Power line inspection method of UAV swarm based on transmission line topology

By using drone cluster patrol methods on transmission lines, topological structure, clustering and genetic algorithms are used to optimize airport location and patrol paths, combined with catenary and expansion algorithm strategies, traditional patrol efficiency and safety hazards are solved. The problem has achieved efficient and accurate power line detection.

CN118763558BActive Publication Date: 2024-11-05NORTH CHINA UNIVERSITY OF SCIENCE AND TECHNOLOGY
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411253169.8
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-09-09
Publication Date
2024-11-05
Estimated Expiration
2044-09-09

AI Technical Summary

Technical Problem

Traditional transmission line inspection requires a lot of manpower, which poses safety hazards, has low detection accuracy, making it difficult to detect hidden problems, and the inspection cost continues to increase as the line length increases.

Method used

The drone cluster inspection method based on the transmission line topology structure is used to determine the optimal number and location of the drone airports using improved fitness function, and the patrol path is planned in combination with clustering and genetic algorithms, and catenary equations and expansion algorithms are used to surround it. Path strategy performs power line detection.

Benefits of technology

It significantly improves the efficiency and quality of inspections, ensures the comprehensiveness and accuracy of inspections, and is especially suitable for large-scale or complex power transmission lines, reducing the burden on a single drone and shortening the inspection time.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118763558B_ABST
    Figure CN118763558B_ABST
Patent Text Reader

Abstract

The present invention discloses a method for power line inspection of a drone group based on the topological structure of a power transmission line, which belongs to the field of power transmission line inspection, and includes the following steps: S1, obtaining power line GIS data, and establishing the topological structure of the power transmission line; S2, according to the topological structure, combining a clustering method and a genetic algorithm, determining the number and location of the best drone airports; S3, planning the drone inspection path according to the geographical location information in the topological structure and the number and location of the determined best drone airports; S4, configuring a drone for each drone airport, making all configured drones fly and inspect along the planned inspection path, and sending the captured data to a remote data center, completing the inspection process including the power line inspection sub-path and the elevated inspection sub-path. The present invention realizes the comprehensiveness and accuracy of the inspection, significantly improves the inspection efficiency, and is particularly suitable for the inspection tasks of large-scale or complex power transmission lines.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the field of power transmission line inspection, and in particular relates to a power line inspection method using a swarm of unmanned aerial vehicles based on a power transmission line topology structure. Background Art

[0002] After the power transmission lines are put into use, problems may occur due to various situations, such as equipment wear, insulation aging, and tower deformation, and these problems are often difficult to detect and deal with in time. Traditional transmission line detection generally uses manual detection, and this traditional transmission line inspection requires a lot of manpower and material resources, has high requirements on personnel quality, and has low inspection efficiency. Due to the scattered geographical location and long lines of transmission lines, it takes a lot of time and energy to inspect; and traditional transmission line inspections require personnel to climb high altitudes and cross rugged mountain roads and other harsh environments, which poses certain safety hazards and is prone to accidents, causing casualties and equipment damage; and traditional transmission line inspections mainly rely on human eye observation, with low detection accuracy, prone to missed detection, false detection, etc., and it is difficult to detect some hidden problems. At the same time, as the length of the line increases, the inspection cost will continue to increase. Therefore, the power line detection method using drones has become a research direction. Summary of the invention

[0003] In order to address the deficiencies in the prior art, the present invention provides a power line inspection method using a drone swarm based on the topological structure of a transmission line. The improved fitness function is used to determine the optimal number and location of drone airports, and multiple drones are allocated accordingly to simultaneously inspect the power lines. The catenary equation is used to detect power lines, and the expansion algorithm is used to use a surrounding path strategy to detect elevated roads. This not only speeds up the inspection process, but also effectively ensures the comprehensiveness and accuracy of the inspection. It is particularly suitable for inspection tasks of large-scale or complex power transmission lines, and significantly improves the inspection efficiency and quality.

[0004] In order to achieve the above-mentioned object, a power line inspection method of a drone swarm based on a power transmission line topology structure of the present invention comprises the following steps:

[0005] S1. Obtain power line GIS data and establish the topological structure of power transmission lines;

[0006] S2. Determine the optimal number and location of drone airports based on the topological structure of power transmission lines, combined with clustering method and genetic algorithm; specifically includes the following sub-steps:

[0007] S21, obtaining the longitude and latitude data of each node in the topological structure;

[0008] S22, clustering the acquired longitude and latitude data using a clustering method to obtain the center point of each cluster as the number and location of potential airports for drones;

[0009] S23, using a genetic algorithm to optimize the number and location of potential airports for drones, and selecting the chromosome with the best fitness from the results of the genetic algorithm to determine the number and location of the optimal drone airports;

[0010] S3, planning the drone inspection route based on the geographic location information in the topological structure and the number and location of the optimal drone airports determined;

