Tower-conductor combined inspection path optimization method based on vehicle-mounted mobile nest

Abstract

The invention discloses a tower-lead combined inspection path optimization method based on a vehicle-mounted mobile nest, and aims to solve the problems that nest site selection and path planning are independent, the optimization target is single, and the solving precision is insufficient in the prior art. According to the method, alternative grid points are screened based on four principles of scientificity and the like; the method comprises the following steps: constructing a dual-target model containing an inspection vehicle ground network and an unmanned aerial vehicle air network through graph theory modeling, and taking the minimum deployment number of machine nests and the total inspection cost as targets; setting multi-dimensional constraints such as coverage constraints and wire continuity constraints; solving by adopting an improved NSGA-II algorithm which introduces a local search strategy and a priority coding and decoding mechanism, and outputting a Pareto frontier and a corresponding scheme; and finally, the transformation of the unmanned aerial vehicle from'tower-by-tower routing inspection along the line 'to'surface-by-grid routing inspection' is realized. According to the method, collaborative optimization of site selection and path is realized, the feasibility and inspection efficiency of machine nest deployment are improved, no blind area of joint inspection is ensured, the comprehensive cost is reduced, and the method is adaptive to a complex power grid inspection scene and has wide application value.

Description

Technical Field

[0001] This invention relates to the field of power grid inspection technology, and in particular to a method for optimizing the joint inspection path of poles and conductors based on vehicle-mounted mobile inspection stations. Background Technology

[0002] In recent years, with the rapid development of drone technology, its application in power line inspection has become increasingly widespread. Drones, with their advantages of maneuverability, safety, efficiency, and excellent visibility, have gradually replaced traditional manual inspection methods, becoming an important means of power transmission line inspection. However, in practical applications, drone inspection technology still faces many challenges, limiting its effectiveness in large-scale continuous inspection tasks.

[0003] Currently, the application of drones in power line inspection mainly relies on the deployment of drone nests and the construction of intelligent inspection and dispatch platforms. As the parking, charging, and data transmission center for drones, the deployment method of the drone nest directly affects the drone's operating range and efficiency. Existing drone nests are mainly divided into two types: fixed nests and mobile nests.

[0004] Fixed drone housings: Fixed drone housings are typically installed in specific locations to provide a stable environment for drone takeoff, landing, and charging. However, their deployment locations are subject to strict requirements, installation and maintenance costs are high, and they have low flexibility, making them difficult to adapt to complex and changing inspection environments.

[0005] Mobile drone shelters: To overcome the limitations of fixed drone shelters, the concept of mobile shelters has emerged. In particular, vehicle-mounted drone mobile shelters, by combining drones with inspection vehicles, expand the service range of drones and enhance the flexibility of on-demand inspections. This model reduces investment in fixed infrastructure construction and improves inspection efficiency. Although mobile shelters have shown great potential in power line inspection, existing technologies still have the following problems: 1. Separate handling of inspection needs: Traditional methods often treat the inspection needs of towers and conductors separately, focusing only on towers for aggregate coverage and site selection, while neglecting the important inspection need of conductors. For example, although CN111080832A proposes an inspection method and system for transmission line towers, this method does not fully consider the joint inspection needs of towers and conductors. This results in the conductor section potentially not being adequately covered during actual inspections, increasing the risk of missed inspections.

[0006] 2. Single-objective location selection model: Existing UAV nest location selection models in the literature usually adopt single-objective optimization, which makes it difficult to take into account multiple location selection factors such as economy, efficiency, and resource utilization. For example, although CN116430903A proposes a mobile nest UAV inspection method and system that supports single-UAV take-off and landing inspection and multi-UAV concurrent control, its location selection model is still mainly based on single-objective optimization, which may result in a non-optimal location selection and fail to meet complex and ever-changing inspection requirements.

[0007] 3. Fixed Coverage Radius: Traditional drone nesting models typically use a fixed coverage radius, neglecting the impact of environmental factors on the drone's operational range. In practical applications, the drone's operational range is affected by various factors such as RTK signals, communication signals, and natural conditions, making a fixed coverage radius unsuitable for complex environments. For example, in mountainous areas or densely populated urban areas with tall buildings, drone signals are easily interfered with, and a fixed coverage radius location model may result in blind spots during inspections.

[0008] 4. Insufficient Capability for Large-Scale Continuous Inspection: Due to a shortage of professional pilots, insufficient drone endurance, and weak autonomous planning and control capabilities, drones are currently unable to complete large-scale continuous inspection tasks. For example, while CN111080832A (Inspection Method and System for Transmission Line Towers) and CN116430903A (A Mobile UAV Inspection Method and System for Cellular Towers) propose methods to improve drone inspection efficiency, neither effectively addresses the issues of insufficient drone endurance and weak autonomous planning and control capabilities, thus limiting the widespread application of drones in power line inspection.

[0009] 5. Independent Decision-Making for Site Selection and Path Planning: In existing technologies, site selection and path planning for drones are mostly decided independently, without a collaborative optimization mechanism. This leads to site selection considering only a single constraint (such as coverage area), ignoring actual deployment needs such as road traffic and terrain environment, making it difficult for drones to be deployed quickly. At the same time, path planning does not take into account the two-tiered operational network of inspection vehicles and drones, resulting in problems such as high operating costs and insufficient battery life. In addition, the optimization objective is singular, failing to simultaneously minimize the number of drones deployed and optimize the total inspection cost, leading to insufficient resource utilization.

[0010] 6. Lack of local search strategies and priority mechanisms in solution algorithms: Existing technologies lack local search strategies and priority mechanisms when solving location path optimization problems, resulting in insufficient accuracy and diversity of Pareto front solutions, making it difficult to adapt to complex inspection scenarios. For example, in CN116430903A, a method and system for inspecting mobile nests of unmanned aerial vehicles, although an optimization method based on power consumption and inspection targets is proposed, the solution algorithm still lacks sufficient flexibility and accuracy, and cannot effectively cope with complex and ever-changing inspection environments.

