Unmanned aerial vehicle power grid inspection communication resource allocation method and system

Abstract

The invention discloses an unmanned aerial vehicle power grid inspection communication resource allocation method and system, and aims to reduce the total communication cost and improve the environmental friendliness and operation efficiency of a power system. The system comprises a task collection module, a coordination module, a relay node processing module and a monitoring and feedback module. According to the method, the flexibility of the unmanned aerial vehicle in inspection is fully utilized, the heterogeneous computing power network is combined, dynamic scheduling is performed by adopting an optimization algorithm, and efficient distribution of calculation tasks is realized; and the flight energy consumption of the unmanned aerial vehicle, communication resources and the load condition of the relay node are fully considered, and an optimal task allocation scheme is generated, so that the communication cost of the system is minimized. According to the method, the efficiency and quality of power grid inspection are improved, reasonable utilization of resources and reduction of energy consumption are effectively promoted, the method has important theoretical and practical significance, and transformation of the society to low energy consumption is promoted.

Description

Technical Field

[0001] The present invention relates to the field of drone inspection, and in particular to a method and system for allocating communication resources for drone power grid inspection, aiming to provide a method and system that can effectively allocate drone computing tasks to reduce the total system communication cost. Background Art

[0002] As the demand for equipment inspection and maintenance in power systems increases, drones, as an efficient inspection tool, are gradually being used in the monitoring and maintenance of power equipment. Drones have the advantages of high flexibility, wide coverage, and low cost, and can collect real-time data in complex environments. However, with the increase in the number of drones and the increase in task complexity, how to effectively allocate the computing tasks of drones has become a problem that needs to be solved.

[0003] In the power system, drones will pass through multiple relay nodes during flight, including BSs (BaseStations) and GNs (Ground Nodes). Drones pass computing tasks to these relay nodes through two methods: licensed band transmission (LBT) and unlicensed band transmission (UBT). After receiving the computing tasks, the relay nodes will send them to the cloud or edge server for computing task processing.

[0004] Traditional computing task allocation methods mostly rely on centralized management systems, which may be effective in handling simple tasks, but often seem to be inadequate when faced with dynamically changing power systems and multi-UAV collaboration. In addition, existing technologies fail to fully consider the optimization of communication costs during task allocation, resulting in low system efficiency and unnecessary communication overhead. Summary of the invention

[0005] In view of this, the present invention provides a method and system for allocating communication resources for UAV power grid inspection, aiming to reduce the total communication cost of the system through an efficient task allocation mechanism, and improve the environmental friendliness and operating efficiency of the power system. This method makes full use of the advantages of heterogeneous computing power networks, combines the fixed flight routes of UAVs with the communication resources of intermediate nodes, and realizes the optimal allocation of computing tasks during the UAV inspection process, thereby achieving the goal of minimizing communication costs.

[0006] In order to achieve the above object, the present invention provides a UAV power grid inspection communication resource allocation system, comprising:

[0007] Task collection module, coordination module, relay node processing module and monitoring and feedback module;

[0008] The task collection module is used to collect computing tasks related to IoT terminal devices in the drone coverage area during the inspection process;

[0009] The coordination module includes an optimization algorithm module, a task allocation module and a communication resource allocation module, wherein:

[0010] The optimization algorithm module is used to establish an energy consumption optimization model for the power system UAV inspection system. The energy consumption optimization model minimizes the energy consumption of the UAV inspection system by rationally allocating the computing tasks carried by the UAV to the relay nodes. The task allocation scheme and the communication resource allocation scheme are obtained by solving the energy consumption optimization model. The task allocation scheme determines the transmission mode selected by each UAV to the relay node on the path and the computing tasks sent, which can minimize the total communication cost. The communication resource allocation scheme includes each relay node evaluating the computing tasks that can be accepted based on its own remaining computing resource information;

[0011] The task allocation module is used to collect current network status information in real time, identify relay nodes that have communication capabilities and are on the flight route of the drone, and transmit computing tasks to the identified relay nodes according to the task allocation plan;

[0012] The communication resource allocation module is used to dynamically evaluate the actual communication resource situation of each relay node and adjust the task allocation plan according to the communication resource allocation plan;

[0013] The relay node processing module is used to transmit the computing task received from the coordination module to the upper cloud or edge server for computing;

[0014] The monitoring and feedback module is used to receive communication network status information in real time and feed it back to the coordination module.