[0011] S4. Configure drones for the drone airport, and make all configured drones fly and inspect along the planned inspection path, and send the captured data to the remote data center to complete the inspection process; the inspection path includes a power line inspection sub-path, and the power line inspection sub-path includes: using drone cameras to perform real-time image recognition and tracking, automatically tracking power lines, real-time capturing and analyzing the current state information of power lines, matching the captured current state information of power lines with the acquired power line catenary model, and when it is found that the matching degree is low, the power line is abnormally detached, or the power line is hung with foreign objects, the drone records the log in time and automatically reports the error; wherein the acquisition process of the power line catenary model includes:

[0012] Get the coordinates and horizontal distances of the two end points of the power line in the topological structure; calculate the catenary section equation of the power line according to the coordinates and horizontal distances of the two end points; calculate the catenary space curve equation of the power line according to the obtained catenary section equation; calculate multiple catenary space plane equations according to the coordinates and horizontal distances of the two end points of all power lines between two points:

[0013] y=acosh(x / a)

[0014] Among them, y is the vertical position of the point on the catenary, x is the horizontal position of the point on the catenary, and a is the parameter of the catenary, which affects the shape of the curve.

[0015] Preferably, step S1 comprises:

[0016] Obtain power line GIS data, use high-voltage transmission towers and power generating stations as nodes, and transmission lines as edges to build the topological structure of power transmission lines.

[0017] Preferably, the fitness in step S23 is determined by an improved fitness function, which is set to determine the function value based on the total distance from all data points to the determined airport and the total distance from the specific location point to the selected airport, with the purpose of minimizing the weighted sum of the objectives.

[0018] Preferably, the fitness in step S23 is determined by an improved fitness function, wherein the improved fitness function includes a penalty mechanism.

[0019] Preferably, in step S3, according to the geographical location information in the topological structure and the number and location of the determined optimal drone airports, the drone inspection path is planned, including:

[0020] S31. Visualize the distribution of airport locations and power transmission lines based on the number and location of the determined optimal drone airports and the geographic location information in the topological structure;

[0021] S32, setting each inspection path to have the determined optimal drone airport location as the starting point and end point of the drone inspection path;

[0022] S33, the power stations in the topological structure between two adjacent drone airports and the interconnected transmission lines and high-voltage transmission towers between the power stations form a drone inspection path, and the inspection paths of the drones do not overlap.

[0023] Preferably, in step S4, the inspection path also includes an elevated inspection sub-path, and the elevated inspection sub-path includes: locating and identifying key points of the elevated structure, using the elevated structure expansion algorithm to calculate the optimal spatial surrounding path for the UAV, using the sensors on the UAV to obtain the distance information between the UAV and the elevated structure in real time, combining the acquired distance information with the known elevated structure model, and making the UAV fly and identify according to the spatial surrounding path of the elevated structure expansion algorithm, ensuring a detailed scan of each key point of the elevated structure from all angles, and timely recording logs and automatically reporting errors.

[0024] Preferably, the overhead structure expansion algorithm surrounding path strategy specifically includes:

[0025] The key points of the elevated structure are determined based on the data from the depth camera and the model of the elevated transmission tower. An expansion area is generated for each key point by adding a virtual three-dimensional sphere around the key point. The A* search algorithm and genetic optimization algorithm are used to generate an optimal spatial path for the drone to surround the expansion area.

[0026] Preferably, the inspection distance between each pair of airports is the sum of the total distance of the power lines between the two airports plus the length of the expansion algorithm space surrounding inspection path of all overhead lines on the power lines.

[0027] The beneficial effects of the present invention are:

[0028] 1. The present invention improves the design efficiency of the UAV airport location by optimizing the fitness function. The fitness function combines BIC (Bayesian Information Criterion) and K-Means clustering algorithm, and can comprehensively consider the total distance from the data point to the airport and the distance from the specific location point to the airport according to the actual application scenario (such as UAV airport location optimization). By combining the advantages of both in a hybrid algorithm, the selection of airport location and quantity is more accurate, which significantly improves the inspection efficiency and reduces the flight cost of the UAV.

[0029] 2. In the process of power line inspection, the present invention adopts a method of real-time tracking of the side of the catenary by a drone. The drone collects real-time image information of the power line through a camera and matches it with the pre-established catenary equation. When it is detected that the power line is detached, a foreign object is hanging, or it does not match the catenary model, the drone will immediately generate a log and issue an early warning, thereby greatly improving the accuracy and timeliness of the inspection and avoiding the problems of missed detection and false detection.

[0030] 3. The present invention adopts an expansion algorithm and a surrounding path strategy in the inspection of elevated structures to ensure a comprehensive inspection of each key structural point. This method is based on the expansion algorithm in image processing, and by generating a virtual three-dimensional expansion area, the drone can capture detailed information of key points from multiple angles. This strategy can ensure that each key point of the elevated structure is accurately covered, enhance the comprehensiveness and accuracy of the inspection, and is particularly suitable for complex elevated line inspections.

