Intelligent line dynamic scheduling optimization method based on AI algorithm

Abstract

The invention provides an intelligent line dynamic scheduling optimization method based on an AI algorithm, and the method comprises the steps: obtaining the temperature, voltage and current, and vibration data, detected by a sensor, of a plurality of nodes on a power transmission line at a plurality of moments, inputting the temperature, voltage and current, and vibration data of the plurality of nodes at a plurality of moments into simulation software, and enabling the simulation software to output state data of each component on the power transmission line at multiple moments, inputting the state data of each component on the power transmission line at multiple moments into the fault prediction model for processing, and obtaining fault prediction data of each component on the power transmission line. The inspection plan of the unmanned aerial vehicle is generated according to the fault prediction data of each component on the power transmission line, and the unmanned aerial vehicle is controlled to inspect the power transmission line according to the inspection plan of the unmanned aerial vehicle, so that the fault of the power transmission line is found more accurately at the minimum inspection cost.

Description

Technical Field

[0001] The present application relates to the field of electric power inspection technology, and in particular to a smart line dynamic scheduling optimization method based on AI algorithm. Background Art

[0002] As the scale of power grids continues to expand and their complexity increases, the safe and stable operation of transmission lines, as a key component of the power system, is of vital importance.

[0003] The traditional manual inspection method is not only time-consuming and laborious, but also greatly affected by factors such as the environment and weather, making it difficult to achieve all-weather, all-angle monitoring. With the development of drone technology, the use of drone inspection technology has been proposed. Usually, the inspection personnel formulate the flight plan of the drone according to experience or inspection specifications, and the pilot controls the drone for inspection according to the flight plan of the drone. However, since the flight plan needs to be formulated based on experience or inspection specifications. If the inspection frequency is too high, it will cause the inspection cost to rise. If the inspection frequency is too low, it is easy to fail to detect faults in time.

[0004] Therefore, the existing flight plan formulation method cannot select the optimal inspection frequency, resulting in the inability to achieve both lower inspection costs and more accurate fault detection. Summary of the invention

[0005] The AI ​​algorithm-based smart line dynamic scheduling optimization method provided in this application uses sensors to collect data on the transmission line, analyzes the patterns of transmission line faults, and formulates inspection plans based on the patterns of faults, so as to achieve the effect of balancing inspection costs and accurate fault detection.

[0006] The present application provides a smart line dynamic scheduling optimization method based on an AI algorithm. Each node of the transmission line is provided with a sensor. The method is applied to a management system. The method includes: Obtain temperature, voltage, current and vibration data of multiple nodes on the transmission line detected by sensors at multiple times; Input the temperature, voltage, current and vibration data of multiple nodes at multiple times into the simulation software, so that the simulation software outputs the status data of each component on the transmission line at multiple times; Inputting the state data of each component on the transmission line at multiple times into the fault prediction model for processing to obtain the fault prediction data of each component on the transmission line; An inspection plan for drones is generated based on the fault prediction data of various components on the transmission line, and drones are dispatched to inspect the transmission lines according to the inspection plan.

[0007] In the above technical solution, sensors are arranged on the transmission line, and the sensors detect the temperature, vibration data, voltage and current of each node on the transmission line. In this way, the simulation software can perform fault prediction based on the temperature, vibration data and voltage and current of each node. After determining the fault prediction data of the transmission line, the optimal UAV inspection plan can be formulated according to the fault prediction data, and the UAV can be controlled to perform inspections according to the inspection plan, thereby increasing the probability of discovering transmission line faults with the lowest inspection cost.

[0008] In a possible implementation, the fault prediction data includes the fault location, the fault start time, the fault type, and the fault expansion rate; Accordingly, the inspection plan of the UAV is generated according to the fault prediction data of each component on the transmission line, and the UAV is controlled to inspect the transmission line according to the inspection plan of the UAV, which specifically includes: For each component, the inspection time of the component is generated according to the fault start time, fault type and fault expansion rate; the shooting posture of the component is determined according to the fault location; The components are grouped according to their locations and inspection times, and the inspection plan of the drone is generated according to the inspection times and shooting postures of the grouped components.