[0015] Furthermore, during the task collection process, the drone will regularly update its power status, and when the battery power is lower than the set threshold, it will automatically adjust the inspection plan and select the optimal route to return to the ground node as soon as possible for battery replacement or charging.

[0016] Furthermore, after collecting the computing tasks, the UAV collects the location information of the relay nodes in the inspection area, the maximum task reception and the energy consumption indicators to provide basic data for the UAV's path planning. The path planning module establishes an energy consumption model, considering the flight energy consumption and the consumption required for communication, and evaluates the energy consumption under different flight conditions; at the same time, by constructing a communication cost model, the communication cost of each UAV is evaluated.

[0017] Furthermore, the energy consumption optimization model models the energy consumption of the drone inspection system when uploading data to the relay node using two transmission modes, the authorized frequency band and the unauthorized frequency band. The optimization algorithm uses the alternating direction multiplier method to solve the optimization problem. The optimization problem is as follows:

[0018]

[0019] Among them, e (1) (x) is the system communication cost using unlicensed frequency bands, e (2) (y) is the system communication cost using the licensed frequency band; l , η n are the maximum data acceptance capacity of GN l and BS n, respectively, i represents the total amount of data sent by UAV i within τ time, ε i Indicates the maximum energy consumption allowed by the drone's battery;

[0020] The alternating direction multiplier method divides the global optimization problem into N sub-problems, each of which is solved by a drone using its private information. Define g ij = <x il ,y in > is the amount of data sent by drone i to relay node j, nj=<ζ l ,η n > represents the maximum data receiving capability set of the relay node, and the optimization problem is simplified to:

[0021]

[0022] Define the feasible set of the i-th UAV The indicator function of the i-th UAV is defined as follows:

[0023]

[0024] Similarly, for other constraints in the joint optimization problem, define the feasible domain Then the indicator function is:

[0025]

[0026] By converting the constraints in the optimization problem into an indicator function, the function in the form of ADMM is obtained as follows:

[0027]

[0028] stg-n=0

[0029] in H(n)=ID (g).

[0030] The final result is an optimized g value, which corresponds to the minimum energy consumption value.

[0031] Furthermore, the current network status information collected by the task allocation module includes the geographical location, load status and communication capability of each relay node, and the relay nodes include base stations BSs and ground nodes GNs.

[0032] Furthermore, the communication resource allocation module is specifically used to dynamically evaluate the actual communication resource situation of each relay node by continuously acquiring the current load status, network bandwidth and processing capacity of each relay node, analyze the communication resources that the relay node can use for task processing within a specific time, and then adjust the task allocation plan to avoid the occurrence of overload.

[0033] A method for allocating communication resources for unmanned aerial vehicle power grid inspection, which is applied to the unmanned aerial vehicle power grid inspection communication resource allocation system described in the claims, the method comprising:

[0034] Step S1, real-time acquisition of the communication resources and current task load information of each relay node during the drone inspection process to ensure real-time monitoring of the working status of each relay node.

[0035] Step S2, the task collection module collects computing tasks related to the Internet of Things terminal devices within the coverage area of ​​the drone through the drone during the inspection process.

[0036] Step S3, the optimization algorithm module of the coordination module establishes an energy consumption optimization model for the power system UAV inspection system, the energy consumption optimization model minimizes the energy consumption of the UAV inspection system by rationally allocating the computing tasks carried by the UAV to the relay nodes, and obtains the task allocation scheme and the communication resource allocation scheme by solving the energy consumption optimization model. The task allocation scheme determines the transmission mode selected by each UAV to the relay node on the path and the computing tasks sent, which can minimize the total communication cost, and the communication resource allocation scheme includes each relay node evaluating the computing tasks that can be accepted according to its own remaining computing resource information;

[0037] The task allocation module of the coordination module collects the current network status information in real time, identifies the relay nodes with communication capabilities and on the flight route of the UAV, and transmits the computing tasks to the identified relay nodes according to the task allocation plan;

[0038] The communication resource allocation module of the coordination module dynamically evaluates the actual communication resource situation of each relay node and adjusts the task allocation plan according to the communication resource allocation plan;

[0039] Step S4, the relay node processing module transmits the computing task received from the coordination module to the upper cloud or edge server for computing;

[0040] Step S5: The monitoring and feedback module receives the communication network status information in real time and feeds it back to the coordination module.