[0031] 4. The present invention realizes fast and efficient inspection of power lines through the collaborative work of multiple drones. Each drone is assigned to a specific area, and uses advanced camera technology and image recognition algorithms to capture and analyze the status of power lines in real time. By clustering the number and location of drones and optimizing the planning of genetic algorithms, the inspection task achieves the highest efficiency with the least resource investment. This method not only significantly reduces the burden of a single drone, but also greatly shortens the inspection time. It is particularly suitable for large-scale or geographically complex transmission lines, ensuring the comprehensiveness and accuracy of the inspection. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 It is a flow chart of an embodiment of a method for power line inspection of a drone swarm based on a power line topology structure of the present invention;

[0033] Figure 2 This is a schematic diagram of the change of various data of fitness with the number of iterations according to an embodiment of the present invention;

[0034] Figure 3 is a schematic diagram of the evolution of fitness with the number of iterations according to an embodiment of the present invention;

[0035] Figure 4 is an airport location optimization diagram of an embodiment of the present invention;

[0036] Figure 5 is a schematic diagram of a power line space plane of an embodiment of the present invention;

[0037] Figure 6 It is a schematic cross-sectional view of a power line catenary according to an embodiment of the present invention.

[0038] Main figure numerals: 100, divergent intersection; 101, drone A; 102, drone B; 103, drone C; 104, drone D; 105 drone E; 106, drone F; 107, drone G. DETAILED DESCRIPTION

[0039] In order to make the purpose, technical solutions and advantages of the present invention clearer, the following Figure 1-6 The invention is further described with examples.

[0040] like Figure 1 As shown, the present invention provides a power line inspection method using a drone swarm based on a power transmission line topology structure, which comprises the following steps:

[0041] S1. Obtain power line GIS (Geographic Information System) data and establish the topological structure of power transmission lines.

[0042] The topological structure of the power transmission line uses high-voltage transmission towers and power generation stations as nodes and power transmission lines as edges. Topological structure methods include: node-based topological structure method, edge-based topological structure method and hybrid topological structure method. The node-based topological structure method models the topological structure based on the nodes in the power transmission line. Common methods include node admittance matrix method, node recoding method, node grouping method, etc. These methods can effectively describe the relationship and connection mode of nodes in the transmission line. The edge-based topological structure method models the topological structure based on the edges in the power transmission line. Common methods include branch admittance matrix method, loop method, edge recoding method, etc. These methods can effectively describe the relationship and connection mode of edges in the transmission line. The hybrid topological structure method considers the topological structure of nodes and edges at the same time, and describes the topological structure of the transmission line by modeling nodes and edges. Common methods include busbar admittance matrix method, cell-based topological structure method, etc. These methods can more comprehensively describe the connection mode and topological structure of the transmission line. Different modeling methods are used to describe the relationship between nodes and edges in power transmission lines, thereby realizing the modeling and analysis of power transmission lines.

[0043] Power transmission lines are generally composed of elevated-line-elevated-line-elevated, etc., and they pass through power stations in the middle. The inspection cycle modeling is carried out based on the power line GIS data. The above method is used to construct the topological structure of the power transmission line and assign weights to each edge. These weights can represent the importance of the power line or the difficulty of inspection. The two entity types of elevated and transmission line are defined in the topological structure of the power transmission line. The elevated contains information such as the type of elevated, the location and number of insulators, GPS information, and the estimated inspection time. The transmission line contains information such as the location, number, length, span, thickness, and standard catenary model of the wire.

[0044] S2. According to the topological structure, clustering method and genetic algorithm are combined to determine the optimal number and location of drone airports.

[0045] In the topological structure of power transmission lines, according to the longitude and latitude data of the nodes obtained in step S1, the location distribution planning of power lines is carried out using clustering method and genetic algorithm, with the aim of determining the location of the airport and allocating a suitable number of drones at the minimum quantity cost.

[0046] S21. Obtain the longitude and latitude data of each node in the topological structure.

[0047] S22. Use a clustering method to cluster the acquired longitude and latitude data to obtain the center point of each cluster as the number and location of potential airports for drones.

[0048] The number of airport locations is determined by K-Means clustering, and the Bayesian Information Criterion is used to select the optimal number of clusters, which is the number of airports. This process determines the most suitable number of airports based on the data itself, rather than a fixed number. Once the optimal number of clusters is determined, the K-Means algorithm will cluster according to this number of clusters and provide an initial center location for each cluster, i.e., the airport.

[0049] S23. Use a genetic algorithm to optimize the number and location of potential airports. From the results of the genetic algorithm, select the chromosome with the best fitness to determine the optimal number and location of airports.

[0050] Next, a genetic algorithm is used to optimize these center locations. The role of the genetic algorithm here is to optimize the initial airport locations derived by K-Means, rather than changing the number of airports. It adjusts these locations to find a better solution that minimizes the objective function.

[0051] The fitness is determined by an improved fitness function, which can be set in two ways:

[0052] As a first setting of the improved fitness function, the fitness function is set to determine the function value based on two objectives: the total distance from all data points to the determined airport and the total distance from a specific location point to the selected airport, with the purpose of minimizing the weighted sum of the two objectives. The specific location point may be some key landmarks, city centers, important facilities or other places of special significance, such as drone maintenance points.