[0011] Therefore, there is an urgent need for a vehicle-mounted mobile drone nest location optimization method that balances scientific site selection with path optimization and adapts to the needs of joint inspections, in order to overcome the shortcomings of existing technologies. This invention proposes a vehicle-mounted mobile drone nest location optimization method considering joint inspections of power poles and conductors. The aim is to improve the efficiency and quality of drone inspections through scientific and reasonable site selection and path planning, thereby ensuring the safe and stable operation of the power system. Summary of the Invention The technical problem this invention aims to solve is to provide a method for optimizing the joint inspection path of power poles and conductors based on vehicle-mounted mobile drone nests, addressing the technical problems of insufficient endurance, complex path planning, and inflexible deployment and scheduling in the field of power line inspection using drone inspection technology. Specifically, existing drone inspection methods are limited by the shortage of professional pilots, the limited endurance of drones, and weak autonomous planning and control capabilities when facing the inspection of large-scale transmission lines, making it difficult to achieve efficient and continuous inspection operations. Furthermore, traditional methods often treat the inspection needs of power poles and conductors separately, neglecting the importance of conductor inspection, resulting in incomplete inspection coverage and increasing the risk of missed inspections. Therefore, this invention aims to overcome the specific limitations of existing technologies and improve the efficiency and quality of power line inspection.

[0012] To achieve the above technical objectives, this invention provides a method for optimizing the joint inspection path of poles and conductors based on vehicle-mounted mobile towers, specifically including the following steps: Step 1: Determine the set of candidate grid points Based on four principles—scientific rigor, operability, safety, and economy—candidate grid points for vehicle-mounted mobile habitats were comprehensively selected: Key influencing factors: distribution of transmission lines, road traffic conditions, terrain and environmental conditions, and inter-grid coverage. Experience-based adaptation: Historical experience of manual drone flights and actual inspection data; Screening criteria: The grid point has good accessibility to power transmission lines, is suitable for road distance, has a terrain slope of ≤30°, and is far from the core area of ​​the power protection zone (no safety hazards).

[0013] Step 2: Construction of the dual-objective location selection path model A simplified model of transmission lines is constructed based on graph theory: transmission towers are abstracted as discrete nodes, and transmission conductors are abstracted as edges connecting these nodes, forming a power grid topology model; on this basis, a bi-objective location path model containing a two-level network is built: Level 1 network (ground operation network): The route of the inspection vehicle departing from the operation and maintenance station, visiting the alternative grid points and parking to wait for the drone to return, and returning to the operation and maintenance station after completing the operation; The second-level network (aerial operation network): The flight path of the drone taking off from the grid point take-off and landing platform, performing joint inspection tasks of towers and conductors within the coverage area, and then returning to the platform.

[0014] Step 3: Constructing the objective function and constraints (1) Biobjective function Objective 1: Minimize the number of grid points deployed. The core objective is to improve the utilization rate of individual grid points and reduce redundant deployments. Its expression is: (1); In the formula, As alternative grid nodes The decision-making variables, if the candidate grid points If selected Otherwise, it is 0. Under the condition of satisfying the full coverage of transmission lines in the operation and maintenance area, the fewer the number of grid points, the more inspection needs are allocated to a single grid point, the more fully the grid points are utilized, and the higher the inspection efficiency.

[0015] Objective 2: Minimize the total cost of inspections The total cost encompasses both the operating costs of the inspection vehicle and the flight costs of the drone, achieving full-process cost control. Its expression is: (2); In the formula, For maintenance stations; For the set of candidate grid points; It is the set of vehicle nodes that can be accessed by the inspection vehicle. ; The unit operating cost of the inspection vehicle; The unit flight cost of the drone; for Nodes and nodes The shortest road network distance between them; For nodes and nodes The Euclidean distance between them; For the route upper node Drive to the node The decision variables. If the inspection vehicle is on the route upper node Drive to the node but ,otherwise ; As a decision variable, if the transmission tower From alternative grid points Providing inspection services ,otherwise .

[0016] (2) Constraints Access restrictions at the maintenance station ensure that inspection vehicles depart from and return to the maintenance station, forming a closed-loop path. (3); In the formula, For the inspection vehicle route assembly; Grid point entry and exit constraints: When an inspection vehicle visits any grid point, the number of entries and exits remains consistent to avoid path interruptions. (4); The patrol vehicle's range is constrained; the total mileage of the patrol vehicle must be less than or equal to its maximum mileage to ensure operational feasibility. (5); In the formula, This is the maximum mileage of the inspection vehicle.

[0017] Inspection demand allocation constraints: (6); In the formula, Assign decision variables to transmission towers, if the transmission tower From alternative grid points Providing inspection services ,otherwise ; Assign decision variables to transmission lines if the transmission lines From alternative grid points Providing inspection services ,otherwise ; A collection of power transmission towers; A collection of power transmission lines; Coverage constraint: A transmission line can only be covered by a grid point if both ends of the towers are within the coverage radius of that grid point, ensuring no blind spots in joint inspections. (7); In the formula, and They are power transmission lines The two ends of the tower , and alternative grid points The Euclidean distance between the endpoints and All are using alternative grid points Centered on If the transmission line is inside a circle with radius [blank], then the transmission line... Can be selected grid points cover; As alternative grid points The coverage radius; Conductor continuity constraints require that the towers at both ends of the transmission line be assigned to the same alternative grid point to ensure the continuity of conductor inspection. (8); In the formula, and They are power transmission lines The two ends of the tower , and alternative grid points The distribution relationship between them.

