Substation inspection method based on unmanned aerial vehicle cluster, electronic device and storage medium

Abstract

The application discloses a transformer substation inspection method based on a UAV cluster, an electronic device and a storage medium. The method comprises the following steps: obtaining an inspection task, geographical environment information and endurance information of each UAV in the UAV cluster of a transformer substation; inputting the inspection task, the geographical environment information and the endurance information into a task planning model to obtain a target inspection area and a target inspection path of each UAV, wherein the task planning model is determined based on a graph diffusion network and a graph convolution network; and controlling the UAV cluster to inspect the transformer substation according to the target inspection area and the target inspection path to obtain an inspection result, wherein the inspection result is used to determine whether the UAV cluster completes the inspection task. The application solves the technical problems of high complexity and low efficiency of the UAV power grid inspection operation in the related art.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of substation inspection, in particular to a substation inspection method based on a UAV cluster, an electronic device and a storage medium. BACKGROUND

[0002] With the rapid development of UAV technology, the application of UAV in power grid inspection has gradually been valued. The existing UAV power grid inspection technology has the following problems:

[0003] 1) The coverage of single UAV inspection is limited, and the efficiency is low: the endurance and operation range of single UAV are relatively limited, especially when facing large-scale power lines and complex geographical environment, single-machine operation cannot quickly complete the full-area inspection task, resulting in low inspection efficiency. And in order to realize large-scale inspection, it needs to take off and land several times and charge, which increases the time cost of inspection.

[0004] 2) The complexity of UAV cluster scheduling is high: although UAV cluster can improve efficiency through cooperative operation, when the number of UAVs is large, the complexity of task allocation, flight path planning, obstacle avoidance and other problems increases exponentially, affecting the speed and reliability of inspection.

[0005] In view of the above problems, no effective solution has been proposed so far. SUMMARY

[0006] The embodiments of the present application provide a substation inspection method based on a UAV cluster, an electronic device and a storage medium, to at least solve the technical problem of high complexity and low efficiency of UAV power grid inspection in related technologies.

[0007] According to an aspect of one of the embodiments of the present application, a substation inspection method based on a UAV cluster is provided, comprising: obtaining the inspection task of the substation, the geographical environment information and the endurance information of each UAV in the UAV cluster; inputting the inspection task, the geographical environment information and the endurance information into a task planning model to obtain the target inspection area and the target inspection path of each UAV, wherein the task planning model is determined based on a graph diffusion network and a graph convolution network; controlling the UAV cluster to inspect the substation according to the target inspection area and the target inspection path, and obtaining the inspection result, wherein the inspection result is used to determine whether the UAV cluster completes the inspection task.

[0008] Optionally, the inputting the inspection task, the geographical environment information and the endurance information into a task planning model to obtain the target inspection area and the target inspection path of each unmanned aerial vehicle comprises: determining the initial inspection area and the initial inspection path of each unmanned aerial vehicle based on the inspection task, the geographical environment information and the endurance information; determining the diffusion capability expression matrix and the role importance expression matrix of the unmanned aerial vehicle cluster based on the initial inspection area and the initial inspection path; determining the comprehensive score of each unmanned aerial vehicle based on the diffusion capability expression matrix and the role importance expression matrix; and adjusting the initial inspection area and the initial inspection path according to the comprehensive score to obtain the target inspection area and the target inspection path.

[0009] Optionally, the determining the initial inspection area and the initial inspection path of each unmanned aerial vehicle based on the inspection task, the geographical environment information and the endurance information comprises: dividing the to-be-inspected area of the transformer substation based on the inspection task and the geographical environment information to obtain a division result; performing task allocation on the unmanned aerial vehicle cluster according to the division result to obtain the initial inspection area; and analyzing the initial inspection area, the geographical environment information and the endurance information by using a path planning algorithm to determine the initial inspection path.

[0010] Optionally, the determining the diffusion capability expression matrix of the unmanned aerial vehicle cluster based on the initial inspection area and the initial inspection path comprises: establishing a communication network of the unmanned aerial vehicle cluster according to the initial inspection area and the initial inspection path, wherein the nodes of the communication network represent each unmanned aerial vehicle in the unmanned aerial vehicle cluster; and simulating the information flow transmission process of the communication network by using a graph diffusion network to obtain the diffusion capability expression matrix.

[0011] Optionally, the determining the role importance expression matrix of the unmanned aerial vehicle cluster based on the initial inspection area and the initial inspection path comprises: performing role division on each unmanned aerial vehicle based on the initial inspection area and the initial inspection path to obtain role information of each unmanned aerial vehicle; constructing a role graph of the unmanned aerial vehicle cluster by using the role information and the initial inspection area, wherein the role graph is used to determine the position information and the importance level of each unmanned aerial vehicle; and inputting the role graph, the initial inspection area and the initial inspection path into a graph convolution network to obtain the role importance expression matrix.

[0012] Optionally, the task planning model at least comprises a multilayer perceptron network, and the determining the comprehensive score of each unmanned aerial vehicle based on the diffusion capability expression matrix and the role importance expression matrix comprises: obtaining a diffusion score of each unmanned aerial vehicle based on the diffusion capability expression matrix; obtaining a role score of each unmanned aerial vehicle based on the role importance expression matrix; and performing data fusion processing on the diffusion score and the role score by using the multilayer perceptron network to obtain the comprehensive score.

[0013] Optionally, the adjusting the initial inspection region and the initial inspection path according to the comprehensive score comprises: obtaining grade information of each sub-inspection region in the division result; establishing a task allocation mapping relationship based on the grade information and the comprehensive score; adjusting the initial inspection region based on the task allocation mapping relationship to obtain the target inspection region; and adjusting the initial inspection path based on the target inspection region to obtain the target inspection path.

[0014] According to one of the embodiments of the present application, a substation inspection device based on a UAV cluster is also provided, which comprises: an obtaining module, configured to obtain an inspection task of a substation, geographical environment information, and endurance information of each UAV in the UAV cluster; a determining module, configured to input the inspection task, the geographical environment information, and the endurance information into a task planning model to obtain a target inspection region and a target inspection path of each UAV, wherein the task planning model is determined based on a graph diffusion network and a graph convolution network; and a control module, configured to control the UAV cluster to perform inspection on the substation according to the target inspection region and the target inspection path to obtain an inspection result, wherein the inspection result is used to determine whether the UAV cluster completes the inspection task.

[0015] Optionally, the determining module is further configured to: determine an initial inspection region and an initial inspection path of each UAV based on the inspection task, the geographical environment information, and the endurance information; determine a diffusion capability expression matrix and a role importance expression matrix of the UAV cluster based on the initial inspection region and the initial inspection path; determine a comprehensive score of each UAV based on the diffusion capability expression matrix and the role importance expression matrix; and adjust the initial inspection region and the initial inspection path according to the comprehensive score to obtain the target inspection region and the target inspection path.

[0016] Optionally, the determining module is further configured to: divide an inspection region of the substation based on the inspection task and the geographical environment information to obtain a division result; allocate tasks to the UAV cluster according to the division result to obtain the initial inspection region; and analyze the initial inspection region, the geographical environment information, and the endurance information by using a path planning algorithm to determine the initial inspection path.