[0053] The fitness function of the first setting is composed of:

[0054] Total distance: An important part of the fitness function is to calculate the total distance of all data points to the selected airport. This reflects the effectiveness of the airport coverage area.

[0055] Distance to specific locations: This section calculates the total distance to the selected airport for certain specific locations, taking into account their proximity to the airport. Special locations can be, for example, important infrastructure or high-demand areas.

[0056] In the design process of the present invention, a penalty mechanism of the fitness function can also be appropriately established as needed. The penalty mechanism of the fitness function includes two layers:

[0057] The first level of penalty mechanism is the distance limit penalty between airports: ensure that the distance between all selected airports does not exceed the maximum flight distance of the drone or aircraft, so as to ensure safety and operational efficiency. For each selected pair of airport topology, if the distance between any two topology exceeds the preset maximum flight distance of 20 kilometers, a penalty is added to the fitness function value. Among them, the distance between any two topology can be the inspection path length of the power line plus the sum of the inspection path lengths of all elevated roads on the line. The inspection path length of an elevated road can be obtained by the total length of the surrounding path generated for the drone by the surrounding path strategy of the drone flying elevated structure expansion algorithm. The size of the penalty is proportional to the distance exceeded. Traverse all airport pairs and calculate the distance between each pair of airports. If the maximum distance is exceeded, the excess distance is accumulated as a penalty.

[0058] The second penalty mechanism is the airport construction cost total budget limit penalty: ensure that the sum of the construction costs of all airports does not exceed the total budget cost. If it exceeds the budget, a penalty is added to the fitness function value. The size of the penalty is proportional to the excess cost.

[0059] As another embodiment of the present invention, the second setting mode of the improved fitness function is: the fitness function is set to determine the function value based on the three objectives of the total distance from all data points to the determined airport, the total distance from the specific location point to the selected airport, and the construction cost of the airport, with the purpose of minimizing the weighted sum of the three objectives. The specific location point may be some key landmarks, city centers, important facilities or other places with special significance such as drone maintenance points, and can be customized as needed.

[0060] The fitness function of the second setting is composed of:

[0061] Total distance: An important part of the fitness function is to calculate the total distance of all data points to the selected airport. This reflects the effectiveness of the airport coverage area.

[0062] Distance from specific locations: Taking into account the proximity of certain specific locations to the airport, this section calculates the total distance from these points to the selected airport.

[0063] Airport construction cost: Because the overall budget of the project usually limits the number of airports, the second setting takes the construction cost of the airport into account in the fitness function.

[0064] The penalty mechanism of the fitness function in the second setting is the same as the first setting above, and will not be repeated here.

[0065] The following is an example of a certain line, which includes four lines:

[0066] de la LIT 225kV N0 1 ILE-NAPOLEON-LUTTERBACH,

[0067] de la LIT 225kV N0 1 LUTTERBACH-SIERENTZ、

[0068] de la LIT 225kV N0 1 ILE-NAPOLEON - OTTMARSHEIM,

[0069] de la LIT 225kV N0 1 CHALAMPE-SIERENTZ.

[0070] The following is an analysis of these four lines as examples:

[0071] (1) Analysis of the optimal number of clusters

[0072] The optimal number of clusters determined using the BIC method is 22, which means that based on the current power line data, 22 locations should be considered for airports to achieve maximum efficiency. These locations were determined using the KMeans algorithm based on the distribution of the data.

[0073] (2) Fitness evolution analysis

[0074] From the output of the given genetic algorithm, we can observe the evolution of the fitness value. During the 40 generations of evolution, the average fitness value decreased from 4.75582 in the first generation to 3.32337 in the last generation, which means that the algorithm gradually found better solutions for the airport locations.

[0075] (3) Fitness distribution

[0076] The fitness range of the first generation of the population, i.e., generation 0, is from 3.15692 to 9.67915. As the iterations proceed, we can see that the average fitness value gradually decreases, indicating that the overall performance of the population is improving. By generation 40, the fitness range has narrowed to 3.15692 to 4.28902. This shows that the overall performance of the population has been significantly improved and the quality of the solution has been enhanced.

[0077] This improvement in fitness usually means that the algorithm is effectively converging to a better solution. A decrease in fitness value and a narrowing range indicate that the algorithm is finding a solution that is closer to the optimal solution. As the number of iterations increases, the individuals in the population become more and more adapted to the given problem, and the quality of the solution improves accordingly.

[0078] (4) Genetic algorithm operation analysis

[0079] The crossover probability is 52% and the mutation probability is 20%. This configuration allows the population to have both sufficient exploration (through mutation) and sufficient utilization (through crossover) during the evolution process.

[0080] like Figure 2 As shown in the figure, the evolution of the fitness value during the iteration process can be observed from the figure:

[0081] ① Average Fitness: As the number of iterations increases, the average fitness tends to decrease, which indicates that the genetic algorithm is optimizing the solution to make it more suitable for the problem.

[0082] ② Minimum Fitness: The minimum fitness value has stabilized in the early iterations and remains at a low level, which indicates that the algorithm has found a relatively good solution in the early iterations.