[0018] Sub-loop constraint elimination uses continuous variables to define the inspection vehicle access order, avoiding path loop redundancy. (9); In the formula, , They are nodes , The continuous variable represents the number of vehicle nodes accessed by the inspection vehicle. , The order; For set The total number of nodes; Decision variable constraints , , All variables are discrete, ranging from 0 to 1, which aligns with practical decision-making logic. (10); In the formula, , They are nodes , The continuous variable represents the number of vehicle nodes accessed by the inspection vehicle. The order; It is a positive integer; Step 4: Solve using the improved N-step GA-II algorithm. An improved N-step GA-II algorithm is used to solve the model. The core optimization mechanism and process are as follows: (1) Optimization mechanism Priority-based encoding and decoding mechanism: adapts to multi-dimensional decision-making needs of location selection, allocation, and path selection, and generates a reasonable initial population; Local search strategy: Introduce 1-opt exchange, 2-opt exchange, and 3-opt exchange operators (adapting to grid point priority adjustment), as well as allocation exchange and reallocation operators (adapting to site selection allocation optimization) to improve the local optimality of the solution; Elite preservation strategy: merge parent and offspring populations, and maintain population diversity and optimal solution preservation through fast non-dominated sorting and crowding distance mechanism; (2) Solution process Generate the initial population and complete the encoding / decoding; Offspring populations are generated through crossover and mutation operations; Merge the parent and child populations and perform a fast non-dominated sort; Perform a local search on the optimal nondominated solution to generate an optimization population; Population updates are based on non-dominated sorting and crowding distance mechanisms; After the iteration terminates, the Pareto front and the corresponding mobile pod location, inspection demand allocation network, and inspection vehicle path are output.

[0019] Step 5: Change of Inspection Mode Based on the inspection requirements allocated to each grid point, independent inspection areas are divided to realize the transformation of drones from "inspecting tower by tower along the line" to "inspecting grid by grid": each grid point corresponds to a dedicated inspection area, and drones complete continuous inspection of all towers and conductors within the area, avoiding duplication or omissions across areas.

[0020] The pole-conductor joint inspection path optimization method based on vehicle-mounted mobile antenna housing provided by this invention has the following beneficial effects: 1. This invention effectively solves the problems hindering large-scale power line inspections in the field, such as insufficient drone endurance, complex path planning, inflexible deployment and scheduling, shortage of professional pilots, and weak autonomous planning and control capabilities. Through scientific and reasonable scheme planning, it improves the efficiency and quality of power line inspections, enabling drones to complete the inspection tasks of large-scale transmission lines more efficiently and stably.

[0021] 2. This invention takes into account practical factors such as the selection principle of candidate grid points, the coverage range of grid points and the range of inspection vehicles. It can accurately calculate the number and location of grid points in advance based on the spatial distribution of power transmission lines in the maintenance area, thereby realizing the reasonable allocation of inspection needs and the optimized planning of inspection vehicle routes.

[0022] 3. This invention aims to minimize the number of grid points deployed and the total inspection cost, maximizing resource utilization while also considering economic efficiency. This results in grid point location that is not only scientific and reasonable but also more practical in real-world applications, effectively reducing inspection costs.

[0023] 4. This invention proposes a comprehensive strategy of "full coverage of transmission lines", which fully considers the spatial distribution characteristics and topological relationships of towers and conductors. It can completely eliminate the risk of missed inspections caused by the lack of full coverage of transmission lines and their channels, effectively ensure the integrity of coverage and the continuity of inspection, effectively reduce the situation of missed inspections and re-flights by drones, and is applicable to a wider range of scenarios.

[0024] 5. This invention reduces the skill requirements for inspection personnel by using unmanned aerial vehicles (UAVs) for grid-based inspections and autonomous flight operations. Personnel do not need to have advanced professional flying skills; they can complete inspection tasks with simple operations, thereby promoting the construction of smart grids and improving the digitalization level and operation and maintenance efficiency of the power system.

[0025] 6. This invention establishes a site selection and path collaborative optimization mechanism, which can ensure that the deployment of the nests closely adapts to the actual inspection scenarios. Based on the characteristics and needs of different scenarios, the nest locations are rationally planned, while minimizing the number of nests deployed and the total inspection cost. This ensures both efficiency and economy while guaranteeing the inspection effect.

[0026] 7. This invention constructs a two-level operation network, which can well adapt to the collaborative operation needs of inspection vehicles and drones. Through a reasonable network architecture design, the two can cooperate efficiently during the inspection process. Furthermore, through coverage constraints and conductor continuity constraints, it ensures that the joint inspection of towers and conductors is free of blind spots and breaks, achieving comprehensive and detailed inspection.

[0027] 8. The present invention optimizes the solution algorithm and improves the NSGA-II algorithm by introducing local search and priority mechanism, which greatly improves the accuracy and diversity of Pareto front solutions. It can achieve efficient solution in complex and ever-changing inspection scenarios and provide a more accurate and comprehensive solution for site selection path optimization.

[0028] 9. This invention transforms the drone inspection mode from the traditional "inspection along the line and tower by tower" to "inspection by surface and grid by grid". This innovative mode can significantly improve inspection efficiency, reduce the drone's round-trip flight between different areas, reduce cross-regional redundant operations, and effectively reduce the drone's endurance pressure, extending its single inspection time and range.

[0029] 10. This invention has strong versatility and can flexibly adapt to the inspection needs of different terrain environments and power grid scales. Whether it is a complex terrain such as mountains, plains or cities, or a small power grid or a large power grid, it has a wide range of engineering application value and can provide effective solutions for various power inspection scenarios. Attached Figure Description

[0030] The present invention will be further described below with reference to the accompanying drawings and embodiments: Figure 1This is a technical flowchart of Embodiment 1 of the present invention; Figure 2 This is a schematic diagram illustrating the location path problem of a vehicle-mounted mobile terminal in Embodiment 1 of the present invention. Figure 3 This invention describes the crossover operation process of the improved NSGA-II algorithm in Embodiment 1 of the present invention. Figure 4 This is a distribution map of transmission lines, alternative grid points, and operation and maintenance stations according to Embodiment 2 of the present invention; Figure 5 This is the location path result under the scheme with the minimum number of grid points deployed in Embodiment 2 of the present invention; Figure 6 This is the location path result under the lowest total inspection cost scheme in Embodiment 2 of the present invention. Detailed Implementation

[0031] The technical solution of the present invention will be further described below with reference to the accompanying drawings and embodiments: Example 1 like Figure 1 As shown, this embodiment provides a method for optimizing the joint inspection path of poles and conductors based on vehicle-mounted mobile towers, including the following steps: Step 1: Determine the candidate grid points for vehicle-mounted mobile cell based on the principles of scientific rigor, operability, safety, and economy; Step 2: Establish a multi-target location path model for vehicle-mounted mobile nests; Step 3: Construct the objective function and constraints of the model; Step 4: Solve the model using the improved NSGA-II (Non-dominated Sorting Genetic Algorithm II, the core of the algorithm is non-dominated sorting technology II) algorithm to obtain the Pareto front, as well as the corresponding vehicle-mounted mobile nest grid point locations, inspection demand allocation network, and inspection vehicle path. Step 5: Divide the area into grids according to the inspection requirements assigned to each grid point, so as to realize the transformation of UAV from "inspecting tower by tower along the line" to "inspecting grid by grid".