[0009] In the above technical solution, the inspection time of the drone is generated according to the fault start time, fault type and fault expansion rate, and the shooting posture of the drone is determined according to the fault location. In this way, an inspection plan can be formulated based on the development of the fault. Compared with the inspection plan formulated based on experience and inspection specifications, the inspection plan formulated by this application can take into account both the inspection cost and the probability of discovering faults. In addition, the components are grouped according to their locations and the inspection times of each component, and the inspection plan is generated according to the grouping results. This can reduce the inspection path of the drone, and timely discover faults, improve the accuracy of fault discovery, and reduce the impact of faults on the operation of the transmission line.

[0010] In a possible implementation, the inspection time of the drone is generated according to the fault start time, fault type and fault expansion rate, specifically including: Determine the failure time when the component fails based on the fault start time, fault type and fault expansion rate; The difference between the fault start time and the preset lower limit buffer time is calculated as the lower limit time of the inspection period; the sum of the failure time and the preset upper limit buffer time is calculated as the upper limit time of the inspection period; During the inspection period, at least one moment is selected as the inspection moment of the drone.

[0011] In the above technical solution, the period between the lower limit time of the inspection period and the upper limit time of the inspection period is the period when component failure occurs. By selecting at least one moment between the lower limit time of the inspection period and the upper limit time of the inspection period as the inspection time of the drone, the probability of discovering faults during the drone inspection process can be increased.

[0012] In a possible implementation, selecting at least one moment during the inspection period as the inspection moment of the drone specifically includes: Calculate the duration between the lower limit time of the inspection period and the upper limit time of the inspection period; If the duration is greater than the preset duration threshold, the product of the first coefficient and the duration is calculated to obtain the first interval time; the sum of the fault start time and the first interval time is used as the inspection time; If the duration is less than or equal to the preset duration threshold, the product of the second coefficient and the duration is calculated to obtain the second interval time; the sum of the fault start time and the second interval time is used as the inspection time; wherein the first coefficient is greater than the second coefficient.

[0013] In the above technical solution, after obtaining the inspection period of each component, the duration of the inspection period is calculated, and a coefficient is selected according to the duration of the inspection period. The longer the duration of the inspection period, the larger the selected coefficient, and the shorter the duration of the inspection period, the smaller the selected coefficient. In this way, the shorter the duration of the inspection period, the closer the inspection time is to the start time of the fault, so that the drone can find the fault in time, reserve time for maintenance, and reduce the probability of power transmission failure due to component failure in the transmission line.

[0014] In a possible implementation, the inspection plan of the drone is generated according to the position of each component, the inspection time of each component and the shooting posture of each component, specifically including: Obtain the inspection time of each component located in the same inspection area according to the position of each component, and sort the inspection time of each component located in the same inspection area in order from first to last; The inspection time of each component is grouped according to the time sorting result, and the inspection time of the component is adjusted according to the position of each component in the same group. The inspection plan of the drone is generated according to the adjusted inspection time and shooting posture.

[0015] In the above technical solution, after calculating the inspection time of each component, the components are spatially grouped according to their positions, and then temporally grouped according to the inspection time of the components in the spatial groups. The inspection time of the components is adjusted according to the positions of the components in the same group. This can not only reduce the flight path of the drone, but also increase the probability of the drone detecting a fault during inspection.