[0083] ③ Maximum Fitness: The maximum fitness also shows a downward trend, but its decrease is larger than the average fitness, which means that the worst solution in the population is also being improved.

[0084] That is, the optimal number of clusters is: 22, the number of clusters used to initially determine potential airport locations. The fitness value of the optimal airport location is: 3.1569216787685224, which is the fitness score of the airport location configuration that was finally selected. The lower this score is, the better the overall performance of the selected location. The table in the output shows the progress of the genetic algorithm, including statistics for each generation: gen: number of generations, that is, the number of iterations of the genetic algorithm. nevals: the number of individuals evaluated in each generation. avg: the average fitness of the population. min: the fitness of the best individual in the population. max: the fitness of the worst individual in the population. The evolution of fitness with the number of iterations is shown in Figure 3 shown.

[0085] (5) If Figure 2 As shown, the final airport location and optimal airport location coordinates are shown in Table 1:

[0086] Table 1 Final airport location Optimal airport location coordinates

[0087]

[0088] In general, as the number of iterations increases, the fitness of the entire population is improving, especially some of the worst solutions. This verifies the effectiveness of the genetic algorithm in finding optimized solutions. After genetic algorithm optimization, the best airport locations were obtained. These locations match the distribution of data points very well and ensure that the inspection drone can complete its mission within the shortest distance.

[0089] S3. Plan the inspection route of the drone based on the geographic location information in the topological structure and the number and location of the determined drone airports; including:

[0090] In the UAV power transmission line inspection solution, path planning is a key technology. Path planning should take into account factors such as line height, tower distribution, terrain complexity, and inspection requirements to maximize inspection efficiency and coverage area. In addition, the UAV's track planning technology should be combined to avoid repeated coverage and missed inspections.

[0091] S31. Based on the number, location and location information of the determined drone airports in the topological structure, the airport locations and the distribution of power transmission lines are visualized, and manual or system route planning can be performed after visualization.

[0092] S32, setting each inspection path to have the determined drone airport location as the starting point and end point of the drone inspection path.

[0093] S33, forming a drone inspection path for each power station in the topological structure between two adjacent drone airports and the interconnected high-voltage transmission towers and transmission lines between the power stations, and the inspection paths of each drone do not overlap.

[0094] Specifically, there are several possible situations:

[0095] 1. If the topological structure path between two airports is a direct connection, then the topological structure path is the inspection path of a drone flight inspection.

[0096] 2. If there is a divergent topological structure path between two airports other than the directly connected topological structure path, and the end point is not an airport, then the divergent intersection point 100 of the topological structure path is recorded. Figure 4 It just coincides with the power station, and the drone inspection path takes off from the take-off airport, and inspects along its inspection topology path until the divergent topology is completed, then flies directly back to the divergent intersection, and then inspects along the remaining inspection topology path to the destination airport to form the drone inspection path.

[0097] 3. In addition to the directly connected topological structure path, there is a divergent topological structure path between the two airports, and the end point is the third airport. That is, there is a "T"-shaped topological structure path between the three airports. At this time, compare the three edges at the intersection of the "T"-shaped topological structure path and record the divergent intersection of the topological structure path. Let the first drone take off from the first departure airport along the inspection path of two of the edges and fly to the second or third airport. Let the second drone fly directly to the divergent intersection and then patrol along the remaining patrol topological structure path to the third or second airport at the destination to form the drone patrol path.

[0098] 4. For the single-sided divergent topological structure path connecting a single airport, merge it into the inspection path of the drone responsible for the airport and plan it in the flight direction so that it can form the drone inspection path with the shortest journey.

[0099] Furthermore, each inspection route is designed so as not to exceed the maximum flight distance of the UAV.

[0100] S4. A drone is configured for each drone airport. Alternatively, due to the above-mentioned planning of the flight direction, the configuration plan can also be fine-tuned. For example, if one of the drone airports is configured with two drones, the other airport will not be configured with a drone. All configured drones are allowed to fly and inspect along the planned inspection path, and the captured data is sent to the remote data center to complete the inspection process.

[0101] The main advantage of multi-machine collaborative power line inspection path planning is that it significantly improves the inspection efficiency. This method effectively reduces the burden of a single drone and speeds up the overall inspection process by assigning multiple drones to inspect power lines simultaneously. Each drone is responsible for a specific area and uses advanced camera technology and image recognition algorithms to capture and analyze the status of power lines in real time. At the same time, the location and number of drones are optimized and planned through clustering methods and genetic algorithms, further improving the efficiency of inspection. This method not only speeds up the inspection process, but also ensures the comprehensiveness and accuracy of the inspection. It is particularly suitable for inspection tasks of large-scale or complex power transmission lines.

[0102] like Figure 4 As shown in the figure, the drone swarm is assigned to start their inspection missions at airports numbered 1 to 7, and each drone takes off from its corresponding airport. Specifically, drones A to G take off from airports 1 to 7 respectively. This strategy ensures wide coverage of inspection activities and optimizes the collaboration and efficiency between drones.