[0041] Compared with the prior art, the present invention has the following characteristics:

[0042] 1. Comprehensive consideration of environmental factors: The present invention fully and comprehensively considers the environmental factors of the UAV during the execution of the mission, including the difference in energy costs of different relay nodes, to generate a more flexible and reasonable task allocation plan and improve the efficiency and environmental protection of the system.

[0043] 2. Real-time adjustment of task allocation: By monitoring the flight status and task status of the UAV, the present invention can flexibly adjust task allocation according to real-time communication cost data, thereby ensuring that the system budget is not exceeded when executing computing tasks and promoting the reasonable transfer of energy consumption among different relay nodes.

[0044] 3. Reduce communication costs: The present invention optimizes the task allocation mechanism through an algorithm, effectively reducing the total communication cost of the system, making the transmission of tasks from drones to intermediate nodes more efficient and achieving optimal configuration of system resources.

[0045] By efficiently allocating computing tasks during drone inspections, the present invention solves the problem of uncontrollable communication costs during drone inspections. It comprehensively considers energy costs, system status, and node requirements, obtains relay node load information through real-time monitoring, and generates an optimal task allocation plan in combination with the data traffic carried by the drone, with the goal of minimizing the total communication cost of the system. This method flexibly responds to different inspection routes of drones, effectively utilizes the available resources of relay nodes, and improves system efficiency and environmental protection. The present invention minimizes communication costs during inspections through algorithm optimization and task allocation, which is of great significance and value to improving the computing efficiency of drone networks and promoting the transformation of the entire society to low energy consumption. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 This is a structural schematic diagram of a UAV power grid inspection communication resource allocation system according to an embodiment of the present invention;

[0047] Figure 2 A schematic diagram of the flow chart of the alternating direction method of multipliers (ADMM) algorithm provided by the present invention;

[0048] Figure 3 This is a flow chart of a method for allocating communication resources for UAV power grid inspection according to an embodiment of the present invention;

[0049] Figure 4The total energy consumption comparison diagram of different transmission methods is shown in Figure 1, where (a) is the communication cost comparison diagram of different transmission methods, and (b) is the task volume comparison diagram of the UAV using LBT and UBT transmission in the MCS scheme;

[0050] Figure 5 This is a comparison chart of the number of iterations of the ADMM algorithm proposed in the present invention and the commonly used interior point method and subgradient method. DETAILED DESCRIPTION

[0051] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0052] See also Figure 1 , an embodiment of the present invention provides a UAV power grid inspection communication resource allocation system, including a task collection module, a coordination module, a relay node processing module and a monitoring and feedback module.

[0053] The task collection module is used to collect computing tasks related to the IoT terminal devices in the coverage area of ​​the drone during the inspection process. During the inspection process, the drone will move according to a pre-planned fixed flight route. In order to ensure the integrity and accuracy of the task collection, the drone will monitor its own battery power in real time during the flight to optimize energy management and ensure the smooth completion of the task. The drone will fly to a predetermined node position at a certain time interval (for example, every hour), so that it can regularly access and retrieve the computing tasks of all IoT terminal devices within the coverage area of ​​the node. Drones usually use wireless communication technology (such as Wi-Fi, LTE or 5G) to connect with IoT terminal devices, receive task instructions in real time and send data. All received computing task information will be cached and recorded in the internal storage system of the drone for use by subsequent task allocation modules and communication resource allocation modules. In this process, the module will also record the priority and requirements of each task to ensure that the actual status of the node can be combined in the next task allocation. Priority allocation.

[0054] During the task collection process, the drone will regularly update its power status, and when the battery power is lower than the set threshold, it will automatically adjust the inspection plan and select the optimal route to return to the ground node as soon as possible for battery replacement or charging, thereby maintaining the efficient operation of the system. After the drone collects the computing task, it needs to transmit the computing task. First, by collecting the location information, maximum task reception and energy consumption indicators of the relay nodes in the inspection area, basic data is provided for the drone's path planning module. The path planning module establishes a detailed energy consumption model, taking into account the two aspects of flight energy consumption and communication consumption, and accurately evaluates the energy consumption under different flight conditions. At the same time, by constructing a communication cost model, the communication cost of each drone is evaluated.