[0017] Optionally, the substation inspection device based on the UAV cluster further comprises an establishing module, configured to establish a communication network of the UAV cluster according to the initial inspection region and the initial inspection path, wherein a node of the communication network represents each UAV in the UAV cluster; and the substation inspection device based on the UAV cluster further comprises a simulation module, configured to simulate an information flow transmission process of the communication network by using the graph diffusion network to obtain the diffusion capability expression matrix.

[0018] Optionally, the determining module is further configured to divide roles of the unmanned aerial vehicles based on the initial inspection area and the initial inspection path, to obtain role information of the unmanned aerial vehicles; and the establishing module is further configured to construct a role graph of the unmanned aerial vehicle cluster by using the role information and the initial inspection area, wherein the role graph is used to determine position information and an importance level of each unmanned aerial vehicle; and the determining module is further configured to input the role graph, the initial inspection area and the initial inspection path into a graph convolution network, to obtain a role importance expression matrix.

[0019] Optionally, the obtaining module is further configured to obtain a diffusion score of each unmanned aerial vehicle based on the diffusion capability expression matrix, and obtain a role score of each unmanned aerial vehicle based on the role importance expression matrix; and the determining module is further configured to perform data fusion processing on the diffusion score and the role score by using a multilayer perceptron network, to obtain a comprehensive score.

[0020] Optionally, the obtaining module is further configured to obtain level information of each sub-inspection area in the division result; the establishing module is further configured to establish a task allocation mapping relationship based on the level information and the comprehensive score; and the determining module is further configured to adjust the initial inspection area based on the task allocation mapping relationship, to obtain a target inspection area, and adjust the initial inspection path based on the target inspection area, to obtain a target inspection path.

[0021] According to an embodiment of the present application, an electronic device is provided, comprising: a memory storing an executable program; and a processor configured to run the program, wherein the program, when running, performs the above-mentioned substation inspection method based on the unmanned aerial vehicle cluster.

[0022] According to an embodiment of the present application, a computer readable storage medium is provided, comprising: a computer readable storage medium including a stored executable program, wherein the executable program, when running, controls a device where the storage medium is located to perform the above-mentioned substation inspection method based on the unmanned aerial vehicle cluster.

[0023] According to an embodiment of the present application, a computer program product is provided, comprising: a computer program, which, when executed by a processor, implements the above-mentioned substation inspection method based on the unmanned aerial vehicle cluster.

[0024] In the embodiment of the present application, the method of obtaining the inspection task of the transformer substation, the geographic environment information and the endurance information of each unmanned aerial vehicle in the unmanned aerial vehicle cluster, and inputting the inspection task, the geographic environment information and the endurance information into the task planning model to obtain the target inspection area and the target inspection path of each unmanned aerial vehicle is adopted, the dynamic adjustment of the task allocation and the inspection path planning of each unmanned aerial vehicle in the unmanned aerial vehicle cluster is performed through the optimization of the scheduling and the collaborative operation of the unmanned aerial vehicle cluster, the purpose of automatically performing the inspection task allocation and the inspection path planning according to the complexity of the power line, the geographic environment and the real-time operation condition is achieved, and the technical effects of reducing the complexity of the unmanned aerial vehicle power grid inspection operation and improving the efficiency of the unmanned aerial vehicle power grid inspection operation are realized, and thus the technical problems of high complexity and low efficiency of the unmanned aerial vehicle power grid inspection operation in the related art are solved. BRIEF DESCRIPTION OF DRAWINGS

[0025] The accompanying drawings, which are included to provide a further understanding of the present application and are incorporated in and constitute a part of this application, illustrate embodiments of the present application and serve to explain the present application. In the drawings:

[0026] Figure 1 is a hardware structure block diagram of a transformer substation inspection method based on an unmanned aerial vehicle cluster according to one of the embodiments of the present application;

[0027] Figure 2 is a flowchart of a transformer substation inspection method based on an unmanned aerial vehicle cluster according to one of the embodiments of the present application;

[0028] Figure 3 is a flowchart of another transformer substation inspection method based on an unmanned aerial vehicle cluster according to one of the embodiments of the present application;

[0029] Figure 4 is a structure block diagram of a transformer substation inspection device based on an unmanned aerial vehicle cluster according to one of the embodiments of the present application. DETAILED DESCRIPTION

[0030] In order to enable persons skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings of the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, but not all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by persons skilled in the art without creative labor should fall within the protection scope of the present application.

[0031] It should be noted that the terms "first", "second", and the like in the description and in the claims of the present application and the above-described accompanying drawings are used to distinguish similar objects, and do not necessarily have to be used to describe a specific order or sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments of the application described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device including a series of steps or units does not have to be limited to only those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0032] The UAV cluster-based substation inspection method provided by the embodiments of the present application can be executed in a mobile terminal, a computer terminal, or a similar computing device. Figure 1 is a hardware structure block diagram of a UAV cluster-based substation inspection method according to an embodiment of the present application. As shown in Figure 1 , the computer terminal 10 (or electronic device 10) can include one or more processors (the processor can include but is not limited to a microprocessor MCU or a programmable logic device FPGA processing device), a memory 104 for storing data, and a transmission module 106 for communication functions. In addition, it can also include a display, an input / output interface (I / O interface), a universal serial bus (USB) port (which can be included as one of the ports of the I / O interface), a network interface, a power supply and / or a camera. Those skilled in the art can understand that Figure 1 The structure shown is only schematic, which does not limit the structure of the above-mentioned electronic device. For example, the computer terminal 10 can also include more or fewer components than those shown in Figure 1 , or have a different configuration than Figure 1 .

[0033] It should be noted that the one or more processors and / or other data processing circuits described above can be referred to herein as "data processing circuits" in general. The data processing circuit can be embodied in whole or in part as software, hardware, firmware or any other combination. In addition, the data processing circuit can be a single independent processing module, or any one of the other elements combined into the computer terminal 10 (or electronic device) in whole or in part. As referred to in the embodiments of the present application, the data processing circuit controls as a processor (for example, the selection of the variable resistance terminal path connected to the interface).

[0034] The memory 104 can be used to store software programs of application software and modules, such as program instructions / data storage devices corresponding to the substation inspection method based on the UAV cluster in the embodiments of the present application. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory 104, that is, implements the substation inspection method based on the UAV cluster as described above. The memory 104 can include a high-speed random access memory, and can also include a non-volatile memory, such as one or more magnetic storage devices, flash memories, or other non-volatile solid-state memories. In some examples, the memory 104 can further include a memory remotely arranged with respect to the processor, which can be connected to the computer terminal 10 through a network. Examples of the above-mentioned network include but are not limited to the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0035] The transmission module 106 is used to receive or send data via a network. The specific examples of the above-mentioned network can include a wireless network provided by a communication provider of the computer terminal 10. In one example, the transmission module 106 includes a network adapter (Network Interface Controller, NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission module 106 can be a radio frequency (Radio Frequency, RF) module, which is used to communicate with the Internet in a wireless manner.