[0016] In a possible implementation, the temperature, voltage, current and vibration data of multiple nodes at multiple times are input into the simulation software, so that the simulation software outputs the status data of each component on the transmission line at multiple times, specifically including: Input the temperature and vibration data of multiple nodes at multiple times into the mechanical vibration simulation software to obtain the stress or force of each component of the transmission line at multiple times; Input the voltage and current of multiple nodes at multiple times into the power simulation software to obtain the voltage and current of each component of the transmission line at multiple times; Accordingly, the state data of each component on the transmission line at multiple times are input into the fault prediction model for processing to obtain fault prediction data, specifically including: Input the stress of the component at multiple times into the fracture failure prediction model to obtain the time when the component starts to crack and the crack growth rate; Input the forces of the component at multiple moments into the loosening fault prediction model to obtain the moment when the component starts to loosen and the loosening rate; The voltage and current of the component at multiple times are input into the power fault prediction model to obtain the time when the component will short-circuit or open-circuit fail.

[0017] In the above technical solution, the temperature and vibration data are input into the mechanical vibration simulation software to obtain the force or stress of each component, and the looseness or break fault is detected based on the force or stress. The voltage and current are input into the power simulation software to obtain the voltage and current of each component at multiple times, so as to predict the short circuit and open circuit faults based on the voltage and current of the components. By predicting the mechanical fault and the short circuit and open circuit fault, the drone is used to inspect the fault, and the fault is discovered in time, reducing the impact of component failure on the operation of the transmission line.

[0018] In a possible implementation, after generating an inspection plan for the drone according to the fault prediction data of each component on the transmission line, and controlling the drone to inspect the transmission line according to the inspection plan of the drone, the method further includes: Obtain inspection images of power transmission lines taken by drones, use the target detection model to perform target detection on the inspection images, and obtain target detection results; The image of the target object is extracted from the inspection image according to the target detection result, and the image of the target object is input into the defect recognition model for fault recognition.

[0019] In the above technical solution, when the drone takes inspection images, in order to ensure the safety of the drone, it is usually far away from the photographed object. Compared with other objects in the inspection image, in order to more accurately identify defects, the target object in the inspection image is first extracted using the target detection model, and then the fault recognition model is used to identify the fault type of the extracted image to improve the accuracy of fault recognition.

[0020] The present application provides a smart line dynamic scheduling optimization system based on AI algorithm, including: An acquisition module, used to acquire temperature, voltage, current and vibration data of multiple nodes on the transmission line at multiple times detected by sensors; A processing module, used for inputting the temperature, voltage, current and vibration data of multiple nodes at multiple times into simulation software, so that the simulation software outputs the status data of each component on the transmission line at multiple times; The processing module is used to input the state data of each component on the transmission line at multiple times into the fault prediction model for processing to obtain the fault prediction data of each component on the transmission line; The processing module is used to generate an inspection plan for the drone according to the fault prediction data of each component on the transmission line, and dispatch the drone to inspect the transmission line according to the inspection plan of the drone.

[0021] The present application provides an electronic device, comprising: a memory, a processor; Memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory, so that the processor executes various possible implementations as described above.

[0022] The present application provides a computer-readable storage medium, in which computer-executable instructions are stored. When the computer-executable instructions are executed by a processor, they are used to implement the various possible implementation methods described above.

[0023] The present application provides a computer program product, including a computer program, which implements various possible implementation modes as described above when executed by a processor. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0025] Figure 1 Application scenarios of the intelligent line dynamic scheduling optimization method based on AI algorithm provided in this application; Figure 2A flowchart of a smart line dynamic scheduling optimization method based on an AI algorithm provided in this application; Figure 3 A schematic diagram of the principle of adjusting the inspection time of each component provided in this application.

[0026] The above drawings have shown clear embodiments of the present application, which will be described in more detail later. These drawings and text descriptions are not intended to limit the scope of the present application in any way, but to illustrate the concept of the present application to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0027] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application. Instead, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0028] With the development of drone technology, the use of drone inspection technology has been proposed. Usually, the inspection personnel formulate the flight plan of the drone according to experience or inspection specifications, and the pilot controls the drone for inspection according to the flight plan. However, since the flight plan needs to be formulated based on experience or inspection specifications. If the inspection frequency is too high, it will cause the inspection cost to rise. If the inspection frequency is too low, it is easy to fail to detect faults in time.