[0055] The coordination module includes an optimization algorithm module, a task allocation module and a communication resource allocation module. After collecting the computing tasks, the drone sends them to the coordination module. The optimization algorithm module first establishes an energy consumption optimization model of the power system drone inspection system, that is, the energy consumption of the drone inspection system when uploading data to the relay node using two transmission modes, the authorized frequency band and the unauthorized frequency band, is modeled.

[0056] In view of the energy consumption optimization problem, the optimization algorithm proposes an alternating direction multiplier method to solve the optimization problem, which can minimize the energy consumption of the drone inspection system by rationally allocating the computing tasks carried by the drone to the relay nodes. The result of the optimization problem generates a task allocation plan (how many computing tasks each drone sends to each relay node on the path through which transmission method) and a communication resource allocation plan (how many computing tasks each relay node can accept based on its own remaining computing resource information) and is implemented through the corresponding modules.

[0057] The task allocation module collects the current network status information in real time, including the geographical location, load and communication capacity of each relay node (base station (BSs) and ground node (GNs)). By analyzing these data, the module can identify the relay nodes with communication capabilities and on the flight path of the drone, and allocate computing tasks to these relay nodes according to the task allocation scheme to avoid network congestion.

[0058] The communication resource allocation module plays a key role in the entire drone inspection system. Its main function is to monitor the communication resources owned by each relay node in real time to ensure that the amount of tasks it receives does not exceed the maximum task receiving capacity. This module can dynamically evaluate the actual communication resource situation of each relay node by continuously obtaining information such as the current load status, network bandwidth, and processing power of each relay node. Specifically, the communication resource allocation module will analyze the communication resources that the relay node can use for task processing within a specific time, and adjust the task allocation plan in time according to these data to avoid overload.

[0059] In the UAV inspection system, each UAV can transmit computing tasks to relay nodes (GNs, BSs) through two methods: licensed frequency band LBT and unlicensed frequency band UBT.

[0060] The unlicensed band, that is, each drone uses a random access to the unlicensed band, and must follow a certain channel access mechanism to avoid conflicts with other coexisting technologies. This mechanism allows each device to dynamically adjust the transmission time and period according to the availability of the channel. Generally, two free bands, 2.4GHz and 5GHz ISM, are used. Each device follows the CSMA / CA protocol to send data signals to support wireless services such as Wi-Fi, ZigBee, Thread, ZWave and Wi-SUN. In these systems, a group of ground nodes (GNs) have been deployed to relay data from drones to edge servers. In CSMA / CA, each device needs to execute a channel access mechanism based on "listen before talking" to avoid collisions with other co-located transmitters.

[0061] Assuming that the average energy consumed by each drone i is δi, the probability of successful transmission in the unlicensed frequency band is P i , that is, the chance of the channel being occupied by other devices or coexisting devices is 1-P i In this case, the drones cannot successfully occupy the channel and therefore do not consume any energy for data transmission. After successfully gaining access to the unauthorized channel, each drone can send its data to nearby GNs. To represent the set of GNs that can receive data traffic for the system. Within τ time, UAV i sends x il bits of data to GN l, and the data traffic λ is sent through the drone i The energy consumption is:

[0062]

[0063] In the formula, B T is the bandwidth of the unlicensed band, h il is the channel gain between UAV i and GN l, σ il is the noise power received at GN l. and Set τ = 1, that is, is equal to the transmission power of UAV i, and its expected energy consumption is:

[0064]

[0065] c i Defined as the conversion rate between energy consumption and cost, the communication cost of the system is:

[0066]

[0067] Where x = {x i} i∈N .