[0036] The display can be, for example, a touch screen type liquid crystal display (LCD), which can enable a user to interact with a user page of the computer terminal 10 (or electronic device).

[0037] It should be noted that in some optional embodiments, the above-mentioned Figure 1 The computer device (or electronic device) shown can include hardware elements (including circuitry), software elements (including computer code stored on a computer-readable medium), or a combination of both hardware and software elements. It should be noted that in some embodiments, the functions of the above-mentioned Figure 1 is only one example of a particular specific embodiment, and is intended to show the types of components that can be present in the above-mentioned computer device (or electronic device).

[0038] According to the embodiments of the present application, a method embodiment of a substation inspection method based on a UAV cluster is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described herein can be executed in an order different from that shown herein.

[0039] Figure 2is a flow chart of a substation inspection method based on a UAV cluster according to an embodiment of the present application, as shown in Figure 2 The method comprises the following steps:

[0040] In step S20, the inspection task of the substation, the geographical environment information and the endurance information of each UAV in the UAV cluster are obtained.

[0041] In step S20, the above-mentioned inspection task is used to represent the specific goal and requirement of the substation inspection. For example, which area of the substation needs to be checked, and which specific equipment or structure needs to be paid attention to.

[0042] The above-mentioned geographical environment information is used to represent the geographical environment data of the substation, including but not limited to: terrain, buildings, vegetation, etc.

[0043] The above-mentioned endurance information of each UAV is used to represent the battery status of the UAV, including the power and health status.

[0044] In step S22, the inspection task, the geographical environment information and the endurance information are input into a task planning model to obtain the target inspection area and the target inspection path of each UAV, wherein the task planning model is determined based on a graph diffusion network and a graph convolution network.

[0045] In step S22, the above-mentioned graph diffusion network (GDN) is a network model constructed based on graph theory, which is used to simulate the propagation process of information in the network. The graph diffusion network is used to establish the communication network among each UAV in the UAV cluster, and according to the information sharing among the UAVs, the diffusion ability of each UAV is determined, i.e. the ability of each UAV to handle and transfer tasks in the whole task. In this communication network, each UAV is a node, and the communication or cooperation relationship between them is an edge.

[0046] The above-mentioned graph convolution network (GCN) understands the position and role of each UAV in the power grid inspection by learning the local (neighbors of nodes) and global (entire network) network structure in the UAV cluster. GCN uses graph convolution operation to aggregate the features of the local neighborhood, and combines the features of the global network structure. In this way, GCN can not only capture the relationship between each UAV and its direct neighbors, but also understand the influence of the topology of the entire network on the UAV task.

[0047] Specifically, the task planning model can analyze geographic environment information, identify potential risk factors such as obstacles, weather conditions, etc., to ensure the safe flight of the unmanned aerial vehicle; match the navigation system of the unmanned aerial vehicle with geographic information system (GIS) or global positioning system (GPS) data to ensure accurate positioning of the unmanned aerial vehicle; calculate the expected flight time of the unmanned aerial vehicle according to the load, flight speed and flight state of the unmanned aerial vehicle, evaluate the endurance of the unmanned aerial vehicle, combine the complexity and range of the inspection task, the geographic environment information and the endurance information, assign the inspection task to each unmanned aerial vehicle in the unmanned aerial vehicle cluster, and plan the target inspection path to ensure that the unmanned aerial vehicle can safely return to the base after completing the task. Among them, the control parameters of the unmanned aerial vehicle in the inspection process are determined according to the target inspection task. For example, flight height, speed, shooting frequency, etc.

[0048] Step S24, controlling the unmanned aerial vehicle cluster to inspect the transformer substation according to the target inspection area and the target inspection path to obtain an inspection result, wherein the inspection result is used to determine whether the unmanned aerial vehicle cluster completes the inspection task.

[0049] Specifically, the operator or the automatic control system will guide the unmanned aerial vehicle cluster to inspect the target inspection area according to the target inspection path, and collect data using the cameras, sensors and other devices carried by the unmanned aerial vehicle. The data collected by the unmanned aerial vehicle during the inspection process will be transmitted back to the control center for analysis and evaluation. Through the analysis of the collected data, it can be determined whether the unmanned aerial vehicle cluster has completed all the inspection tasks. If there are omissions or places that need further inspection, the unmanned aerial vehicle cluster will be guided to re-inspect.

[0050] Based on the above steps S20 to S24, the inspection task of the transformer substation, the geographic environment information and the endurance information of each unmanned aerial vehicle in the unmanned aerial vehicle cluster are obtained, and the inspection task, the geographic environment information and the endurance information are input into the task planning model to obtain the target inspection area and the target inspection path of each unmanned aerial vehicle. Through optimizing the scheduling and collaborative work of the unmanned aerial vehicle cluster, the task allocation and the inspection path planning of each unmanned aerial vehicle in the unmanned aerial vehicle cluster are dynamically adjusted. The purpose of automatically assigning the inspection task and planning the inspection path according to the complexity of the power line, the geographic environment and the real-time operation situation is achieved, thereby realizing the technical effect of reducing the complexity of the unmanned aerial vehicle power grid inspection operation and improving the efficiency of the unmanned aerial vehicle power grid inspection operation, and further solving the technical problems of high complexity and low efficiency of the unmanned aerial vehicle power grid inspection operation in the related art.

[0051] Optionally, in step S22, inputting the inspection task, the geographic environment information and the endurance information into the task planning model to obtain the target inspection area and the target inspection path of each unmanned aerial vehicle comprises:

[0052] Step S221, determine the initial inspection area and initial inspection path of each unmanned aerial vehicle based on the inspection task, geographical environment information and endurance information;

[0053] Specifically, the initial inspection area, geographical environment information and endurance information of each unmanned aerial vehicle are analyzed using a path planning algorithm to obtain an analysis result. For example, A* algorithm, Dijkstra algorithm or graph-based path planning algorithm. Based on the analysis result, the initial inspection path of each unmanned aerial vehicle is determined to ensure the efficiency of the path and the endurance of the unmanned aerial vehicle.

[0054] Step S222, determine the diffusion capability expression matrix and role importance expression matrix of the unmanned aerial vehicle cluster based on the initial inspection area and initial inspection path;

[0055] Specifically, based on the initial inspection area and path of the unmanned aerial vehicle, a communication network is constructed, and GDN is used to simulate the information flow transmission process, thereby obtaining the diffusion capability expression matrix of each unmanned aerial vehicle. The diffusion capability expression matrix reflects the capability of the unmanned aerial vehicle in information transmission and task cooperation. The role importance of the unmanned aerial vehicle is analyzed through GCN, a role graph is constructed, and a role importance expression matrix is obtained. The role importance expression matrix represents the role and importance of the unmanned aerial vehicle in the inspection task.