[0029] Therefore, the existing flight plan formulation method cannot select the optimal inspection frequency, resulting in the inability to balance lower inspection costs and a higher probability of fault detection.

[0030] This application provides a smart line dynamic scheduling optimization method based on AI algorithm. By using sensors to collect data on the transmission line, analyzing the patterns of transmission line faults, and formulating inspection plans based on the patterns of faults, it is possible to achieve the effect of balancing inspection costs and fault detection accuracy.

[0031] The technical solution of the present application and how the technical solution of the present application solves the above-mentioned technical problems are described in detail below with specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below in conjunction with the accompanying drawings.

[0032] Figure 1 The application scenarios of the intelligent line dynamic scheduling optimization method based on AI algorithm provided in this application are as follows: Figure 1As shown, the transmission line 200 includes a transmission tower 210 and a conductor 220, and the conductor 220 is used to transmit high voltage. Sensors 230 are arranged at various nodes of the transmission line 200, for example: vibration sensors are arranged at the stress nodes of the transmission line to detect the magnitude of vibration near the stress nodes. Temperature sensors are arranged at the high-voltage contact nodes of the transmission line, and the temperature sensors are used to detect the temperature on the high-voltage contact nodes. Voltage and current sensors are also arranged on the transmission line to detect the voltage and current on the transmission line. The intelligent line dynamic scheduling optimization system based on AI algorithm (hereinafter referred to as: management system 100) provided in this application is communicated with each sensor 230 on the transmission line. The management system 100 obtains the data collected by each sensor and formulates the inspection plan of the drone based on the data collected by the sensor.

[0033] Figure 2 A flow chart of the intelligent line dynamic scheduling optimization method based on AI algorithm provided in this application, such as Figure 2 As shown, the method includes: S101. The management system obtains the temperature, vibration data, and voltage and current of each node on the transmission line detected by sensors at multiple times.

[0034] Among them, sensors are arranged at each node of the transmission line, and the sensors detect the temperature, voltage, current and vibration data of each node at multiple times. The sensors are connected to the management system in communication, and after the sensors detect the temperature, voltage, current and vibration data of each node, the above data are reported to the management system.

[0035] S102. The management system inputs the temperature, voltage, current and vibration data of each node at multiple times into the simulation software, so that the simulation software outputs the status data of each component on the transmission line at multiple times.

[0036] The management system runs simulation software, which builds a transmission line model based on the design data of the transmission line. The temperature, vibration data, voltage and current detected by the sensor are input into the simulation software, so that the simulation software can analyze the state of the transmission line and obtain the state data of each component in the transmission line at multiple times.

[0037] In some examples, the simulation software includes mechanical vibration simulation software, in which a transmission line model is constructed, and vibration data is input into the mechanical vibration simulation software, and the mechanical vibration simulation software predicts the stress magnitude of each component.

[0038] The simulation software includes power simulation software. A transmission line model is constructed in the power simulation software, and voltage and current are input into the power simulation software. The power simulation software analyzes the voltage and current borne by each component.

[0039] S103. The management system inputs the status data of each component on the transmission line at multiple times into the fault prediction model for processing to obtain fault prediction data of each component.

[0040] The state data of the component at multiple moments include the stress of the component at multiple moments, the magnitude of the force at multiple moments, and the voltage and current at multiple moments.

[0041] The fault prediction model is obtained by training the machine learning model. The fault prediction model includes a fracture fault prediction model, a loose fault prediction model, and a power fault prediction model. The machine learning model is trained with different training data to obtain the corresponding fault prediction model.

[0042] The stress of the component at multiple times is input into the fracture failure prediction model to obtain the time when the crack appears in the component and the crack propagation rate.

[0043] The force magnitudes of the components at multiple moments are input into the loosening fault prediction model to obtain the moment when the components loosen and the loosening rate.

[0044] The voltage and current of the component at multiple times are input into the power fault prediction model to obtain the time when the component will short-circuit or open-circuit fail.

