and is of great significance for ensuring the stability and reliability of power supply.

[0051] Additional aspects and advantages of the invention will become apparent from the description which follows, or may be learned by practice of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the following description of the embodiments with reference to the accompanying drawings, in which:

[0053] Fig. 1 This is a flow chart of a method for artificial intelligence path planning and line selection of a power transmission line using multi-source heterogeneous data fusion according to an embodiment of the present invention;

[0054] Fig. 2 It is an execution block diagram of a multi-source heterogeneous data fusion transmission line artificial intelligence path planning and line selection system according to an embodiment of the present invention. DETAILED DESCRIPTION

[0055] In order to more clearly understand the above-mentioned objects, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that, in the absence of conflict, the embodiments of the present application and the features therein can be combined with each other.

[0056] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the scope of protection of the present invention is not limited to the specific embodiments disclosed below.

[0057] Refer to the following Figs. 1-2 The following describes a method for artificial intelligence path planning and line selection of a power transmission line based on multi-source heterogeneous data fusion provided in accordance with some embodiments of the present invention.

[0058] Some embodiments of the present application provide a method for artificial intelligence path planning and line selection of power transmission lines by fusion of multi-source heterogeneous data.

[0059] like Fig. 1As shown, the first embodiment of the present invention proposes an artificial intelligence path planning and line selection method for power transmission lines based on multi-source heterogeneous data fusion, including the following steps S1 to S4.

[0060] S1. Acquire multi-source heterogeneous data related to transmission line path planning, wherein the multi-source heterogeneous data includes basic geographic data, obstacle marking data, meteorological data, environmental data, and the starting point and end point of the target planned transmission line.

[0061] In some embodiments, the basic geographic data can be acquired through satellites, drone aerial photography, aerial surveys, radar, airborne LiDAR (Light Detection and Ranging), and other methods. Basic geographic data can include high-definition satellite aerial images, digital elevation models (DEMs), and laser point cloud data, used to display the geospatial information of the planning area. Meteorological data primarily includes ice cover, wind speed, and weather type throughout the planning area, and can be acquired through ice and wind zone distribution maps and on-site surveys. Environmental data primarily includes the landform type of the planning area, such as flat land, hills, and deserts. Obstacle marking data primarily includes the location, size, and importance of obstacles within the area. It should be understood that this obstacle marking data only provides basic information about the obstacles; subsequent analysis of their potential impact on the transmission line and the extent of that impact will be required based on their characteristic information. Furthermore, the obstacle marking data may also include data for classifying each type of obstacle, such as the criteria for classifying obstacles.

[0062] In addition, the present invention plans the transmission line with a determined starting point and end point.

[0063] S2. Identify and classify obstacles according to the obstacle marking data.

[0064] Step S2 can be completed manually or automatically based on input data by training a neural network model.

[0065] In some embodiments, identified obstacles can be annotated using data annotation tools. Deep learning architectures such as Transformers can also be used to intelligently perceive and identify obstacles. Specifically, an obstacle recognition model suitable for this system is trained based on image type (high-resolution aerial images / satellite images (0.5-2m resolution)), annotation category (based on obstacle type), sample size, and annotation specifications. It should be noted that existing obstacle recognition models can also be directly used, as long as they can distinguish the impact of different obstacles. Furthermore, obstacles that have not been previously labeled but may affect transmission lines can also be identified using this model.

[0066] S3. Construct a multi-level cost map system based on the multi-source heterogeneous data; the multi-level cost map system includes multiple layers of cost maps representing different influencing factors, and the multiple layers of the cost map include a base layer, an obstacle layer, a meteorological layer, and an environmental layer; mark the cost coefficient of the corresponding influencing factor at each node in each layer of the cost map.

[0067] Specifically, in the multi-level costmap system, several costmaps of the same size as the map are created. The planned area, including the starting and ending points of the target planned transmission line, is divided into several grids, with each grid defined as a node. In other words, in this disclosure, the transmission line extends from the starting point to the end point in increments of one node. The cost coefficients of the nodes are marked in the costmap at the corresponding level based on the basic geographic data, obstacle marking data, meteorological data, and environmental data within the corresponding area of ​​the node.