[0103] In terms of inspection route arrangement, drone A 101 is tasked to fly from airport 1 to airport 2, and drone B 102 is tasked to fly from airport 2 to airport 3. In addition to being responsible for the straight inspection route from airport 2 to airport 3, drone B 102 will also perform additional divergent topology branch line inspections, which requires flying from airport 2 to the end of the path, then flying directly back to the divergent intersection, and then inspecting along the remaining inspection topology path to the destination airport 3. After that, drone C 103, drone D 104, drone E 105, drone F 106, and drone G 107 will respectively perform the tasks of airport 3 to airport 4, airport 4 to airport 5, airport 5 to airport 6, airport 6 to airport 7, and airport 7 back to airport 1. This coherent route layout forms a complete closed-loop inspection network, ensuring that the power lines are thoroughly and effectively inspected.

[0104] In addition, the inspection path includes a power line inspection sub-path and an overhead inspection sub-path.

[0105] The power line inspection sub-path includes: using drone cameras for real-time image recognition and tracking, automatically tracking power lines, capturing and analyzing the status of power lines in real time, matching the model of the power line catenary based on the power line information captured by the camera from the side, and if the power line is found to have a low degree of match with the catenary equation, or the power line is abnormally detached, or there are foreign objects hanging on the power line, the drone will record the log in time and automatically report the error. Considering the impact of wind and other weather conditions on the status of power lines, inspections are generally carried out when there is no wind or light wind. Among them, the low degree of match can be, for example, the power line is detached or drooping.

[0106] A catenary curve is a specific mathematical curve that usually describes the shape of a uniform flexible chain that is fixed at both ends and is only subject to gravity. The length of the catenary can be solved based on the horizontal distance of the catenary.

[0107] The process of obtaining the model of the catenary of the power line includes: obtaining the coordinates and horizontal distances of the two end points of the power line in the topological structure; calculating the catenary cross-sectional equation of the power line according to the coordinates and horizontal distances of the two end points; calculating the catenary space curve equation of the power line according to the obtained catenary cross-sectional equation; calculating multiple catenary space plane equations according to the coordinates and horizontal distances of the two end points of all power lines between two points:

[0108] y=acosh(x / a)

[0109] Among them, y refers to the vertical position of the point on the catenary, x refers to the horizontal position of the point on the catenary, and a is the parameter of the catenary, which affects the shape of the curve. Figure 5 The power line space plan shown in the figure, Figure 6 The catenary cross section of the power line shown, the coordinates of the end points of the catenary and the horizontal distance:

[0110] Catenary 1: Starting point coordinates: (-40.0, 75.26781426947176)

[0111] End point coordinates: (159.99000000010233, 75.26473753413867)

[0112] Horizontal distance: 199.99000000010233

[0113] Catenary 2: Starting point coordinates: (-25.0, 75.26781426947176)

[0114] End point coordinates: (174.99000000010233, 75.26473753413867)

[0115] Horizontal distance: 199.99000000010233

[0116] Catenary 3: Starting point coordinates: (10.0, 75.26781426947176)

[0117] End point coordinates: (209.99000000010233, 75.26473753413867)

[0118] Horizontal distance: 199.99000000010233

[0119] Catenary 4: Starting point coordinates: (25.0, 75.26781426947176)

[0120] End point coordinates: (224.99000000010233, 75.26473753413867)

[0121] Horizontal distance: 199.99000000010233

[0122] Catenary 5: Starting point coordinates: (-32.5, 105.26781426947176)

[0123] End point coordinates: (167.49000000010233, 105.26473753413867)

[0124] Horizontal distance: 199.99000000010233

[0125] Catenary 6: Starting point coordinates: (17.5, 105.26781426947176)

[0126] End point coordinates: (217.49000000010233, 105.26473753413867)

[0127] Horizontal distance: 199.99000000010233

[0128] Carry out tracking inspections on the sides of multiple power catenaries, match the power line information collected by the camera on the side, detect abnormal power line detachment, mismatch with the catenary equation, and foreign objects hanging on the power lines, and record logs and make forecasts in a timely manner.

[0129] The elevated inspection sub-path includes: positioning and identifying key points of the elevated structure. The key points of the elevated structure include tower top equipment, main beams and cross beams, supporting structures, grounding systems, fasteners and connectors, wires and cables, and insulators. Among them, the tower top social security is not lightning arrester, insulator string and ground wire, which are key components to ensure the safe operation of the transmission line. The main beam and cross beam are the main load-bearing parts of the elevated structure, bearing the weight of the entire structure and the influence of environmental factors. The supporting structure includes the columns, cross arms, brackets and other parts of the transmission tower, which are the supporting points of the entire elevated structure. The grounding system ensures that the current can be safely guided to the ground in the event of lightning or other electrical faults. Fasteners and connectors include bolts, screws, fastening rings, brackets, etc., which are used to firmly connect the components together. Wires and cables are the main part of the transmission line, used to transmit electrical energy. Insulators are used to isolate the conductive parts from the supporting structure to prevent current leakage.