[0056] Step S223, determine the comprehensive score of each unmanned aerial vehicle based on the diffusion capability expression matrix and role importance expression matrix;

[0057] Specifically, the diffusion score and role score of each unmanned aerial vehicle are data fusion processed using a Multilayer Perceptron (MLP) network in combination with the diffusion capability expression matrix and role importance expression matrix to obtain a comprehensive score. The comprehensive score takes into account the performance of the unmanned aerial vehicle in terms of information transmission efficiency and task importance, and is used to evaluate the overall contribution of each unmanned aerial vehicle.

[0058] Step S224, adjust the initial inspection area and initial inspection path according to the comprehensive score to obtain the target inspection area and target inspection path.

[0059] Specifically, the initial inspection area and initial inspection path are adjusted according to the comprehensive score of the unmanned aerial vehicle. High-scoring unmanned aerial vehicles are assigned to more important areas, and low-scoring unmanned aerial vehicles are assigned to relatively less important areas. Through this adjustment, the inspection coverage is optimized to ensure that key areas receive more attention, and the overall efficiency of the inspection task is improved.

[0060] Based on the steps S221 to S224, the initial inspection area and the initial inspection path of each unmanned aerial vehicle are determined based on the inspection task, the geographical environment information and the endurance information; the diffusion capability expression matrix and the role importance expression matrix of the unmanned aerial vehicle cluster are determined based on the initial inspection area and the initial inspection path; the comprehensive score of each unmanned aerial vehicle is determined based on the diffusion capability expression matrix and the role importance expression matrix; the initial inspection area and the initial inspection path are adjusted according to the comprehensive score, and the target inspection area and the target inspection path are obtained, thereby achieving the purpose of automatically performing the inspection task allocation and the inspection path planning according to the complexity of the power line, the geographical environment and the real-time operation condition, so as to realize the technical effects of reducing the complexity of the unmanned aerial vehicle power grid inspection operation and improving the efficiency of the unmanned aerial vehicle power grid inspection operation, and further solve the technical problems of high complexity and low efficiency of the unmanned aerial vehicle power grid inspection operation in the related art.

[0061] Optionally, in step S221, determining the initial inspection area and the initial inspection path of each unmanned aerial vehicle based on the inspection task, the geographical environment information and the endurance information comprises:

[0062] In step S2211, the to-be-inspected area of the substation is divided based on the inspection task and the geographical environment information, and a division result is obtained.

[0063] Specifically, the to-be-inspected area of the substation is logically divided according to the inspection task and the geographical environment information. The division can be based on geographical zoning (such as division according to latitude and longitude), functional zoning (such as transformer area, power transmission line area) or risk level zoning (high risk area and low risk area). For example, according to the inspection task and the geographical environment information, the substation is divided into four areas: A area (high-voltage power transmission line), B area (transformer area), C area (low-voltage power distribution area) and D area (control center and auxiliary facility area). Each area is assigned a different priority according to the importance of the equipment and the complexity of the inspection.

[0064] In step S2212, the unmanned aerial vehicle cluster is allocated tasks according to the division result, and an initial inspection area is obtained.

[0065] Specifically, according to the region division result, the inspection demand and characteristics of each region are determined. Then the performance of each unmanned aerial vehicle in the unmanned aerial vehicle cluster is evaluated, including endurance capability, payload, sensor type, etc. Based on the region division result and the unmanned aerial vehicle capability evaluation, the task allocation of the unmanned aerial vehicle cluster is performed, so that each unmanned aerial vehicle is responsible for a specific inspection region. For example, according to the performance of the unmanned aerial vehicle and the division of the region, unmanned aerial vehicle 1 is allocated to region A (high-voltage transmission line) because it has longer endurance capability and high-performance camera suitable for long-distance flight; unmanned aerial vehicle 2 is allocated to region B (transformer area) because it is equipped with an infrared camera suitable for detecting heat anomalies; unmanned aerial vehicle 3 and unmanned aerial vehicle 4 are responsible for regions C and D respectively.

[0066] In step S2213, an initial inspection path is determined by analyzing the initial inspection region, geographical environment information and endurance information using a path planning algorithm.

[0067] Specifically, a path planning algorithm is selected. For example, A* algorithm, Dijkstra algorithm or graph-based path planning algorithm. The path planning algorithm is used to analyze the initial inspection region, geographical environment information and endurance information of each unmanned aerial vehicle to obtain an analysis result. Based on the analysis result, the initial inspection path of each unmanned aerial vehicle is determined to ensure the efficiency of the path and the endurance capability of the unmanned aerial vehicle. For example, in the path planning of unmanned aerial vehicle 1 in region A, the geographical environment (such as avoiding high mountains and buildings), endurance capability (such as planning a charging point on the way back), and inspection task (such as prioritizing the inspection of key transmission towers) are considered, and the path planning algorithm plans an optimal path from the transformer substation to cover all high-voltage transmission lines. Unmanned aerial vehicle 1 will fly along this path to perform the inspection task and return to the base station for charging when the power allows.

[0068] Based on the above steps S2211 to S2213, the to-be-inspected regions of the transformer substation are divided based on the inspection task and the geographical environment information to obtain a division result; the task allocation of the unmanned aerial vehicle cluster is performed according to the division result to obtain the initial inspection region; and the initial inspection path is determined by analyzing the initial inspection region, geographical environment information and endurance information using a path planning algorithm, which can determine reasonable initial inspection region and initial inspection path for the unmanned aerial vehicle cluster, thereby improving the efficiency and coverage of the inspection, and ensuring the endurance capability and safety of the unmanned aerial vehicle.

[0069] Optionally, in step S222, the diffusion capability expression matrix of the unmanned aerial vehicle cluster is determined based on the initial inspection region and the initial inspection path, which includes:

[0070] In step S2221, a communication network of the unmanned aerial vehicle cluster is established according to the initial inspection region and the initial inspection path, wherein the nodes of the communication network represent each unmanned aerial vehicle in the unmanned aerial vehicle cluster;

[0071] Specifically, a communication network is constructed based on the initial patrol areas and paths of the drones. Each drone is represented as a node in the communication network. The connections between nodes represent the communication paths between drones. Each drone is equipped with a communication module, allowing them to establish direct or indirect communication links with other drones. These communication links can be wireless, such as Wi-Fi, Bluetooth, or a dedicated drone communication protocol. Each node not only processes its own task information, but also transmits this information to surrounding nodes. The information shared between drones includes task status, location data, sensor readings, etc. This information sharing can help drones collaborate to complete tasks, for example, if a drone discovers an anomaly, it can quickly pass this information to other drones so that they can adjust their paths or behaviors. In this process, the communication range and communication capabilities of drones need to be considered to determine whether they can communicate directly or need to relay information through relay drones. For example, drone 1, drone 2, and drone 3 are responsible for different areas, and their flight paths overlap, meaning they may communicate with each other during flight. The flight paths of drone 1 and drone 2 are adjacent and can communicate directly; while drone 3 is far away and needs to relay information through drone 1 or drone 2.

[0072] Step S2222, the information flow transmission process of the communication network is simulated using the graph diffusion network to obtain a diffusion capability expression matrix.