[0032] In this embodiment, in step 1, for any candidate point, factors such as power transmission line distribution, road traffic conditions, terrain environment conditions, and inter-grid coverage are considered (see Table 1), and a screening judgment is made in conjunction with previous experience from manual drone flights and actual inspection results. This yields a set of candidate grid points.

[0033] Table 1. Main influencing factors in the selection of candidate grid points

[0034] In this embodiment, step 2 includes the following sub-steps: Step 2.1: Simplify the modeling of the transmission line based on graph theory. The transmission towers are abstracted as discrete nodes, and the transmission conductors are abstracted as edges connecting these nodes, that is, line segments with towers as endpoints.

[0035] Step 2.2, Problem Description. In a power line inspection scenario, assume there is one maintenance station within the maintenance area. , One alternative grid point , One power transmission tower and One transmission line Each candidate grid point has a coverage radius. Drones can only provide inspection services for power equipment within their coverage area. The inspection vehicle, carrying multiple drones, departs from the maintenance station to visit grid points, and then parks at those points to await the drones' return, forming the first-level network. The aerial operation trajectories of multiple drones taking off from the landing platform, performing inspection tasks at the grid points, and returning to the landing platform form the second-level network. After reaching the inspection vehicle's range limit or completing all inspection work in the area, the inspection vehicle, carrying the drones, returns to the maintenance station. Figure 2 As shown.

[0036] Step 2.3, Problem Assumptions. Considering the complexity of model building and solving, to increase the model's adaptability, and to closely reflect the actual operation scenario of power line inspection, the following problem assumptions are made: (1) Location assumption: The locations of the operation and maintenance station, alternative grid points and transmission lines are known, which is a static location problem.

[0037] (2) Coverage radius assumption: The coverage radius of the candidate grid points is affected by factors such as RTK signal, communication signal, UAV endurance, and natural conditions, and the coverage radius of each candidate grid point is different.

[0038] (3) Access assumptions: Inspection vehicles can only access maintenance stations and grid points, and cannot access inspection requests; drones must be transported to each grid point by inspection vehicles before they can access inspection requests.

[0039] (4) Path assumptions: The shortest road network distance is used for the travel distance of the inspection vehicle between nodes. This distance is calculated by the Network Analyst tool of ArcGlS. The UAV is not restricted by the road network, does not consider the influence of obstacles, and ignores the flight distance of the UAV during take-off and landing. Its flight path is measured by Euclidean distance.

[0040] In this embodiment, step 3 includes the following sub-steps: Step 3.1, the objective function of the model is as follows: 1) Objective function one: Minimize the number of grid points deployed: (1); In the formula, For decision variables, if candidate grid points If selected ,otherwise Under the condition of achieving full coverage of transmission lines within the operation and maintenance area, the fewer the number of grid points, the more inspection needs are allocated to a single grid point, the more fully the grid points are utilized, and the higher the inspection efficiency.

[0041] 2) Objective function two: Minimize the total inspection cost: (2); In the formula, For maintenance stations; For the set of candidate grid points; It is the set of vehicle nodes that can be accessed by the inspection vehicle. ;; The unit operating cost of the inspection vehicle; The unit flight cost of the drone; for Nodes and nodes The shortest road network distance between them; For nodes and nodes The Euclidean distance between them; As a decision variable, if the inspection vehicle is on the route upper node Drive to the node but ,otherwise ; As a decision variable, if the transmission tower From alternative grid points Providing inspection services ,otherwise The total cost of completing the inspection task for the entire maintenance area consists of two parts: the cost of the inspection vehicle's operation and the cost of the drone's flight.

[0042] Step 3.2, the constraints of the model include: 1) Access restrictions for the maintenance station: (3); In the formula, For the inspection vehicle route assembly; 2) Grid point in / out constraints: (4); 3) Range constraints of inspection vehicles: (5); In the formula, This refers to the maximum mileage of the inspection vehicle. 4) Constraints on the allocation of inspection requirements: (6); In the formula, Assign decision variables to transmission towers, if the transmission tower From alternative grid points Providing inspection services ,otherwise ; Assign decision variables to transmission lines if the transmission lines From alternative grid points Providing inspection services ,otherwise ; A collection of power transmission towers; A collection of power transmission lines; 5) Coverage constraints: (7); In the formula, and They are power transmission lines The two ends of the tower , and alternative grid points The Euclidean distance between the endpoints and All are using alternative grid points Centered on If the transmission line is inside a circle with radius [blank], then the transmission line... Can be selected grid points cover; As alternative grid points The coverage radius; 6) Conductor continuity constraints: (8) In the formula, and They are power transmission lines The two ends of the tower , and alternative grid points The allocation relationship between them; this constraint means that if the towers at both ends of the transmission line are assigned to the same grid point, then the line is also assigned to that grid point to ensure the continuity of the inspection task.

[0043] 7) Eliminating constraints in sub-loops: (9) In the formula, , They are nodes , The continuous variable represents the vehicle points visited by the inspection vehicle. , The order of; For set The total number of nodes; 8) Decision variable constraints: (10) In the formula, , They are nodes , The continuous variable represents the number of vehicle nodes accessed by the inspection vehicle. The order; It is a positive integer; In this embodiment, step 4, the improved NSGA-II algorithm, introduces a local search strategy and a priority-based encoding / decoding mechanism during the iteration process of the NSGA-II algorithm, specifically including the following sub-steps: Step 4.1: Chromosome encoding / decoding and initial population generation Chromosomes use a real number encoding method, and each chromosome... Depend on and It consists of two parts. This indicates the priority of each candidate grid point, and its length is the number of candidate grid points. The value of each gene locus ranges from 1 to The natural numbers are randomly arranged and generated, with smaller values ​​having higher priority; This represents the number of alternative grid points assigned to each inspection request, with a length equal to the number of inspection requests. The value range for each gene locus is: The initial population size is denoted as . Randomly generated according to the encoding rules .