[0068] In some embodiments, in the base layer, the flight line deviation cost of each node in the planning area is marked, and the marking method includes:

[0069] Calculate the vertical distance from the node to the aerial line; the direct line connecting the starting point and the end point of the target planned transmission line is defined as the aerial line, and the direction of the aerial line is the direction from the starting point to the end point;

[0070] Calculate the angle between the node and the direction of the air line;

[0071] The deviation cost of the node from the air line is calculated based on the vertical distance from the node to the air line and the angle between the node and the air line direction. The calculation method is:

[0072]

[0073] in, represents the air line deviation cost of node i; represents the vertical distance from node i to the aviation line; represents the distance weight coefficient; represents the angle weight coefficient; Represents the angle between node i and the direction of the air line.

[0074] In some embodiments, in the obstacle layer, an obstacle cost is assigned to a node based on the level of obstacles within the node's range. Obstacles are generally categorized as significant and general obstacles. For nodes located near impassable obstacles, i.e., significant obstacles, the obstacle cost is set to infinity or a very large number (much larger than the obstacle cost for general obstacles). For nodes located in obstacle-free areas, the obstacle cost is set to 0. For example, significant obstacles such as the core area of ​​a nature reserve, a primary water source protection area, and an urban planning area are assigned an obstacle cost of 1000.

[0075] In a specific embodiment, for general obstacles, the obstacle cost is set to a constant greater than 0 and less than 10 based on its obstacle level. For example, the value range for experimental areas or general control areas of nature reserves is 6-10, the value range for second- and third-level water source protection areas is 2-5, the value range for forest parks is 2-6, the value range for scenic spots is 1-5, and the value range for rural tourism planning areas is 1-3.

[0076] In some embodiments, in the meteorological layer, the meteorological cost of the node is assigned according to the meteorological area type within the node range; the meteorological area types include ice areas, wind areas and micro-meteorological areas; similarly, according to the severity of the weather, the meteorological cost of the node is set to a constant greater than 0 and less than 10; for example, the meteorological cost range of the node in the ice area is 2-8, the meteorological cost range of the node in the wind area is 1-6, and the meteorological cost range of the node in the micro-meteorological area is 1-3.

[0077] In other embodiments, in the meteorological layer, the meteorological cost of a node is assigned based on the historical average wind speed and the historical average ice thickness within the node range. For example, for nodes with a historical average ice thickness of 5mm, 10mm, 15mm, 20mm, 25mm, 30mm, 40mm, 50mm, 60mm, 70mm, and 80mm, the meteorological cost range is set to 1-8, with the greater the ice thickness, the greater the meteorological cost. For nodes with a historical average wind speed of 25m / s, 26m / s, 27m / s, 28m / s, 29m / s, 30m / s, 31m / s, 32m / s, 33m / s, 34m / s, 35m / s, 36m / s, 37m / s, and 38m / s, the meteorological cost range is set to 1-1.26, with the greater the wind speed, the greater the meteorological cost. For areas affected by both ice cover and wind, the two can be summed or weighted to form the meteorological cost of crossing the node.

[0078] In some embodiments, in the environmental layer, the environmental cost of the node is assigned according to the landform type within the node range; because in different landforms, different slopes will affect the construction difficulty and tower foundation stability, different altitudes affect material transportation and operation and maintenance costs, the terrain undulation reflects the complexity of the surface, and the slope affects wind load and sunshine; based on the above differences, the environmental cost can also be assigned to a constant greater than 0 and less than 10. For example: the environmental cost of flat land can be set to 1-2, the environmental cost of hills can be set to 2-3, the environmental cost of mountains can be set to 4-5, the environmental cost of high mountains can be set to 6-8, the environmental cost of steep mountains can be set to 8-9, the environmental cost of deserts can be set to 3-5, the environmental cost of swamps can be set to 4-6, and the environmental cost of river networks can be set to 4-6. The above assignment process can be used to quantify the degree of influence of different landforms on the installation of transmission lines.

[0079] It is understandable that in the above-mentioned multi-level cost map system, the assignment process can be completed manually, or it can be completed automatically by the machine and then manually reviewed. The specific assignment method is not limited.

[0080] S4. Path planning for the target planned transmission line is performed using the improved A* algorithm based on the multi-level cost map system. A cost calculation function is constructed based on the cost coefficients at each node corresponding to the impact factors marked in each level of the cost map, and a combined cost function in the improved A* algorithm is defined based on the cost calculation function. Specifically, the cost calculation function can be used to calculate the actual cost from the starting point to the current node in the improved A* algorithm, and can also be used as a heuristic function to estimate the cost from the current node to the end point. In this disclosure, the improvement to the improved A* algorithm primarily resides in integrating the multi-level cost map system into the combined cost function of the improved A* algorithm. The specific execution process of the improved A* algorithm can continue to follow conventional methods. In one specific embodiment, the cost calculation function is used to calculate the actual cost from the starting point to the current node in the improved A* algorithm, and the heuristic function in the improved A* algorithm can be selected as Euclidean distance or Manhattan distance, for example. In another specific embodiment, the cost calculation function is used as a heuristic function in the improved A* algorithm to estimate the cost from the current node to the end point, and the actual cost from the starting point to the current node in the improved A* algorithm can continue to be calculated using conventional methods.