[0073] Specifically, a graph diffusion network (GDN) model is used to simulate the information flow transmission process between drones. In GDN, information starts from the source node and gradually diffuses to other nodes through communication links between drones. This process can simulate the speed and range of information propagation. By simulating the information flow transmission process, a diffusion capability expression matrix can be obtained, which describes the diffusion capability of each drone in the drone swarm, i.e. their role and influence in information transmission. For example, drone 1, drone 2, and drone 3 form a small drone swarm. Through GDN model simulation, we can obtain the diffusion capability expression matrix: In this matrix, the values represent the influence of a drone on the information propagation of other drones. For example, the influence of drone 1 on the information propagation of drone 2 is 0.8, while the influence of drone 2 on drone 1 is 0.8, indicating that their communication is bidirectional and the influence is similar. The influence of drone 3 on drone 1 is smaller (0.5), which may mean that their communication is affected by distance or obstacles.

[0074] Based on the steps S2221 to S2222, the communication network of the UAV cluster is established according to the initial inspection area and the initial inspection path; the information flow transmission process of the communication network is simulated by using the graph diffusion network to obtain a diffusion capability expression matrix, and the importance and role of each UAV in the information transmission process are evaluated by using a quantitative method, so that the inspection task allocation and the communication strategy of the UAVs can be optimized.

[0075] Optionally, in step S222, determining the role importance expression matrix of the UAV cluster based on the initial inspection area and the initial inspection path comprises:

[0076] In step S2223, the initial inspection area and the initial inspection path are used to divide the roles of the UAVs, and the role information of the UAVs is obtained.

[0077] Specifically, the role extraction algorithm (ROIX) is used to automatically extract the roles of the UAVs in the inspection task. The ROIX algorithm analyzes the functional roles of the UAVs according to the inspection area where the UAVs are located and the importance of the adjacent devices. The ROIX algorithm divides the UAVs into different roles by analyzing the number of power grid devices covered by the UAVs, the task complexity, the location, the inspection area where the UAVs are located, and the importance of the adjacent devices. For example, key area inspector, secondary area monitor, emergency responder, etc. The obtained role information will include the specific roles and responsibilities of each UAV, and provide basic data for subsequent role graph construction and role importance evaluation. For example, UAV 1 is responsible for inspecting high-voltage transmission lines, and is therefore divided into a “key area inspector”; UAV 2 is responsible for monitoring the secondary area around the substation, and is divided into a “secondary area monitor”; UAV 3 is an emergency responder, and is ready to assist other UAVs at any time.

[0078] In step S2224, a role graph of the UAV cluster is constructed by using the role information and the initial inspection area, wherein the role graph is used to determine the position information and the importance level of each UAV.

[0079] Specifically, based on the role information extracted by the ROIX algorithm, a role graph is constructed to describe the task relationship and importance between the UAVs. The role graph shows the functional roles and importance of different UAVs in the power grid inspection network, including their positions and importance levels in the inspection task. Through the role graph, the mutual relationship and role of the UAVs in the inspection task are further analyzed to ensure that the key areas and devices are inspected first. For example, the constructed role graph may show that UAV 1 (key area inspector) has more direct connections with other UAVs because it is responsible for the inspection of the most critical part of the power grid; UAV 2 (secondary area monitor) has relatively fewer connections because its role is more independent; and UAV 3 (emergency responder) is in the center of the graph because it needs to be ready to assist other UAVs at any time.

[0080] Step S2225, input the role graph, the initial inspection area and the initial inspection path into the graph convolution network to obtain a role importance expression matrix.

[0081] Specifically, after the role graph is constructed, the role importance of the nodes is further analyzed through the GCN network. The GCN model learns the local and global network structure in the UAV cluster to calculate the importance of each UAV in the power grid inspection. The GCN can generate a role importance score based on the relationship between the UAVs and the complexity of their inspection tasks. For example, through the analysis of the GCN model, the role importance score of UAV 1 is the highest because it is responsible for inspecting the key area that is crucial to the stable operation of the power grid; UAV 3, as an emergency responder, has a relatively high role importance score because it needs to be ready to handle emergencies at any time; the role importance score of UAV 2 is relatively low because it is responsible for a relatively secondary area.

[0082] Based on the above steps S2223 to S2225, the initial inspection area and the initial inspection path are used to divide the roles of the UAVs to obtain the role information of each UAV; the role information and the initial inspection area are used to construct a role graph of the UAV cluster, wherein the role graph is used to determine the position information and the importance level of each UAV; the role graph, the initial inspection area and the initial inspection path are input into the graph convolution network to obtain a role importance expression matrix, which can determine the role importance expression matrix of the UAV cluster, provide decision support for the optimization of the power grid inspection task and the allocation of resources, help to improve the inspection efficiency, ensure that the key areas and equipment are given priority for inspection, and reasonably allocate the UAV resources to improve the overall inspection effect.

[0083] Optionally, in step S223, the task planning model at least includes a multi-layer perceptron network, and determining the comprehensive score of each UAV based on the diffusion capability expression matrix and the role importance expression matrix includes:

[0084] Step S2231, obtaining a diffusion score of each UAV based on the diffusion capability expression matrix;

[0085] Specifically, based on the diffusion capability expression matrix, a diffusion capability score is generated for each UAV. The diffusion score reflects the efficiency of the UAV in information transmission and task allocation. The diffusion capability score takes into account the communication range, information transmission speed, task coverage range and other factors of the UAV. The UAV with a high diffusion score can cooperate more effectively and cover a wider inspection area. For example, UAV 1 has extensive communication links with other UAVs in the diffusion network and can cover multiple key areas, so it obtains a higher diffusion capability score.

[0086] Step S2232, obtaining the role score of each UAV based on the role importance expression matrix;

[0087] Specifically, based on the role importance analysis, the role importance expression matrix is obtained. According to the role importance expression matrix, the role importance score of each UAV is generated. The role importance score reflects whether the role of the UAV in the inspection task is critical and the priority of the UAV in executing the task in the inspection network. The role importance score takes into account the importance of the area responsible by the UAV, the complexity of the task, and the position of the UAV, etc. The UAV with a higher score plays a more important role in the inspection network. For example, UAV 2 is responsible for inspecting a critical transformer area, so it obtains a higher role importance score.

[0088] Step S2233, data fusion processing of the diffusion score and the role score by using a multi-layer perceptron network to obtain a comprehensive score.

[0089] Specifically, in order to determine the target inspection area and the target inspection path of each UAV, the diffusion ability score and the role importance score are fused. The fusion process is completed by the MLP network, and finally the comprehensive task score of each UAV is generated. The MLP network learns the nonlinear relationship between the diffusion score and the role score, and weights and fuses the two scores to generate a comprehensive score. The UAV with a higher comprehensive score will preferentially execute a critical task, while the UAV with a lower score will execute a secondary task or as a supplement to the inspection. For example, the diffusion ability score of UAV 1 is 0.9, and the role importance score is 0.7; the diffusion ability score of UAV 2 is 0.6, and the role importance score is 0.8. Through the fusion processing of the MLP model, the comprehensive score of UAV 1 may be 0.8, and the comprehensive score of UAV 2 may be 0.7. According to the comprehensive score, UAV 1 will preferentially execute a critical task, while UAV 2 will execute a secondary task or as a supplement.