[0045] Input temperature and vibration data into mechanical vibration simulation software to obtain the force or stress of each component, and detect the fall-off or break fault based on the force or stress. Input voltage and current into power simulation software to obtain the voltage and current of each component at multiple times, so as to predict short-circuit and open-circuit faults based on the voltage and current of the component. By predicting mechanical faults and short-circuit and open-circuit faults, using drones to inspect faults, timely detect faults, and reduce the impact of component failures on the operation of transmission lines.

[0046] It is worth mentioning that by integrating temperature, voltage, current and vibration data, the above scheme provides comprehensive monitoring of the status of the transmission line. These data reflect the working status and potential problems of the components from different angles, providing a solid foundation for subsequent simulation and prediction. Based on the output of mechanical vibration simulation software and power simulation software, the fault prediction model can accurately predict faults such as component breakage, looseness, short circuit and open circuit. This accurate prediction helps operation and maintenance personnel take measures in advance to avoid the occurrence or expansion of faults. Through simulation and prediction, operation and maintenance personnel can clearly know which components have potential failure risks, so that they can conduct more targeted inspections during the inspection process. This helps to reduce inspection blind spots and improve inspection efficiency. Based on the understanding of component status, operation and maintenance personnel can optimize the inspection path to ensure that inspection tools such as drones can cover all key components with the shortest path, further reducing inspection costs. Through simulation and prediction, operation and maintenance personnel can detect problems at the early stage of the fault, so as to take timely and effective measures to repair or replace components to avoid further expansion of the fault or evolution into a serious accident. Among them, early fault detection and repair can help reduce the power outage time caused by faults, reduce the impact on power supply, and thus reduce economic losses and social impact. The above scheme realizes the automatic monitoring and intelligent prediction of the status of transmission lines, reduces the dependence on manual inspections, and improves the intelligence level of operation and maintenance work. Based on the simulation and prediction results, operation and maintenance personnel can formulate inspection plans and maintenance strategies more scientifically to achieve data-driven decision support. By timely detecting and repairing potential faults, this method helps to improve the stability of transmission lines and reduce the risk of system fluctuations or collapse caused by faults. Stable transmission lines help to ensure the reliability of power supply and meet the electricity demand of social production and life.

[0047] S104. The management system generates an inspection plan for the drone according to the fault prediction data of each component on the transmission line, and controls the drone to inspect the transmission line according to the inspection plan.

[0048] Among them, the fault prediction data output by the simulation software includes information such as the fault start time, fault type, fault expansion rate, and fault location. The inspection frequency is determined based on the fault start time, fault expansion rate, and fault type in the fault prediction data, and the inspection posture of the drone is determined based on the fault location. In this way, the drone inspection plan is formulated based on the fault prediction data of each component on the transmission line.

[0049] After formulating the inspection plan for the drone, the management system sends the inspection plan to the drone control platform. The drone control platform generates flight instructions based on the inspection plan and sends the flight instructions to the drone. The drone inspects the transmission lines according to the flight instructions and takes inspection images.

[0050] In the above technical solution, sensors are arranged on the transmission line, and the sensors detect the temperature, vibration data, voltage and current of each node on the transmission line. In this way, the simulation software can perform fault prediction based on the temperature, vibration data and voltage and current of each node. After determining the fault prediction data of the transmission line, the optimal UAV inspection plan can be formulated according to the fault prediction data, and the UAV can be controlled to perform inspections according to the inspection plan, so as to accurately detect faults on the transmission line with the lowest inspection cost.