[0044] Step 4.2, Record the iteration number .

[0045] Step 4.3: Generate offspring population through crossover and mutation operations. A partial mapping crossover operator is used: two individuals are randomly selected from the parent population, the crossover point determines the interval, and genes are exchanged to generate offspring. If the offspring are invalid after crossover ( If duplicate genes are present, then the crossover interval is used. A mapping rule is established for the correspondence between paternal genes, and duplicate genes are corrected, specifically as follows: Figure 3As shown. To enhance population diversity and prevent premature convergence, this paper introduces multimodal mutation after the crossover operation. Each mutation process randomly uses a predefined operation operator, the detailed design of which is described in Section 4.5. Each operation operator is selected with the same probability.

[0046] Step 4.4: Merge the parent and offspring populations to obtain... and to Perform a fast non-dominated sort.

[0047] Step 4.5: Perform a local search on the optimal non-dominated solution in the current population to obtain the population. The following two types of neighborhood operators are used to support local search and mutation operations: (1) Applied to the priority order of candidate grid points The operation operator. 1-opt swap operator: randomly generates two position indices. The 2-opt swap operator reverses the gene order between two positions. The 3-opt swap operator randomly orders the genes between two positions. Chromosomes obtained after performing these operations are still valid.

[0048] (2) Applied to site selection and allocation The operation operators include two types: assignment / exchange and redistribution. The assignment / exchange operator randomly selects two inspection requests and swaps their assigned candidate grid points. The redistribution operator randomly selects one inspection request and redistributes it to a different candidate grid point. It is then determined whether the chromosomes after performing these operations satisfy the model's coverage constraints. Chromosomes that violate the constraints are penalized with large values, reducing their fitness.

[0049] Step 4.6: Merge populations and get .

[0050] Step 4.7, Population Renewal. Non-dominated sorting and crowding distance mechanisms are used to hierarchically sort individuals within the population and maintain diversity. An elite retention strategy is then combined to construct a new generation of the population to ensure the continuation of superior genes, thereby accelerating population evolution.

[0051] Step 4.8: Determine if the iteration count has been reached; if so, let... If the calculation ends, return to step 4.3; otherwise, end the calculation, output the Pareto front, and use the decoding mechanism to obtain the corresponding vehicle-mounted mobile nest grid point locations, inspection demand allocation network, and inspection vehicle path.

[0052] Example 2 In another preferred embodiment, based on the above embodiment 1, such as Figures 4-6 As shown, this embodiment provides a method for optimizing the joint inspection path of poles and conductors based on vehicle-mounted mobile towers, including the following steps: Step 1: Determine the candidate grid points for vehicle-mounted mobile cell based on the principles of scientific rigor, operability, safety, and economy; Step 2: Establish a multi-target location path model for vehicle-mounted mobile nests; Step 3: Construct the objective function and constraints of the model; Step 4: Solve the model using the improved NSGA-II algorithm to obtain the Pareto front, as well as the corresponding vehicle-mounted mobile nest grid point locations, inspection demand allocation network, and inspection vehicle path; Step 5: Divide the area into grids according to the inspection requirements assigned to each grid point, so as to realize the transformation of UAV from "inspecting tower by tower along the line" to "inspecting grid by grid".

[0053] In this embodiment, step 1 involves selecting actual transmission line distribution data from a specific region for case analysis. For example... Figure 4 As shown, the area of ​​this maintenance zone is approximately 200 km². 2 It includes 11 transmission lines, 236 towers, and 237 transmission conductors. Based on the principle of selecting alternative grid points, 32 alternative grid points were obtained.

[0054] In this embodiment, step 2 includes the following sub-steps: Step 2.1: Simplify the modeling of the transmission line based on graph theory. The transmission towers are abstracted as discrete nodes, and the transmission conductors are abstracted as edges connecting these nodes, that is, line segments with towers as endpoints.

[0055] Step 2.2, Problem Description. In a power line inspection scenario, assume there is one maintenance station within the maintenance area. , One alternative grid point , One power transmission tower and One transmission line Each candidate grid point has a coverage radius. Drones can only provide inspection services for power equipment within their coverage area. The inspection vehicle, carrying multiple drones, departs from the maintenance station to visit grid points, and then parks at those points to await the drones' return, forming the first-level network. The aerial operation trajectories of multiple drones taking off from the landing platform, performing inspection tasks at the grid points, and returning to the landing platform form the second-level network. After reaching the inspection vehicle's range limit or completing all inspection work in the area, the inspection vehicle, carrying the drones, returns to the maintenance station.

[0056] Step 2.3, Problem Assumptions. Considering the complexity of model building and solving, to increase the model's adaptability, and to closely reflect the actual operation scenario of power line inspection, the following problem assumptions are made: (1) Location assumption: The locations of the operation and maintenance station, alternative grid points and transmission lines are known, which is a static location problem.

[0057] (2) Coverage radius assumption: The coverage radius of the candidate grid points is affected by factors such as RTK signal, communication signal, UAV endurance, and natural conditions, and the coverage radius of each candidate grid point is different.

[0058] (3) Access assumptions: Inspection vehicles can only access maintenance stations and grid points, and cannot access inspection requests; drones must be transported to each grid point by inspection vehicles before they can access inspection requests.

[0059] (4) Path assumptions: The shortest road network distance is used for the travel distance of the inspection vehicle between nodes. This distance is calculated by the Network Analyst tool of ArcGlS. The UAV is not restricted by the road network, does not consider the influence of obstacles, and ignores the flight distance of the UAV during take-off and landing. Its flight path is measured by Euclidean distance.

[0060] In this embodiment, step 3 includes the following sub-steps: Step 3.1, the objective function of the model is as follows: 1) Objective function one: Minimize the number of grid points deployed.