[0090] Based on the above steps S2231 to S2233, the diffusion score of each UAV is obtained based on the diffusion ability expression matrix; the role score of each UAV is obtained based on the role importance expression matrix; the data fusion processing of the diffusion score and the role score by using a multi-layer perceptron network to obtain a comprehensive score. By evaluating the diffusion ability and the role importance of each UAV and combining the scoring and fusion mechanism, the dynamic adjustment of the inspection task not only improves the coverage and efficiency of the inspection, but also ensures that critical areas and equipment are preferentially inspected, thereby improving the safety and reliability of the power grid.

[0091] Optionally, in step S224, the initial inspection area and the initial inspection path are adjusted according to the comprehensive score to obtain the target inspection area and the target inspection path, including:

[0092] Step S2241, obtaining the level information of each sub-inspection area in the division result;

[0093] Specifically, the sub-inspection areas of the substation are divided based on the inspection task and geographical environment information to obtain the division result. Each sub-inspection area in the division result is assigned level information according to its importance, risk level, and historical failure records, etc. These information is collected from the power grid management system, historical inspection data, and field investigation, providing basis for the determination of the level information, wherein the level information can be adjusted in real time according to the importance, risk level, and historical failure records, etc. of the sub-inspection area. The level information is used to distinguish which areas need more frequent or detailed inspection, and which areas can be inspected at a lower frequency. For example, the substation is divided into four areas, A area is a high-voltage transmission line with high level; B area is a transformer area with medium level; C area is a low-voltage distribution area with low level; and D area is a control center and auxiliary facility area with medium level.

[0094] Step S2242, establishing a task allocation mapping relationship based on the level information and the comprehensive score;

[0095] Specifically, a task allocation mapping relationship is established based on the comprehensive score of the unmanned aerial vehicle and the level information of the sub-inspection area. The unmanned aerial vehicle with high score is allocated to the area with high level to ensure that important areas are fully focused and inspected. In this way, the allocation of unmanned aerial vehicle resources is optimized to ensure that the task of each unmanned aerial vehicle matches its ability and importance. For example, unmanned aerial vehicle 1 has the highest comprehensive score and is allocated to A area with high level; unmanned aerial vehicle 2 and unmanned aerial vehicle 3 have lower scores and are allocated to B area and D area with medium level, respectively; and unmanned aerial vehicle 4 has the lowest score and is allocated to C area with low level.

[0096] Step S2243, adjusting the initial inspection area based on the task allocation mapping relationship to obtain the target inspection area;

[0097] Specifically, the initial inspection area is adjusted according to the task allocation mapping relationship to ensure that the area responsible for each unmanned aerial vehicle matches its comprehensive score and area level. Through adjustment, the efficiency and coverage of inspection are improved to ensure that critical areas receive more attention. The inspection area of each unmanned aerial vehicle is re-allocated to be consistent with the established mapping relationship. For example, the initial inspection area of unmanned aerial vehicle 1 may include part of A area and B area, and after adjustment, unmanned aerial vehicle 1 will only be responsible for the inspection of A area to concentrate its ability in the most important area.

[0098] Step S2244, adjusting the initial inspection path based on the target inspection area to obtain the target inspection path.

[0099] Specifically, based on the adjusted target inspection area, the initial inspection path of each unmanned aerial vehicle is adjusted. A path planning algorithm is used to calculate a new target inspection path to improve inspection efficiency. For example, A* algorithm, Dijkstra algorithm, etc. During the path adjustment process, the endurance of the unmanned aerial vehicle and the geographical environment information need to be considered to ensure the feasibility of the path.

[0100] Based on the above steps S2241 to S2244, the level information of each sub-inspection area in the division result is obtained; a task allocation mapping relationship is established based on the level information and the comprehensive score; the initial inspection area is adjusted based on the task allocation mapping relationship to obtain a target inspection area; and the initial inspection path is adjusted based on the target inspection area to obtain a target inspection path, which can intelligently adjust the inspection area and path of the unmanned aerial vehicle, optimize the execution of the inspection task, ensure the efficiency and accuracy of the power grid inspection, improve the coverage and efficiency of the inspection, and ensure that key areas and equipment are given priority for inspection, thereby improving the safety and reliability of the power grid.

[0101] Figure 3 is a flowchart of another substation inspection method based on a cluster of unmanned aerial vehicles according to an embodiment of the present application, as shown in Figure 3 The method comprises the following steps:

[0102] Step S301, obtaining the inspection task of the substation, the geographical environment information, and the endurance information of each unmanned aerial vehicle in the cluster of unmanned aerial vehicles;

[0103] Step S302, dividing the inspection area of the substation based on the inspection task and the geographical environment information to obtain a division result;

[0104] Step S303, task allocation to the cluster of unmanned aerial vehicles according to the division result to obtain an initial inspection area;

[0105] Step S304, analyzing the initial inspection area, the geographical environment information, and the endurance information by using a path planning algorithm to determine an initial inspection path;

[0106] Step S305, establishing a communication network of the cluster of unmanned aerial vehicles according to the initial inspection area and the initial inspection path;

[0107] Step S306, simulating the information flow transmission process of the communication network by using a graph diffusion network to obtain a diffusion capability expression matrix;

[0108] Step S307, obtaining the diffusion score of each unmanned aerial vehicle based on the diffusion capability expression matrix;

[0109] Step S308, role division to each unmanned aerial vehicle by using the initial inspection area and the initial inspection path to obtain role information of each unmanned aerial vehicle;

[0110] Step S309, constructing a role graph of the UAV cluster by using the role information and the initial inspection area;

[0111] Step S310, inputting the role graph, the initial inspection area and the initial inspection path into a graph convolution network to obtain a role importance expression matrix;

[0112] Step S311, obtaining a role score of each UAV based on the role importance expression matrix;

[0113] Step S312, performing data fusion processing on the diffusion score and the role score by using a multilayer perceptron network to obtain a comprehensive score;

[0114] Step S313, obtaining grade information of each sub-inspection area in the division result;

[0115] Step S314, establishing a task allocation mapping relationship based on the grade information and the comprehensive score;

[0116] Step S315, adjusting the initial inspection area based on the task allocation mapping relationship to obtain a target inspection area;

[0117] Step S316, adjusting the initial inspection path based on the target inspection area to obtain a target inspection path;

[0118] Step S317, controlling the UAV cluster to inspect the transformer substation according to the target inspection area and the target inspection path to obtain an inspection result.

[0119] Based on the above steps S301 to S317, the inspection task of the transformer substation, the geographical environment information and the endurance information of each UAV in the UAV cluster are obtained, and the inspection task, the geographical environment information and the endurance information are input into a task planning model to obtain the target inspection area and the target inspection path of each UAV. Through optimizing the scheduling and collaborative operation of the UAV cluster, the task allocation and the inspection path planning of each UAV in the UAV cluster are dynamically adjusted. The purpose of automatically performing the inspection task allocation and the inspection path planning according to the complexity of the power line, the geographical environment and the real-time operation condition is achieved. Therefore, the technical effects of reducing the complexity of the UAV power grid inspection operation and improving the efficiency of the UAV power grid inspection operation are achieved. Thus, the technical problems of high complexity and low efficiency of the UAV power grid inspection operation in the related art are solved.