[0081] In some embodiments, the cost calculation function is constructed based on the cost coefficient of each node corresponding to the impact factor marked in each layer of the cost map, including:

[0082]

[0083] in, represents the cost calculation function of node i; Indicates the cost weight of the air line deviation, which is used to adjust the weight of the air line deviation. , which means we allow a certain deviation, but do not want the deviation to be too large; represents the air line deviation cost of node i; represents the obstacle cost weight; represents the obstacle cost of node i; represents the meteorological cost weight; represents the meteorological cost of node i; represents the weight of environmental cost; represents the environmental cost of node i.

[0084] In some other embodiments, a tortuosity coefficient is introduced into the aerial deviation cost of the base layer to control the route length. The cost calculation function is constructed based on the cost coefficient of each node corresponding to the impact factor marked in each layer of the cost map, including:

[0085]

[0086] in, represents the cost calculation function of node i; Indicates the cost weight of air route deviation; represents the air line deviation cost of node i; represents the tortuosity weight; The tortuosity coefficient of node i is used to indicate the tortuosity of the current path. The specific value is the ratio of the actual length of the current path to the aviation distance of the current path. represents the obstacle cost weight; represents the obstacle cost of node i; represents the meteorological cost weight; represents the meteorological cost of node i; represents the weight of environmental cost; represents the environmental cost of node i.

[0087] In some embodiments, the step S4 of performing path planning for the target planned transmission line based on the multi-level costmap system using an improved A* algorithm includes:

[0088] According to the set constraints, the improved A* algorithm is used to plan the path of the target transmission line to obtain the obstacle avoidance path and the economic path;

[0089] Among them, in the improved A* algorithm, nodes with a cost of avoiding all obstacles greater than the set obstacle cost are configured to obtain an obstacle avoidance path; among the several path results output by the improved A* algorithm, the path with the smallest cumulative cost is selected as the economic path.

[0090] It can be understood that the obstacle avoidance path is the planned path that meets the most reliable requirement (completely avoiding the specified obstacles), and the economic path is the planned path that meets the most economical requirement (weighing the cost).

[0091] Other embodiments of the present invention provide a multi-source heterogeneous data fusion transmission line artificial intelligence path planning and line selection system, which is applied to the multi-source heterogeneous data fusion transmission line artificial intelligence path planning and line selection method as described in any of the above embodiments. The execution process of the system is as follows: Fig. 2 As shown; the system includes:

[0092] Data input module, used to obtain multi-source heterogeneous data related to transmission line path planning;

[0093] A data processing module is used to perform data processing, including pre-training models, obstacle recognition processing, point cloud processing, and terrain parameter calculation;

[0094] Costmap construction module, used to build a multi-level costmap system;

[0095] Intelligent path planning module, used for intelligent path planning using improved A* algorithm;

[0096] The output module is used to output the results and provide a visual display. For example, it can output path planning diagrams, 3D visualization results, engineering reports, and drawings.

[0097] In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any suitable manner in any one or more embodiments or examples.

[0098] Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A transmission line artificial intelligence path planning and line selection method based on multi-source heterogeneous data fusion, characterized by: include: Acquire multi-source heterogeneous data related to transmission line path planning, wherein the multi-source heterogeneous data includes basic geographic data, obstacle marking data, meteorological data, environmental data, and the starting point and end point of the target planned transmission line; Identifying and grading obstacles according to the obstacle marking data; Constructing a multi-level cost map system based on the multi-source heterogeneous data; The multi-level cost map system includes multi-layer cost maps representing different influencing factors, and the multi-layer cost maps include a base layer, an obstacle layer, a weather layer, and an environment layer; the cost coefficient of the corresponding influencing factor at each node is marked in each layer of the cost map; Based on the multi-level cost map system, an improved A* algorithm is used to perform path planning for the target planned transmission line; wherein a cost calculation function is constructed based on the cost coefficient of each node corresponding to the impact factor marked in each layer of the cost map, and a combined cost function in the improved A* algorithm is defined based on the cost calculation function; In the multi-level cost map system, the planning area including the starting point and end point of the target planned transmission line is divided into several grids, each grid is defined as a node, and the cost coefficient of the node is marked in the cost map of the corresponding level according to the basic geographic data, obstacle marking data, meteorological data, and environmental data in the corresponding area of ​​the node; In the base layer, the flight line deviation cost of each node in the planning area is marked, and the marking method includes: Calculate the vertical distance from the node to the aerial line; the direct line connecting the starting point and the end point of the target planned transmission line is defined as the aerial line, and the direction of the aerial line is the direction from the starting point to the end point; Calculate the angle between the node and the direction of the air line; The deviation cost of the node from the air line is calculated based on the vertical distance from the node to the air line and the angle between the node and the air line direction. The calculation method is: in, represents the air line deviation cost of node i; represents the vertical distance from node i to the aviation line; represents the distance weight coefficient; represents the angle weight coefficient; represents the angle between node i and the direction of the air line; In the obstacle layer, the obstacle cost of the node is assigned according to the obstacle level within the node range; Among them, for nodes located in an impassable obstacle, the obstacle cost is set to infinity; for nodes located in an obstacle-free area, the obstacle cost is set to 0.