[0130] The elevated structure expansion algorithm is used to calculate the optimal spatial orbiting path for the UAV. The sensors on the UAV are used to obtain the distance information between the UAV and the elevated road in real time. The distance information is combined with the known elevated structure model to enable the UAV to fly and identify the orbiting path according to the elevated structure expansion algorithm, ensuring a detailed scan of each key point from all angles, and timely recording of logs and automatic error reporting.

[0131] In the UAV power line inspection, the expansion algorithm surrounding path strategy is adopted to ensure that every key point of interest is fully and comprehensively inspected. This method is based on the expansion algorithm in image processing, which generates a virtual three-dimensional expansion area to highlight each point of interest, ensuring that the drone can capture detailed information of these points from multiple angles. Specifically, first, the points of interest are determined based on the data of the depth camera and the model of the transmission tower. Then, an expansion area is generated for each point, which is usually achieved by adding a virtual three-dimensional sphere around it. Next, an advanced path planning algorithm, such as A* or spatial genetic algorithm, is used to generate a path for the drone to surround these expansion areas. Finally, the drone flies along this path to ensure that each point of interest is scanned exhaustively from all angles. This strategy ensures complete and detailed coverage of each point of interest during the inspection.

[0132] As another embodiment of the present invention: Further, a total flight distance of a standard elevated path around these expansion areas can be calculated as a standard inspection distance of an elevated road. When calculating the fitness function, the standard inspection distances of all elevated roads involved between two airports are also added to the distance between the selected airports for calculation to improve the measurement accuracy.

[0133] Furthermore, considering the flight performance of the drone and possible environmental changes, the inspection path can also be adjusted in real time to ensure the safe flight of the drone. The drone uses basic sensors, such as GPS and simple obstacle avoidance sensors, which allow the drone to avoid basic obstacles and adjust the path according to predetermined rules in specific situations.

[0134] Furthermore, the drone camera is used to perform real-time image recognition and tracking of power lines, ensuring that the drone always flies along the power lines and achieves tracking flight of the power lines. When the drone encounters an elevated road, it automatically identifies the key structural points of the elevated road and performs local path planning. When performing local path planning, it combines the known elevated structure model to give priority to the path flying over the elevated road to avoid collision with the elevated road.

[0135] At the same time, data processing is also an indispensable part of the UAV power transmission line detection solution, which mainly includes image processing, signal processing, data analysis, etc. By processing and analyzing the collected data, faults and hidden dangers of the power transmission line can be discovered and handled in a timely manner, ensuring the accuracy and reliability of the detection results.

[0136] During the inspection process, the drone will transmit the collected data to the data center in real time through the cellular network. At the same time, the drone will also store the data in its CD / ROM. When the drone returns to the data center, it will upload the stored point cloud data to prepare for subsequent data recognition and analysis.

[0137] In the context of combining GIS technology, applying the topological structure information of power transmission lines and the network of power transmission line maps to drone path planning can further improve the efficiency and accuracy of drone transmission line anomaly detection. This drone power line inspection path planning method based on the topological structure of power transmission lines can select and customize drone models according to the characteristics and requirements of the transmission lines, and conduct inspection path planning to maximize the detection efficiency and coverage area. By processing and analyzing the collected data, the faults and hidden dangers of the transmission lines can be discovered and handled in a timely manner. UAV technology combined with the geographic information topological structure of power transmission lines for power line inspection has broad application prospects and promotion value. With the continuous development and maturity of drone technology, drone transmission line detection technology will be more widely used and promoted, providing strong support for the safe and stable operation of transmission lines.

[0138] The present invention discloses a method for power line inspection using a drone swarm based on the topological structure of a power transmission line. First, the inspection cycle is modeled according to the power line data, the topological structure of the power transmission line is constructed, and the weights are assigned, wherein the overhead and the transmission line are two core entities, each containing relevant parameter information. Then, clustering and genetic algorithms are used to plan the location and quantity of drones in the power line topological structure, aiming to minimize the number of airports and reasonably allocate drones. Furthermore, combined with geographic location information, a global inspection plan for drones is formulated and subdivided into subtasks for power lines and overhead. In the power line inspection, real-time image recognition matches the catenary model to ensure tracking flight, and the power line status is judged and abnormal reports are made based on the horizontal distance of the catenary. In addition, the expansion algorithm surround path strategy is introduced to ensure that the drone fully inspects each key point. Finally, the drone transmits data back in real time through the cellular network during flight, and uploads point cloud data after returning to prepare for subsequent analysis. The present invention combines GIS data, K-means algorithm, genetic algorithm, expansion algorithm and real-time image recognition technology to propose a new power line inspection planning and execution method, which ensures the efficiency and accuracy of the inspection, and has real-time feedback and data upload functions. At the same time, based on multi-machine collaboration, the power line inspection path planning is carried out, and the location and number of drones are optimized and planned through clustering methods and genetic algorithms, which further improves the efficiency of the inspection. This method not only speeds up the inspection process, but also ensures the comprehensiveness and accuracy of the inspection. It is particularly suitable for inspection tasks of large-scale or complex power transmission lines.