[0120] Through the description of the above embodiments, those skilled in the art can clearly understand that the method according to the above embodiments can be realized by means of software on a general hardware platform as necessary, and of course can also be realized by hardware, but in many cases the former is a better embodiment. Based on such understanding, the technical solutions of the present application can be embodied in the form of a software product in essence or in the part that contributes to the prior art, and the computer software product is stored in a storage medium (such as a ROM / RAM, a magnetic disk, or an optical disk), and includes a plurality of instructions for causing an end device (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the method of each embodiment of the present application.

[0121] In the embodiments of the present application, a substation inspection device based on a UAV cluster is also provided, which is used to implement the above-mentioned embodiments and preferred embodiments, and will not be described again. As used below, the term "module" can be a combination of software and / or hardware that implements a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware, or a combination of software and hardware is also possible and contemplated.

[0122] Figure 4 is a structural block diagram of a substation inspection device based on a UAV cluster according to one of the embodiments of the present application, as shown in Figure 4 , the device comprises:

[0123] The acquisition module 401 is configured to acquire an inspection task of a substation, geographical environment information, and endurance information of each UAV in a UAV cluster.

[0124] The determination module 402 is configured to input the inspection task, the geographical environment information, and the endurance information into a task planning model to obtain a target inspection area and a target inspection path of each UAV, wherein the task planning model is determined based on a graph diffusion network and a graph convolution network.

[0125] The control module 403 is configured to control the UAV cluster to perform inspection on the substation according to the target inspection area and the target inspection path to obtain an inspection result, wherein the inspection result is used to determine whether the UAV cluster completes the inspection task.

[0126] Optionally, the determination module 402 is further configured to determine an initial inspection area and an initial inspection path of each UAV based on the inspection task, the geographical environment information, and the endurance information; determine a diffusion ability expression matrix and a role importance expression matrix of the UAV cluster based on the initial inspection area and the initial inspection path; determine a comprehensive score of each UAV based on the diffusion ability expression matrix and the role importance expression matrix; and adjust the initial inspection area and the initial inspection path according to the comprehensive score to obtain the target inspection area and the target inspection path.

[0127] Optionally, the determining module 402 is further configured to divide the to-be-inspected area of the transformer substation based on the inspection task and the geographic environment information to obtain a division result, and perform task allocation on the UAV cluster according to the division result to obtain an initial inspection area, and analyze the initial inspection area, the geographic environment information and the endurance information by using a path planning algorithm to determine an initial inspection path.

[0128] Optionally, the transformer substation inspection device based on the UAV cluster further comprises an establishing module 404 configured to establish a communication network of the UAV cluster according to the initial inspection area and the initial inspection path, wherein a node of the communication network represents each UAV in the UAV cluster, and the transformer substation inspection device based on the UAV cluster further comprises a simulation module configured to simulate an information flow transmission process of the communication network by using a graph diffusion network to obtain a diffusion capability expression matrix.

[0129] Optionally, the determining module 402 is further configured to perform role division on each UAV by using the initial inspection area and the initial inspection path to obtain role information of each UAV, and the establishing module 404 is further configured to construct a role graph of the UAV cluster by using the role information and the initial inspection area, wherein the role graph is used to determine position information and an importance level of each UAV, and the determining module is further configured to input the role graph, the initial inspection area and the initial inspection path into a graph convolution network to obtain a role importance expression matrix.

[0130] Optionally, the obtaining module 401 is further configured to obtain a diffusion score of each UAV based on the diffusion capability expression matrix, and obtain a role score of each UAV based on the role importance expression matrix, and the determining module 402 is further configured to perform data fusion processing on the diffusion score and the role score by using a multilayer perceptron network to obtain a comprehensive score.

[0131] Optionally, the obtaining module 401 is further configured to obtain level information of each sub-to-be-inspected area in the division result, and the establishing module is further configured to establish a task allocation mapping relationship based on the level information and the comprehensive score, and the determining module 402 is further configured to adjust the initial inspection area based on the task allocation mapping relationship to obtain a target inspection area, and adjust the initial inspection path based on the target inspection area to obtain a target inspection path.

[0132] It should be noted that each of the above modules can be implemented by software or hardware, and for the latter, the following implementation manners can be used, but are not limited thereto: all the above modules are located in the same processor; or the above modules are located in different processors in any combination.

[0133] According to one of the embodiments of the present application, an electronic device is further provided, comprising: a memory storing an executable program; and a processor configured to run the program, wherein the program performs the above-mentioned transformer substation inspection method based on the UAV cluster when running.

[0134] Optionally, in the embodiment, the processor can be configured to execute the following steps by a computer program:

[0135] Step S1, obtaining the inspection task of the transformer substation, the geographical environment information and the endurance information of each unmanned aerial vehicle in the unmanned aerial vehicle cluster;

[0136] Step S2, inputting the inspection task, the geographical environment information and the endurance information into a task planning model to obtain the target inspection area and the target inspection path of each unmanned aerial vehicle, wherein the task planning model is determined based on a graph diffusion network and a graph convolution network;

[0137] Step S3, controlling the unmanned aerial vehicle cluster to perform the inspection on the transformer substation according to the target inspection area and the target inspection path to obtain the inspection result, wherein the inspection result is used to determine whether the unmanned aerial vehicle cluster completes the inspection task.

[0138] According to one of the embodiments of the present application, a computer readable storage medium is also provided, comprising: the computer readable storage medium comprises a stored executable program, wherein when the executable program is running, the device where the storage medium is located is controlled to execute the above-mentioned transformer substation inspection method based on the unmanned aerial vehicle cluster.

[0139] Optionally, in the embodiment, the storage medium can be configured to store a computer program for executing the following steps:

[0140] Step S1, obtaining the inspection task of the transformer substation, the geographical environment information and the endurance information of each unmanned aerial vehicle in the unmanned aerial vehicle cluster;

[0141] Step S2, inputting the inspection task, the geographical environment information and the endurance information into a task planning model to obtain the target inspection area and the target inspection path of each unmanned aerial vehicle, wherein the task planning model is determined based on a graph diffusion network and a graph convolution network;

[0142] Step S3, controlling the unmanned aerial vehicle cluster to perform the inspection on the transformer substation according to the target inspection area and the target inspection path to obtain the inspection result, wherein the inspection result is used to determine whether the unmanned aerial vehicle cluster completes the inspection task.

[0143] Optionally, in the embodiment, the storage medium can include but is not limited to: a U disk, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk and various computer program storage media.

[0144] According to one of the embodiments of the present application, a computer program product is also provided, comprising: a computer program, which, when executed by a processor, implements the above-mentioned substation inspection method based on a UAV cluster.