2. The method for power transmission line artificial intelligence path planning and line selection based on multi-source heterogeneous data fusion according to claim 1 is characterized in that: In the meteorological layer, the meteorological cost of the node is assigned according to the meteorological region type within the node range; the meteorological region type includes ice region, wind region and micro-meteorological region; Alternatively, in the meteorological layer, the meteorological cost of the node is assigned according to the historical average wind speed and the historical average ice thickness within the node range.

3. The method for power transmission line artificial intelligence path planning and selection based on multi-source heterogeneous data fusion according to claim 2 is characterized in that: In the environmental layer, the environmental cost of a node is assigned according to the landform type within the node range.

4. The method for power transmission line artificial intelligence path planning and line selection based on multi-source heterogeneous data fusion according to claim 1 is characterized in that: The cost calculation function is constructed based on the cost coefficient of each node of the corresponding impact factor marked in each layer of the cost map, including: in, represents the cost calculation function of node i; Indicates the cost weight of air route deviation; represents the air line deviation cost of node i; represents the obstacle cost weight; represents the obstacle cost of node i; represents the meteorological cost weight; represents the meteorological cost of node i; represents the weight of environmental cost; represents the environmental cost of node i.

5. The method for power transmission line artificial intelligence path planning and selection based on multi-source heterogeneous data fusion according to claim 1 is characterized in that: The cost calculation function is constructed based on the cost coefficient of each node of the corresponding impact factor marked in each layer of the cost map, including: in, represents the cost calculation function of node i; Indicates the cost weight of air route deviation; represents the air line deviation cost of node i; represents the tortuosity weight; The tortuosity coefficient of node i is used to indicate the tortuosity of the current path; represents the obstacle cost weight; represents the obstacle cost of node i; represents the meteorological cost weight; represents the meteorological cost of node i; represents the weight of environmental cost; represents the environmental cost of node i.

6. The method for power transmission line artificial intelligence path planning and selection based on multi-source heterogeneous data fusion according to claim 1 is characterized in that: The identification and classification of obstacles includes: Use data annotation tools to mark various obstacles; Use the Transformer deep learning architecture to intelligently perceive and identify obstacles.

7. The method for power transmission line artificial intelligence path planning and selection based on multi-source heterogeneous data fusion according to claim 1 is characterized in that: The method of performing path planning for a target planned transmission line using an improved A* algorithm based on the multi-level cost map system includes: According to the set constraints, the improved A* algorithm is used to plan the path of the target transmission line to obtain the obstacle avoidance path and the economic path; Among them, in the improved A* algorithm, the node whose cost of avoiding all obstacles is greater than the set obstacle cost is configured to obtain the obstacle avoidance path; Among the several path results output by the improved A* algorithm, the path with the minimum cumulative cost is selected as the economic path.

8. A multi-source heterogeneous data fusion power transmission line artificial intelligence path planning and line selection system, characterized by: A method for artificial intelligence path planning and line selection of a power transmission line based on multi-source heterogeneous data fusion according to any one of claims 1 to 7, wherein the system comprises: Data input module, used to obtain multi-source heterogeneous data related to transmission line path planning; A data processing module is used to perform data processing, including pre-training models, obstacle recognition processing, point cloud processing, and terrain parameter calculation; Costmap construction module, used to build a multi-level costmap system; Intelligent path planning module, used for intelligent path planning using improved A* algorithm; The results output module is used to output the results and display them visually.

Images