[0051] It is worth mentioning that multiple sensor nodes are arranged on the transmission line. These sensors can detect the temperature, voltage, current and vibration data of the line at multiple times in real time. These data reflect the operating status of each component of the transmission line and provide basic data support for subsequent fault prediction. Through real-time monitoring, the system can capture subtle changes in the transmission line, providing the possibility of timely detection of potential faults. The collected sensor data are input into the simulation software, which outputs the status data of each component on the transmission line at multiple times based on these data. These data include key parameters such as stress, force, voltage and current of the components, which are important basis for fault prediction. Subsequently, the status data is input into the fault prediction model for processing. The model analyzes the data through AI algorithms such as deep learning to predict the fault location, fault start time, fault type and fault expansion rate of each component on the transmission line. This process realizes the accurate prediction of transmission line faults and provides a scientific basis for the formulation of efficient inspection plans. Based on the fault prediction data, the system can generate the optimal drone inspection plan. For each component predicted to have a fault, the system determines the inspection time according to the fault start time, fault type and fault expansion rate, and determines the shooting posture of the drone according to the fault location. At the same time, the system will group the components according to their locations and inspection times to reduce the flight path and inspection time of the drone. By comprehensively considering the fault prediction data and component location information, the system can develop an economical and efficient inspection plan to ensure that the drone inspects key components at the best time. In addition, traditional inspection methods often rely on manual experience or fixed inspection cycles, which is not only costly but also difficult to ensure the timeliness and accuracy of inspections. The above scheme achieves effective control of inspection costs through intelligent scheduling and dynamic optimization. The system can flexibly adjust the inspection plan according to the fault prediction data to avoid unnecessary inspection frequency and cost waste. At the same time, replacing manual inspections with drone inspections significantly improves inspection efficiency and safety, and reduces labor and time costs during the inspection process. In addition, since the above scheme can accurately predict transmission line faults based on AI algorithms and formulate the optimal inspection plan based on the prediction results, it can significantly improve the fault detection rate. The drone conducts accurate inspections of key components according to the inspection plan, which can timely detect and handle potential faults and avoid the impact of fault expansion on the operation of the transmission line. In addition, the system can also perform target detection and fault identification through inspection images taken by drones, further improving the accuracy and efficiency of fault identification.

[0052] In some embodiments, S104, generating an inspection plan for the drone according to the fault prediction data of each component on the transmission line, and controlling the drone to inspect the transmission line according to the inspection plan of the drone, specifically includes: S201. For each component, generate the inspection time of the drone according to the fault start time, fault type and fault expansion rate, and determine the shooting posture of the drone according to the fault location.

[0053] The component inspection plan includes the inspection time of the drone and the shooting position of the drone. The fault prediction data includes: fault location, fault start time, fault type and fault expansion rate.

[0054] For each component, the fault development process can be determined based on the fault start time and fault expansion rate. According to the fault type, the key moment in the fault development process can be selected and inspections can be carried out at the key moment.

[0055] The fault position refers to the position where the fault occurs on the faulty component. For example, if a crack appears on an insulator, the fault position is the position where the crack appears on the insulator. After obtaining the fault position of the faulty component, the shooting posture of the drone is generated according to the external installation parameters of the camera on the drone, the internal shooting parameters and the fault position of the faulty component.

[0056] S202: Group the components according to their positions and inspection times, and generate an inspection plan for the drone according to the inspection times and shooting postures of the grouped components.

[0057] Among them, each component is grouped according to its location and inspection time, and then the inspection plan of the drone is generated according to the inspection time and shooting posture of the grouped components. By grouping the components based on their location, components located in the same area can be grouped into the same group, reducing the flight path of the drone. By grouping the components based on their inspection time, components with the same inspection time can be grouped into the same group, which can realize timely inspection when a component fails, timely detection of the fault, and reducing the impact of component failure on the operation of the transmission line.

[0058] In the above technical solution, the inspection time of the drone is generated according to the fault start time, fault type and fault expansion rate, and the shooting posture of the drone is determined according to the fault location. In this way, an inspection plan can be formulated based on the development of the fault. Compared with the inspection plan formulated based on experience and inspection specifications, the inspection plan formulated by this application can take into account both the inspection cost and the accuracy of fault detection. In addition, the components are grouped according to the location of each component and the inspection time of each component, and the inspection plan is generated according to the grouping results. This can reduce the inspection path of the drone, detect faults in a timely manner, improve the accuracy of fault detection, and reduce the impact of faults on the operation of the transmission line.