[0068] Regarding the licensed frequency band, 3GPP has proposed a variety of IoT solutions such as NB-IoT, EC-GPRS, LTE eMTC, etc., all of which operate on licensed spectrum that has been exclusively allocated to mobile operators. Mobile operators have carefully deployed their network infrastructure to minimize interference between cells. Each device does not have to sense the channel, but will follow the MNO's channel access schedule. Unlike unlicensed band services, MNOs typically charge mobile devices based on the total amount of data traffic passing through their networks. Since the use of licensed frequency bands must always be fully controlled and monitored by mobile network operators, QoS can always be guaranteed. Each drone signs a contract with a mobile operator, allowing each drone to offload its data traffic to neighboring BSs. Let B represent the set of BSs. At this time, the energy consumption of each drone is:

[0069]

[0070] In the formula, y in Determine the number of data bits that UAV i sends to BS via the licensed frequency band, y i ={y in} n∈Β , and∑ n∈Β {y in}=λ i . Let β n is the price charged by the MNO for each bit of data sent through the BS. The sum of the system communication cost and infrastructure rental cost of this solution is:

[0071]

[0072] The present invention assumes that each drone can use one or more of the above options to send data. Each drone can divide the total amount of data it collects into multiple parts and send them through UBT, LBT or a combination of the two, that is, using Multi-Connectivity Sharing (MCS) to send data. The task allocation scheme needs to determine the data flow sent by each method to minimize the data communication cost of the drone inspection system, so the optimization problem is as follows:

[0073]

[0074] Among them, l , η nare the maximum data acceptance capacity of GN l and BS n, respectively, i represents the total amount of data sent by UAV i within τ time, ε i Indicates the maximum energy consumption allowed by the drone's battery;

[0075] The optimization algorithm is the alternating direction method of multipliers (ADMM) algorithm, which accepts computing tasks from the task collection module, uses the known information of the system to optimize the algorithm, and obtains the corresponding task allocation plan and communication resource allocation plan. The drone sends the computing tasks to the corresponding relay nodes through two plans: task allocation plan and communication resource allocation plan, and finally completes the computing tasks.

[0076] Figure 2 Figure 1 is a flow chart of the ADMM algorithm. In order to solve the above optimization problem, it is necessary to understand the overall information of the network, such as the maximum amount of data transmitted by each drone per unit time and the maximum amount of data received by each relay node. In order to coordinate and solve the optimization problem at a faster convergence speed under the condition of protecting privacy, the present invention proposes a distributed ADMM algorithm. In this algorithm, the global optimization problem can be divided into N sub-problems, and each sub-problem can be solved by the drone using its private information. For the convenience of expression, define g ij = <x il ,y in > is the amount of data sent by drone i to relay node j, nj=<ζ l ,η n > represents the maximum data receiving capability set of the relay node. The optimization problem can be simplified as:

[0077]

[0078] Define the feasible set of the i-th UAV The indicator function of the i-th UAV is defined as follows:

[0079]

[0080] Similarly, for other constraints in the joint optimization problem, define the feasible domain Then the indicator function is:

[0081]

[0082] By converting the constraints in the optimization problem into an indicator function, the function in the form of ADMM is obtained as follows:

[0083]

[0084] stg-n=0

[0085] Said H(n)=I D (g).

[0086] The final result is an optimized g value, which corresponds to the minimum energy consumption. The ADMM algorithm is used to calculate how to reasonably distribute the amount of computing task data transmitted by the drone to each relay node to minimize the final communication cost, that is, the energy consumption e(x)+e(y).

[0087] The relay node processing module mainly transmits the received computing tasks to the corresponding edge server for computing. Assume that each edge server in the system is connected to a subset of relay nodes J close to it in the spatial range.<l,n> , and only receives data traffic from a subset of nodes.

[0088] The monitoring and feedback module is used to receive communication network status information in real time. In this embodiment, the monitoring and feedback module obtains the communication network status information of the communication network at the current moment, and the communication network status information includes the remaining battery capacity of each drone in the network and the amount of data received by the relay node at the current moment. These communication network status information will be sorted and analyzed, and fed back to the coordination module, so that the coordination module can make more accurate communication resource allocation decisions based on real-time data. In this way, the coordination module can adjust the task allocation of the drone in a timely manner to optimize network performance and extend the operation time of the drone.