[0145] Optionally, in the embodiment, the above-mentioned computer program product can be configured to store computer program instructions for executing the following steps:

[0146] Step S1, obtaining an inspection task of a substation, geographical environment information, and endurance information of each UAV in a UAV cluster;

[0147] Step S2, inputting the inspection task, the geographical environment information, and the endurance information into a task planning model to obtain a target inspection area and a target inspection path of each UAV, wherein the task planning model is determined based on a graph diffusion network and a graph convolution network;

[0148] Step S3, controlling the UAV cluster to perform inspection on the substation according to the target inspection area and the target inspection path to obtain an inspection result, wherein the inspection result is used to determine whether the UAV cluster completes the inspection task.

[0149] Optionally, specific examples in the embodiment can refer to examples described in the above-mentioned embodiments and optional implementation manners, and the embodiment will not be described here.

[0150] The serial numbers of the above-mentioned embodiments of the present application are only for description, and do not represent the advantages and disadvantages of the embodiments.

[0151] In the above-mentioned embodiments of the present application, the description of each embodiment has its own emphasis, and the parts not described in detail in a certain embodiment can be referred to the related description of other embodiments.

[0152] In several embodiments provided by the present application, it should be understood that the disclosed technology can be implemented in other ways. Of course, the unit described as the division is only a logical function division, and there can be other division manners in actual implementation, for example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the units shown or discussed can be indirect coupling or communication connection through some interfaces, units or modules, and can be electrical or other forms.

[0153] The units described as separate components can or can not be physically separate, and the components shown as units can or can not be physical units, that is, they can be located in one place, or can be distributed on multiple units. According to actual needs, part or all of the units can be selected to achieve the purpose of the embodiment.

[0154] In addition, each functional unit in each embodiment of the present application can be integrated in one processing unit, or each unit can be physically present separately, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of a software functional unit.

[0155] When the integrated unit is realized in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application, essentially or the part that contributes to the prior art, or all or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in each embodiment of the present application. The foregoing storage medium includes: a U disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a mobile hard disk, a magnetic disk or an optical disk, and various media that can store program codes.

[0156] The above is only the preferred embodiment of the present application, and it should be pointed out that for those skilled in the art, without departing from the principles of the present application, a number of improvements and refinements can be made, and these improvements and refinements should be considered as the protection scope of the present application.

Claims

1. A substation inspection method based on a UAV cluster, characterized in that, The application comprises the following steps: acquiring an inspection task of a substation, geographical environment information, and endurance information of each unmanned aerial vehicle in a cluster of unmanned aerial vehicles; inputting the inspection task, the geographical environment information, and the endurance information into a task planning model to obtain a target inspection area and a target inspection path of each unmanned aerial vehicle, wherein the task planning model is determined based on a graph diffusion network and a graph convolution network; controlling the cluster of unmanned aerial vehicles to inspect the substation according to the target inspection area and the target inspection path to obtain an inspection result, wherein the inspection result is used to determine whether the cluster of unmanned aerial vehicles completes the inspection task; wherein inputting the inspection task, the geographical environment information, and the endurance information into the task planning model to obtain the target inspection area and the target inspection path of each unmanned aerial vehicle comprises: determining an initial inspection area and an initial inspection path of each unmanned aerial vehicle based on the inspection task, the geographical environment information, and the endurance information; determining a diffusion capability expression matrix and a role importance expression matrix of the cluster of unmanned aerial vehicles based on the initial inspection area and the initial inspection path; determining a comprehensive score of each unmanned aerial vehicle based on the diffusion capability expression matrix and the role importance expression matrix; and adjusting the initial inspection area and the initial inspection path according to the comprehensive score to obtain the target inspection area and the target inspection path. 2.The unmanned aerial vehicle cluster-based substation inspection method of claim 1, wherein, determining the initial inspection area and the initial inspection path of each unmanned aerial vehicle based on the inspection task, the geographical environment information, and the endurance information comprises: dividing a to-be-inspected area of the substation based on the inspection task and the geographical environment information to obtain a division result; performing task allocation on the cluster of unmanned aerial vehicles according to the division result to obtain the initial inspection area; analyzing the initial inspection area, the geographical environment information, and the endurance information by using a path planning algorithm to determine the initial inspection path. 3.The unmanned aerial vehicle cluster-based substation inspection method of claim 1, wherein, determining the diffusion capability expression matrix of the cluster of unmanned aerial vehicles based on the initial inspection area and the initial inspection path comprises: establishing a communication network of the cluster of unmanned aerial vehicles according to the initial inspection area and the initial inspection path, wherein a node of the communication network represents each unmanned aerial vehicle in the cluster of unmanned aerial vehicles; simulating an information flow transmission process of the communication network by using the graph diffusion network to obtain the diffusion capability expression matrix. 4.The unmanned aerial vehicle cluster-based substation inspection method of claim 1, wherein, determining the role importance expression matrix of the cluster of unmanned aerial vehicles based on the initial inspection area and the initial inspection path comprises: performing role division on each unmanned aerial vehicle by using the initial inspection area and the initial inspection path to obtain role information of each unmanned aerial vehicle; constructing a role graph of the cluster of unmanned aerial vehicles by using the role information and the initial inspection area, wherein the role graph is used to determine position information and an importance level of each unmanned aerial vehicle; inputting the role graph, the initial inspection area, and the initial inspection path into the graph convolution network to obtain the role importance expression matrix. 5.The unmanned aerial vehicle cluster-based substation inspection method of claim 1, wherein, The task planning model at least comprises a multi-layer perceptron network, and determining the comprehensive score of each unmanned aerial vehicle based on the diffusion capability expression matrix and the role importance expression matrix comprises: obtaining a diffusion score of each unmanned aerial vehicle based on the diffusion capability expression matrix; obtaining a role score of each unmanned aerial vehicle based on the role importance expression matrix; performing data fusion processing on the diffusion score and the role score by using the multi-layer perceptron network to obtain the comprehensive score. 6.The unmanned aerial vehicle cluster-based substation inspection method of claim 2, wherein, Adjusting the initial inspection area and the initial inspection path according to the comprehensive score to obtain the target inspection area and the target inspection path comprises: obtaining grade information of each sub-inspection area in the division result; establishing a task allocation mapping relationship based on the grade information and the comprehensive score; adjusting the initial inspection area based on the task allocation mapping relationship to obtain the target inspection area; adjusting the initial inspection path based on the target inspection area to obtain the target inspection path.

7. An electronic device, comprising: comprise: a memory storing an executable program; a processor configured to execute the program, wherein the program, when executed, performs the method for substation inspection based on a cluster of unmanned aerial vehicles according to any one of claims 1 to 6.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium comprises a stored executable program, wherein the executable program, when executed, controls the device in which the storage medium is located to perform the method for substation inspection based on a cluster of unmanned aerial vehicles according to any one of claims 1 to 6.

9. A computer program product, characterised in that, comprise a computer program that, when executed by a processor, implements the method for substation inspection based on a cluster of unmanned aerial vehicles according to any one of claims 1 to 6.

Images