[0061] (1); In the formula, For decision variables, if candidate grid points If selected ,otherwise Under the condition of achieving full coverage of transmission lines within the operation and maintenance area, the fewer the number of grid points, the more inspection needs are allocated to a single grid point, the more fully the grid points are utilized, and the higher the inspection efficiency.

[0062] 2) Objective function two: Minimize the total inspection cost (2); In the formula, For maintenance stations; For the set of candidate grid points; It is the set of vehicle nodes that can be accessed by the inspection vehicle. ;; The unit operating cost of the inspection vehicle; The unit flight cost of the drone; for Nodes and nodes The shortest road network distance between them; For nodes and nodes The Euclidean distance between them; As a decision variable, if the inspection vehicle is on the route upper node Drive to the node but ,otherwise ; As a decision variable, if the transmission tower From alternative grid points Providing inspection services ,otherwise The total cost of completing the inspection task for the entire maintenance area consists of two parts: the cost of the inspection vehicle's operation and the cost of the drone's flight.

[0063] Step 3.2, the constraints of the model include: Access restrictions for maintenance stations: (3); Grid point in / out constraints: (4); Inspection vehicle range constraints: (5); Inspection demand allocation constraints: (6); Coverage constraints: (7); Conductor continuity constraints: (8); Sub-loop constraint elimination: (9); Decision variable constraints: (10); In this embodiment, step 4 includes the following sub-steps: Step 4.1: Input the obtained coordinate data and related parameters into the improved NSGA-II algorithm. The algorithm obtains 12 Pareto optimal solutions, and the solution set is evenly distributed in the target space.

[0064] Step 4.2: Considering the different tendencies of power sectors in selecting multi-objective optimization schemes under different conditions, this paper selects two extreme optimal solutions of objective functions from the Pareto solution set as typical schemes for analysis. Scheme 1 is the scheme with the fewest grid point deployments, with 13 grid points and a total inspection cost of 22,534.91 yuan / time. Scheme 2 is the scheme with the lowest total inspection cost, with 24 grid points and a total inspection cost of 17,074.25 yuan / time. Figure 5 and Figure 6 The images show the site selection results for Scheme 1 and Scheme 2, respectively. Green dots represent selected vehicle-mounted UAV grid points, and black circular areas represent the coverage area of ​​each grid point. Different colors are used to distinguish the inspection requirements assigned to each grid point. Finally, based on the inspection requirements assigned to each grid point, the area is divided into grids, realizing the transformation of UAVs from "inspecting towers along the line" to "inspecting grids by area".

[0065] This embodiment verifies the feasibility and superiority of the method of the present invention through specific parameter configuration and process execution: the grid points selected in the site selection stage are adapted to the terrain and road conditions; the dual-objective model achieves a balance between the number of grid cells and the cost; the improved NSGA-II algorithm has high solution accuracy, and the Pareto front solution covers different demand scenarios; the "surface-by-grid inspection" mode significantly improves the inspection efficiency and completeness, and is adapted to the power grid inspection needs of complex mountainous and hilly areas.

[0066] In the preferred embodiment, the selection of candidate grid points in step 1 is determined by screening based on the principles of scientific rigor, operability, safety, and economy, combined with factors such as power transmission line distribution, road traffic conditions, terrain conditions, inter-grid coverage, and experience with manual drone operations and actual inspection data. This approach ensures that candidate grid points meet actual inspection needs while maximizing coverage of key areas and minimizing blind spots. Furthermore, by comprehensively considering multiple factors, the accuracy and rationality of grid point selection are improved, laying a solid foundation for efficient drone inspection operations in the future.

[0067] In the preferred embodiment, the screening criteria include: accessibility of power transmission lines within the grid coverage area, suitability for road distance, terrain slope ≤ 30°, and absence of safety hazards (far from the core area of ​​power protection zones). These settings ensure that the screened area has both construction feasibility and operational convenience, while mitigating legal risks. By overlaying and calculating various elements using GIS (Geographic Information System) spatial analysis tools, a list of candidate areas meeting the criteria is automatically generated, providing data support for subsequent site selection.

[0068] In the preferred embodiment, the model construction in step 2 is based on simplified graph theory modeling: power transmission towers are abstracted as discrete nodes, and power transmission lines are abstracted as edges connecting the nodes; the dual-objective site selection path model includes two-level networks: the first level is the ground operation network for inspection vehicles, and the second level is the aerial operation network for unmanned aerial vehicles (UAVs); this setup enables collaborative ground and aerial inspections, improving coverage capabilities in complex terrains. The first-level network optimizes the inspection vehicle's docking points and travel paths, while the second-level network plans the UAV's take-off and landing points and flight trajectories, balancing inspection efficiency and energy consumption costs through node weight allocation.

[0069] In the preferred scheme, the first-level network is the inspection vehicle trajectory: starting from the maintenance station, visiting alternative grid points, parking and waiting for the drone to return, and then returning to the maintenance station; the second-level network is the drone trajectory: taking off from the grid point take-off and landing platform, performing inspection tasks within the coverage area, and then returning to the platform; the above settings can achieve efficient collaboration between inspection vehicles and drones. The inspection vehicle provides mobile take-off and landing points and material resupply for the drone, while the drone expands the inspection range and improves inspection efficiency. The two complement each other's advantages, effectively reducing maintenance costs and time, and improving the overall inspection quality.

[0070] In the preferred embodiment, the objective function in step 3 is a dual objective: minimizing the number of grid points deployed and minimizing the total inspection cost. Minimizing the number of grid points aims to improve the utilization rate of individual grid points and reduce redundant deployments. The total cost of completing the inspection task for the entire maintenance area consists of two parts: the cost of the inspection vehicle's travel and the cost of the drone's flight. This setup ensures both the economic efficiency of the grid point layout and the optimization of inspection efficiency. Through the collaborative optimization of these dual objectives, efficient resource allocation can be achieved, reducing overall maintenance costs while increasing the inspection coverage and frequency, providing more scientific and reasonable decision support for maintenance management.