[0059] It is worth noting that for each component predicted to have a fault, the system can intelligently generate an inspection time based on the fault start time, fault type and fault expansion rate. This method ensures that the drone conducts inspections at the critical stage of fault development, avoiding both the waste of resources that may be caused by premature inspections and the expansion of faults that may be caused by late inspections. According to the fault location, the system can accurately set the drone's shooting posture to ensure that the drone can accurately capture the image information of the faulty component. This step is crucial for subsequent fault identification and location. Through intelligent scheduling based on fault prediction data, the system can avoid frequent inspections of fault-free or low-risk components, thereby significantly reducing the inspection cost. Grouping components according to their locations and inspection times and generating inspection plans based on them can effectively reduce the flight path and inspection time of the drone, further reducing the inspection cost. Since the inspection plan is generated based on fault prediction data, the drone can conduct inspections at the early stage of fault development, thereby timely discovering potential faults and avoiding the impact of fault expansion on the operation of the transmission line. By setting the drone's shooting posture, the system can ensure that high-quality image information of the faulty component is captured, thereby improving the accuracy of fault identification. The intelligent generation and execution of inspection plans reduces the need for manual intervention, reduces the risk of human error, and improves operation and maintenance efficiency. Once a fault is discovered, the system can quickly generate an inspection plan and control the drone to conduct inspections, providing strong support for the rapid handling of faults. By promptly discovering and handling potential faults, the system can effectively prevent the impact of fault expansion on the operation of transmission lines and ensure the stable operation of the power system. The long-term implementation of drone inspection plans based on fault prediction data can significantly improve the overall reliability of the transmission line system and reduce power outages and economic losses caused by faults.

[0060] In some embodiments, S201, generating an inspection time of the drone according to the fault start time, the fault type and the fault expansion rate, specifically includes: S301. Determine a failure time when a component fails according to a fault type, a fault expansion rate, and a fault start time.

[0061] If the fault type is crack generation, the crack length at each moment is determined according to the start time of crack generation and the crack growth rate, and the moment when the crack length reaches a first preset length is the failure moment of the component failure.

[0062] Among them, if the fault type is nut falling off, the nut loosening distance at each moment is determined according to the starting moment of nut loosening and the thread loosening rate, and the moment when the nut loosening distance reaches the second preset length is the failure moment when the nut fastening fails.

[0063] If the fault type is the melting of the fitting, the area of ​​the fitting where the melting occurs at each moment is determined according to the start time of the melting of the fitting and the melting rate of the fitting. The moment when the area of ​​the fitting where the melting occurs is greater than the preset area is the failure moment of the fitting where the melting occurs.

[0064] S302: Calculate the difference between the fault start time and the preset lower limit buffer time as the lower limit time of the inspection period. Calculate the sum of the failure time and the preset upper limit buffer time as the upper limit time of the inspection period.

[0065] The preset lower limit buffer time and the preset upper limit buffer time are preset and determined according to the duration from the fault start time to the failure time. The upper buffer time is obtained by multiplying the upper buffer coefficient by the duration, and the lower buffer time is obtained by multiplying the lower buffer coefficient by the duration.

[0066] The lower limit time Tx1 during the inspection period is calculated according to formula (1): Tx1=Tu1-△T1 (1) Among them, Tu1 is the fault start time, and △T1 is the lower limit buffer time.

[0067] The upper limit time Tx2 during the inspection period is calculated according to formula (2): Tx2=Tu2+△T2 (2) Among them, Tu2 is the failure time, and △T2 is the lower limit buffer time.

[0068] S303: Select at least one time during the inspection period as the inspection time of the drone.

[0069] In some examples, at least one time is randomly selected during the inspection period [Tu1, Tu2]. In other examples, the inspection time of the drone is calculated according to the formula t=Tu1+λ(Tu2-Tu1). Where 0<λ<1. By setting different λ, different inspection times are set.