[0139] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

[0140] The present invention and its embodiments are described above, and such description is not restrictive. The drawings show only one embodiment of the present invention, and the actual structure is not limited thereto. In short, if ordinary technicians in the field are inspired by it, without departing from the purpose of the invention, they can design a structure and embodiment similar to the technical solution without creativity, which should belong to the protection scope of the present invention.

Claims

1. A method for inspecting power lines using drone swarms based on the topology of power transmission lines, characterized in that: It includes the following steps: S1. Obtain power line GIS data and establish the topological structure of power transmission lines; S2, based on the topological structure of power transmission lines, combined with clustering method and genetic algorithm, determine the optimal number and location of drone airports; It includes the following sub-steps: S21, obtaining the longitude and latitude data of each node in the topological structure; S22, clustering the acquired longitude and latitude data using a clustering method to obtain the center point of each cluster as the number and location of potential airports for drones; S23, using a genetic algorithm to optimize the number and location of potential airports for drones, and selecting the chromosome with the best fitness from the results of the genetic algorithm to determine the number and location of the optimal drone airports; S3, planning the drone inspection route based on the geographic location information in the topological structure and the number and location of the optimal drone airports determined; S4. Configure drones for the drone airport, and make all configured drones fly and inspect along the planned inspection paths, and send the captured data to the remote data center to complete the inspection process; The inspection path includes a power line inspection sub-path, which includes: using a drone camera to perform real-time image recognition and tracking, automatically tracking the power line, capturing and analyzing the current state information of the power line in real time, matching the captured current state information of the power line with the acquired catenary model of the power line, and when it is found that the matching degree is low, the power line is abnormally detached, or a foreign object is hung on the power line, the drone records the log in time and automatically reports the error; wherein the acquisition process of the power line catenary model includes: Get the coordinates and horizontal distances of the two end points of the power line in the topological structure; calculate the catenary section equation of the power line according to the coordinates and horizontal distances of the two end points; calculate the catenary space curve equation of the power line according to the obtained catenary section equation; calculate multiple catenary space plane equations according to the coordinates and horizontal distances of the two end points of all power lines between two points: y=acosh(x / a) Among them, y is the vertical position of the point on the catenary, x is the horizontal position of the point on the catenary, and a is the parameter of the catenary, which affects the shape of the curve.

2. The power line inspection method of drone swarm based on power transmission line topology structure according to claim 1 is characterized in that: Step S1 includes: Obtain power line GIS data, use high-voltage transmission towers and power generating stations as nodes, and transmission lines as edges to build the topological structure of power transmission lines.

3. The power line inspection method of UAV swarm based on power transmission line topology structure according to claim 1 is characterized in that: The fitness in step S23 is determined by an improved fitness function, which is set to determine the function value based on the total distance from all data points to the determined airport and the total distance from the specific location point to the selected airport, with the purpose of minimizing the weighted sum of the objectives.

4. The power line inspection method of UAV swarm based on power transmission line topology structure according to claim 1 is characterized in that: The fitness in step S23 is determined by an improved fitness function, which includes a penalty mechanism.

5. The power line inspection method of drone swarm based on power transmission line topology structure according to claim 1 is characterized in that: In step S3, the drone inspection path is planned according to the geographic location information in the topological structure and the number and location of the determined optimal drone airports, including: S31. Visualize the distribution of airport locations and power transmission lines based on the number and location of the determined optimal drone airports and the geographic location information in the topological structure; S32, setting each inspection path to have the determined optimal drone airport location as the starting point and end point of the drone inspection path; S33, the power stations in the topological structure between two adjacent drone airports and the interconnected transmission lines and high-voltage transmission towers between the power stations form a drone inspection path, and the inspection paths of the drones do not overlap.

6. The power line inspection method of drone swarm based on power transmission line topology structure according to claim 1 is characterized in that: In step S4, the inspection path also includes an elevated inspection sub-path, which includes: locating and identifying key points of the elevated structure, using the elevated structure expansion algorithm to calculate the optimal spatial surround path for the drone, using the sensors on the drone to obtain real-time distance information between the drone and the elevated structure, combining the acquired distance information with the known elevated structure model, and making the drone fly and identify according to the spatial surround path of the elevated structure expansion algorithm, ensuring a detailed scan of each key point of the elevated structure from all angles, and timely recording logs and automatically reporting errors.

7. The power line inspection method of UAV swarm based on power transmission line topology structure according to claim 6 is characterized in that: The overhead structure expansion algorithm surrounding path strategy specifically includes: The key points of the elevated structure are determined based on the data from the depth camera and the model of the elevated transmission tower. An expansion area is generated for each key point by adding a virtual three-dimensional sphere around the key point. The A* search algorithm and genetic optimization algorithm are used to generate an optimal spatial path for the drone to surround the expansion area.

8. The power line inspection method of UAV swarm based on power transmission line topology structure according to claim 6 is characterized in that: The inspection distance between each pair of airports is the total distance of the power lines between the two airports plus the sum of the lengths of the expansion algorithm space surrounding inspection paths of all overhead lines on the power lines.