[0071] In the preferred embodiment, the constraints in step 3 include maintenance station access constraints, grid point access constraints, inspection vehicle range constraints, inspection demand allocation constraints, coverage constraints, conductor continuity constraints (the towers at both ends of the transmission conductor must be assigned to the same candidate grid point), sub-loop elimination constraints, and decision variable constraints. These settings ensure that the scheme balances efficiency and safety during execution. The maintenance station and grid point access rules prevent resource waste, the range constraints prevent mid-journey stoppages, the demand allocation and coverage constraints ensure comprehensive inspections, the conductor continuity and sub-loop elimination constraints optimize path rationality, and the decision variable constraints ensure model solvability.

[0072] In the preferred embodiment, the optimization mechanism for improving the NSGA-II algorithm in step 4 includes: a priority-based encoding / decoding mechanism, a local search strategy (including 1-opt / 2-opt / 3-opt swap operators and allocation / redistribution operators), an elite retention strategy, non-dominated sorting, and a crowding distance mechanism. These settings can effectively balance the algorithm's global exploration and local development capabilities. Priority encoding / decoding improves the diversity of solutions, the local search strategy enhances the accuracy of local optimization, elite retention ensures the inheritance of high-quality solutions, and non-dominated sorting and crowding distance maintain the uniformity of the solution set distribution.

[0073] In the preferred scheme, the algorithm solution process in step 4 is as follows: initial population generation → crossover and mutation to generate offspring → merging populations and fast non-dominated sorting → local search optimization → population update → iterative termination to output the Pareto front, the location of the vehicle-mounted mobile nest grid points, the inspection demand allocation network, and the inspection vehicle path. This setup effectively balances the algorithm's global search and local optimization capabilities, avoiding getting trapped in local optima. Fast non-dominated sorting ensures solution set diversity, while local search further refines the accuracy of high-quality solutions. The final Pareto front output provides decision-makers with a basis for multi-objective trade-offs.

[0074] In the preferred embodiment, the UAV inspection mode in step 5 is changed from "inspection along the line tower by tower" to "inspection by area grid by grid," with each grid point corresponding to an independent inspection area. The UAV completes continuous inspection of all towers and conductors within the coverage area. This setting can significantly improve inspection efficiency, reduce repeated flight paths, and lower energy consumption. At the same time, dividing the network by grid facilitates accurate location of fault points, enabling rapid response and maintenance, improving the stability and security of power grid operation, and providing stronger guarantees for power supply.

[0075] In summary, the pole-conductor joint inspection path optimization method based on vehicle-mounted mobile drone nests provided by this invention precisely addresses the key problems existing in UAV inspection technology in the power inspection field, effectively solving difficulties such as insufficient endurance, complex path planning, and inflexible deployment and scheduling. In current power inspection scenarios, large-scale transmission line inspections face numerous challenging situations. On the one hand, the shortage of professional drone pilots, the limitations of UAV endurance, and the weakness of autonomous planning and control capabilities make it difficult for UAVs to achieve efficient and continuous inspection operations. On the other hand, traditional methods treat pole and conductor inspection needs in a fragmented manner, failing to recognize the importance of conductor inspection, leading to incomplete inspection coverage and a significantly increased risk of missed inspections. This invention addresses these practical difficulties by innovatively proposing a vehicle-mounted mobile drone nest location optimization method that considers pole-conductor joint inspection, aiming to improve the overall efficiency and quality of power inspection.

[0076] This invention demonstrates unique advantages in several aspects. First, it breaks free from traditional thinking, abandoning the fixed pattern of treating tower and conductor inspection needs separately, and comprehensively considering the inspection requirements of transmission lines, thus opening up a completely new direction for power line inspection. Second, it constructs a two-level operation network, cleverly incorporating the collaborative operation needs of inspection vehicles and drones into the overall planning scope, forming a unique operation system that contrasts sharply with previous inspection methods using single equipment or simple combinations. Third, it boldly innovates the drone inspection mode, upgrading from the traditional "inspection along the line and tower by tower" to "inspection by area and grid by grid." This innovative mode is unique in the field of power line inspection, significantly improving inspection efficiency and ensuring the integrity of coverage. Fourth, it establishes a site selection and path collaborative optimization mechanism, cleverly achieving the goals of minimizing the number of drone deployments and the lowest total inspection cost simultaneously, proposing a novel and practical collaborative optimization concept that balances efficiency and economy.

[0077] Furthermore, this invention is also remarkably innovative. In terms of objective setting, minimizing the number of grid point deployments and the total inspection cost is the core optimization objective, fully balancing resource utilization and economy. Through scientific and reasonable objective planning, it provides a highly innovative solution for the optimal allocation of power inspection resources. At the algorithm application level, an improved NSGA-II algorithm is introduced, cleverly incorporating local search strategies and priority-based encoding / decoding mechanisms during algorithm iteration, greatly improving the algorithm's solution efficiency and accuracy. This provides a more effective tool for site selection path optimization, demonstrating unique pioneering value in the field of algorithm application. In terms of technical assurance, coverage constraints and conductor continuity constraints ensure that the joint inspection of towers and conductors is free of blind spots and discontinuities, guaranteeing the comprehensiveness and continuity of inspections from a technical perspective, and innovatively solving the problem of missed inspections that may occur in traditional inspections. In terms of modeling methodology, candidate grid points are determined based on the principles of scientific rigor, operability, safety, and economy in site selection. Graph theory is used to simplify the modeling of transmission lines, abstracting transmission towers as discrete nodes and transmission conductors as connecting edges, providing an innovative and scientifically sound modeling method for site selection path planning.