[0089] like Figure 3 As shown, an embodiment of the present invention further provides a method for allocating communication resources for unmanned aerial vehicle power grid inspection, which is applied to the unmanned aerial vehicle power grid inspection communication resource allocation system described in an embodiment of the present invention, and the method includes:

[0090] Step S1, real-time acquisition of the communication resources and current task load information of each relay node during the drone inspection process to ensure real-time monitoring of the working status of each relay node.

[0091] Step S2, the task collection module collects computing tasks related to the Internet of Things terminal devices within the coverage area of ​​the drone through the drone during the inspection process.

[0092] Step S3, the optimization algorithm module of the coordination module establishes an energy consumption optimization model for the power system UAV inspection system, the energy consumption optimization model minimizes the energy consumption of the UAV inspection system by rationally allocating the computing tasks carried by the UAV to the relay nodes, and obtains the task allocation scheme and the communication resource allocation scheme by solving the energy consumption optimization model. The task allocation scheme determines the transmission mode selected by each UAV to the relay node on the path and the computing tasks sent, which can minimize the total communication cost, and the communication resource allocation scheme includes each relay node evaluating the computing tasks that can be accepted according to its own remaining computing resource information;

[0093] The task allocation module of the coordination module collects the current network status information in real time, identifies the relay nodes with communication capabilities and on the flight route of the UAV, and transmits the computing tasks to the identified relay nodes according to the task allocation plan;

[0094] The communication resource allocation module of the coordination module dynamically evaluates the actual communication resource situation of each relay node and adjusts the task allocation plan according to the communication resource allocation plan;

[0095] Step S4, the relay node processing module transmits the computing task received from the coordination module to the upper cloud or edge server for computing;

[0096] Step S5, the monitoring and feedback module receives the communication network status information in real time and feeds it back to the coordination module, so that the coordination module can continuously adjust the task allocation strategy according to the communication cost of the inspection process to cope with possible environmental changes or network status fluctuations. Through feedback adjustment, it is ensured that the entire task execution link always maintains the best balance between communication cost and task efficiency.

[0097] The present invention reduces the total communication cost of the system and improves the operation efficiency and inspection quality of the power system by reasonably scheduling the task allocation of the unmanned aerial vehicle. Figure 4 By comparing the total energy consumption of different transmission methods, it can be seen that the energy consumption reduction performance of using MCS is better than that of using LBT and UBT methods alone to send data. Figure 5 Comparing the ADMM algorithm proposed in the present invention with the commonly used interior point method and subgradient method, it can be clearly seen that the ADMM algorithm requires fewer iterations to reach the optimal value. This method can not only improve the working efficiency of drones, but also effectively reduce resource waste, which has important theoretical and practical significance.

[0098] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed by the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A UAV power grid inspection communication resource allocation system, characterized in that: It includes task collection module, coordination module, relay node processing module and monitoring and feedback module; The task collection module is used to collect computing tasks related to IoT terminal devices in the drone coverage area during the inspection process; The coordination module includes an optimization algorithm module, a task allocation module and a communication resource allocation module, wherein: The optimization algorithm module is used to establish an energy consumption optimization model for the power system UAV inspection system. The energy consumption optimization model minimizes the energy consumption of the UAV inspection system by rationally allocating the computing tasks carried by the UAV to the relay nodes. The task allocation scheme and the communication resource allocation scheme are obtained by solving the energy consumption optimization model. The task allocation scheme determines the transmission mode selected by each UAV to the relay node on the path and the computing tasks sent, which can minimize the total communication cost. The communication resource allocation scheme includes each relay node evaluating the computing tasks that can be accepted based on its own remaining computing resource information; The task allocation module is used to collect current network status information in real time, identify relay nodes that have communication capabilities and are on the flight route of the drone, and transmit computing tasks to the identified relay nodes according to the task allocation plan; The communication resource allocation module is used to dynamically evaluate the actual communication resource situation of each relay node and adjust the task allocation plan according to the communication resource allocation plan; The relay node processing module is used to transmit the computing task received from the coordination module to the upper cloud or edge server for computing; The monitoring and feedback module is used to receive communication network status information in real time and feed it back to the coordination module.

2. The UAV power grid inspection communication resource allocation system according to claim 1, characterized in that: During the task collection process, the drone will regularly update its power status, and when the battery power is lower than the set threshold, it will automatically adjust the inspection plan and select the optimal route to return to the ground node as soon as possible for battery replacement or charging.