[0078] Furthermore, this invention can accurately calculate the deployment quantity and location of grid points in advance based on the spatial distribution of transmission lines within the maintenance area, enabling the allocation of inspection needs and the planning of inspection vehicle routes. This effectively solves problems such as insufficient drone endurance, weak autonomous planning and control capabilities, and inflexible deployment and scheduling, powerfully promoting the development of grid-based drone inspection in the power industry towards high efficiency. Simultaneously, the deep integration of Geographic Information Systems (GIS) with drone inspection technology fully utilizes GIS's powerful spatial analysis and data processing capabilities, providing a new perspective and scientific method for the site selection and deployment of vehicle-mounted mobile drone nests. When constructing the multi-objective site selection path model for vehicle-mounted mobile drone nests, practical factors such as the selection principles of candidate grid points, grid point coverage, and the endurance of the inspection vehicle are fully considered. The optimization objective is to minimize the number of grid points deployed and the total inspection cost, balancing resource utilization and economy, making the model more scientific and practical. Compared to the traditional "full coverage of transmission towers" strategy, the "full coverage of transmission lines" comprehensive strategy proposed in this invention fully considers the spatial distribution characteristics and topological relationships of towers and conductors, ensuring the integrity of coverage and the continuity of inspection, and can adapt to a wider range of scenarios. The implementation of this invention's technical solution not only successfully solves problems such as insufficient endurance and excessively high skill requirements for inspection personnel during drone inspections, but also reduces inspection costs by optimizing drone nesting site selection and reducing redundant investment. This powerfully promotes the construction and development of smart grids, provides the power industry with intelligent and efficient inspection solutions, facilitates the deep integration of drone technology and smart grids, and injects new vitality into the digital transformation of the power industry.

Claims

1. A method for optimizing a tower-conductor combined inspection path based on a vehicle-mounted mobile robot nest, characterized in that, The method comprises the following steps: Step 1: determining a set of candidate grid points of the mobile vehicle nest; Step 2: constructing a double-target site selection path model of the mobile vehicle nest; Step 3: constructing a target function and constraint condition of the model; Step 4: solving the model by using an improved NSGA-II algorithm, and outputting a Pareto front and a corresponding scheme; Step 5: dividing a regional grid based on the distribution result of the inspection demand, and realizing a change of the UAV inspection mode.

2. The tower-conductor joint inspection path optimization method based on the vehicle-mounted mobile nest according to claim 1, characterized in that, The determination mode of the set of candidate grid points in the step 1 is: based on the principles of scientificity, operability, safety and economy, and in combination with the distribution of the power transmission line, the road traffic condition, the terrain environmental condition, the inter-network coverage condition and the UAV manual flying experience and the actual inspection condition, screening and judgment are performed.

3. The method of claim 2, wherein the method further comprises: determining a path of the tower-conductor joint inspection based on the mobile nest. The screening and judgment standard comprises: the accessibility of the power transmission line in the grid point coverage range, the distance adaptation to the road, the terrain slope ≤ 30°, and no safety hidden danger.

4. The method of claim 1, wherein the method further comprises: The model construction in the step 2 is based on a graph theory simplified modeling: the power transmission tower is abstracted as a discrete node, and the power transmission conductor is abstracted as an edge connecting the nodes; the double-target site selection path model comprises two levels of networks, the first level is a ground operation network of the inspection vehicle, and the second level is an aerial operation network of the UAV.

5. The method of claim 4, wherein the method further comprises: The first level network is a track of the inspection vehicle: starting from an operation station, visiting the candidate grid points, parking and waiting for the UAV to return, and then returning to the operation station; the second level network is a track of the UAV: taking off from a grid point take-off platform, performing an inspection task in the coverage range, and then returning to the platform.

6. The method of claim 1, wherein the method further comprises: The target function in the step 3 is a double target. Minimizing grid point deployment number : （1）； In the formula, alternative grid node decision variable; Minimizing total cost of inspection : （2）； wherein, is a set of operation and maintenance stations; is a set of alternative grid points; is a set of vehicle nodes that can be accessed by the inspection vehicle; is a unit travel cost of the inspection vehicle; is a unit flight cost of the unmanned aerial vehicle; is is the shortest road network distance between a node and a node is the Euclidean distance between a node and a node is the Euclidean distance between a node and a node is a decision variable for traveling from a node to a node on a route .

7. The method of claim 1, wherein the method further comprises: determining a path of a tower-conductor joint inspection based on the mobile nest. The constraint condition in the step 3 comprises an operation station access constraint, a grid point access constraint, an inspection vehicle endurance constraint, an inspection demand distribution constraint, a coverage constraint, a conductor continuity constraint, a sub-loop elimination constraint and a decision variable constraint, and specifically as follows: The operation station access constraint: （3）； In the formula, is a set of inspection vehicle routes; The grid point access constraint: （4）； The inspection vehicle endurance constraint: （5）； In the formula, is the maximum driving range of the inspection vehicle; The inspection demand distribution constraint: （6）； wherein assigning decision variables to transmission towers; assigning decision variables to transmission lines; a set of transmission towers; a set of transmission lines; The coverage constraint: （7）； wherein, and are the two end towers of the transmission line , , the Euclidean distance between the candidate grid point and the transmission line is the coverage radius of the candidate grid point . The conductor continuity constraint: （8）； wherein and are the two ends of the transmission line tower , and the distribution relationship between the alternative grid points ; The sub-loop elimination constraint: （9）； wherein , are continuous variables of nodes , , is the total number of nodes of the set . The decision variable constraint: （10）； wherein , are continuous variables of the nodes , respectively; is a positive integer.

8. The method of claim 1, wherein the method further comprises: determining a path of a tower-conductor joint inspection based on the mobile nest. The optimization mechanism of the improved NSGA-II algorithm in the step 4 comprises: a coding and decoding mechanism based on priority, a local search strategy, an elite reservation strategy, a non-dominated sorting and crowding distance mechanism.

9. The method of claim 1, wherein the method further comprises: determining a path of a tower-conductor joint inspection based on the mobile nest. The algorithm solving process in the step 4 is: initial population generation → cross variation to generate offspring → merging of the population and fast non-dominated sorting → local search optimization → population updating → iteration termination to output a Pareto front, a grid point position of the mobile vehicle nest, an inspection demand distribution network and an inspection vehicle path.

10. The method of claim 1, wherein the method further comprises: determining a path of a tower-conductor joint inspection based on the mobile nest. The change of the UAV inspection mode in the step 5 is: from "tower-by-tower inspection along the line" to "grid-by-grid inspection by area", each grid point corresponds to an independent inspection area, and the UAV completes continuous inspection of all towers and conductors in the coverage range.

Images