3. The UAV power grid inspection communication resource allocation system according to claim 1, characterized in that: After receiving the computing tasks, the UAV collects the location information of the relay nodes in the inspection area, the maximum task reception and the energy consumption indicators to provide basic data for the UAV's path planning, optimizes the algorithm module to establish an energy consumption model, considers the flight energy consumption and communication consumption, and evaluates the energy consumption under different flight conditions; at the same time, by constructing a communication cost model, the communication cost of each UAV is evaluated.

4. The UAV power grid inspection communication resource allocation system according to claim 1, characterized in that: The energy consumption optimization model models the energy consumption of the UAV inspection system when uploading data to the relay node using two transmission modes, the authorized frequency band and the unauthorized frequency band. The optimization algorithm uses the alternating direction multiplier method to solve the optimization problem. The optimization problem is as follows: Among them, e (1) (x) is the system communication cost using unlicensed frequency bands, e (2) (y) is the system communication cost using the licensed frequency band; l , η n are the maximum data acceptance capabilities of GN1 and BSn, respectively, i represents the total amount of data sent by UAV i within τ time, ε i Indicates the maximum energy consumption allowed by the drone's battery; The alternating direction multiplier method divides the global optimization problem into N sub-problems, each of which is solved by a drone using its private information. Define g ij = <x il ,y in > is the amount of data sent by drone i to relay node j, nj=<ζ l ,η n > represents the maximum data receiving capability set of the relay node, and the optimization problem is simplified to: Define the feasible set of the i-th UAV The indicator function of the i-th UAV is defined as follows: Similarly, for other constraints in the joint optimization problem, define the feasible domain Then the indicator function is: By converting the constraints in the optimization problem into an indicator function, the function in the form of ADMM is obtained as follows: stg-n=0 in The final result is an optimized g value, which corresponds to the minimum energy consumption value.

5. The UAV power grid inspection communication resource allocation system according to claim 1, characterized in that: The current network status information collected by the task allocation module includes the geographical location, load status and communication capability of each relay node, and the relay nodes include base stations BSs and ground nodes GNs.

6. The UAV power grid inspection communication resource allocation system according to claim 1, characterized in that: The communication resource allocation module is specifically used to dynamically evaluate the actual communication resource situation of each relay node by continuously acquiring the current load status, network bandwidth and processing capacity of each relay node, analyze the communication resources that the relay node can use for task processing within a specific time, and then adjust the task allocation plan to avoid the occurrence of overload.

7. A method for allocating communication resources for unmanned aerial vehicle power grid inspection, characterized in that: The UAV power grid inspection communication resource allocation system applied to any one of claims 1 to 6, the method comprising: Step S1, real-time acquisition of the communication resources and current task load information of each relay node during the drone inspection process to ensure real-time monitoring of the working status of each relay node. Step S2, the task collection module collects computing tasks related to the Internet of Things terminal devices within the coverage area of ​​the drone through the drone during the inspection process. Step S3, the optimization algorithm module of the coordination module establishes an energy consumption optimization model for the power system UAV inspection system, the energy consumption optimization model minimizes the energy consumption of the UAV inspection system by rationally allocating the computing tasks carried by the UAV to the relay nodes, and obtains the task allocation scheme and the communication resource allocation scheme by solving the energy consumption optimization model. The task allocation scheme determines the transmission mode selected by each UAV to the relay node on the path and the computing tasks sent, which can minimize the total communication cost, and the communication resource allocation scheme includes each relay node evaluating the computing tasks that can be accepted according to its own remaining computing resource information; The task allocation module of the coordination module collects the current network status information in real time, identifies the relay nodes with communication capabilities and on the flight route of the UAV, and transmits the computing tasks to the identified relay nodes according to the task allocation plan; The communication resource allocation module of the coordination module dynamically evaluates the actual communication resource situation of each relay node and adjusts the task allocation plan according to the communication resource allocation plan; Step S4, the relay node processing module transmits the computing task received from the coordination module to the upper cloud or edge server for computing; Step S5: The monitoring and feedback module receives the communication network status information in real time and feeds it back to the coordination module.

Images