[0070] In the above technical solution, the period between the lower limit time of the inspection period and the upper limit time of the inspection period is the period when component failure occurs. By selecting at least one moment between the lower limit time of the inspection period and the upper limit time of the inspection period as the inspection time of the drone, the probability of discovering faults during the drone inspection process can be increased, and the accuracy of discovering faults can be improved.

[0071] It is worth noting that the system can accurately calculate the failure time of the component according to the fault start time, fault type and fault expansion rate. This step provides an important basis for the subsequent inspection time selection, ensuring that the drone can perform inspections at the critical stage of fault development. By calculating the difference between the fault start time and the preset lower limit buffer time, the system determines the lower limit time of the inspection period; at the same time, by calculating the sum of the failure time and the preset upper limit buffer time, the system determines the upper limit time of the inspection period. This step defines a reasonable time range for the selection of drone inspection time, ensuring that the inspection activity is carried out at the critical stage of fault development. Between the lower limit time and the upper limit time of the inspection period, the system selects at least one time as the drone inspection time. Since this time range covers the entire process of component failure, drone inspections within this time range can significantly increase the probability of finding faults. The system can flexibly adjust the inspection time selection strategy according to different fault types and expansion rates to adapt to the fault development of different components. By reasonably setting the inspection period and selecting the inspection time, the system can avoid excessive inspections of the same component, thereby reducing the inspection cost. In the selection of inspection time, the system can comprehensively consider the fault prediction data of multiple components, optimize the inspection path and time allocation of drones, and further improve resource utilization efficiency. Through intelligent inspection time selection, the system can ensure that drones arrive at the site for inspection in a timely manner at the critical stage of fault development, providing strong support for the rapid handling of faults. Drone inspections reduce the need for manual inspections, reduce the operating risks of operation and maintenance personnel in high-risk environments, and improve the safety of the operation and maintenance process.

[0072] In some embodiments, S303, selecting at least one moment during the inspection period as the inspection moment of the drone, specifically includes: S401, calculate the duration between the lower limit time of the inspection period and the upper limit time of the inspection period, and determine whether the duration is greater than a preset duration threshold. If so, proceed to S402; if not, proceed to S403.

[0073] The difference between the upper limit time of the inspection period and the lower limit time of the inspection period is calculated to obtain the duration between the lower limit time of the inspection period and the upper limit time of the inspection period.

[0074] S402: If the duration is greater than a preset duration threshold, calculate the product of the first coefficient and the duration to obtain a first interval time, and use the sum of the fault start time and the first interval time as the inspection time.

[0075] The first coefficient is preset, and the first coefficient is greater than 0 and less than 1. When it is determined that the duration is greater than the preset duration threshold, the product of the first coefficient and the duration is calculated, and the product is used as the first interval time. The sum of the fault start time and the first interval time is calculated, and the sum result is used as the inspection time. In this way, it is ensured that the inspection time is within the inspection period.

[0076] S403: If the duration is less than or equal to the preset duration threshold, calculate the product of the second coefficient and the duration to obtain the second interval time. The sum of the fault start time and the second interval time is used as the inspection time. The first coefficient is greater than the second coefficient.

[0077] The second coefficient is preset, and the second coefficient is greater than 0 and less than 1. When it is determined that the duration is less than or equal to the preset duration threshold, the product of the second coefficient and the duration is calculated, and the product is used as the second interval time. The sum of the fault start time and the second interval time is calculated, and the sum result is used as the inspection time. In this way, it is ensured that the inspection time is within the inspection period.

[0078] Among them, the first coefficient is greater than the second coefficient, which ensures that the duration of the inspection is shorter and the inspection time is closer to the beginning time of the fault, so that the drone can find the fault in time, reserve time for maintenance, and reduce the probability of power transmission line failure due to component failure.

[0079] In the above technical solution, after obtaining the inspection period of each component, the duration of the inspection period is calculated, and a coefficient is selected according to the duration of the inspection period. The longer the duration of the inspection period, the larger the selected coefficient, and the shorter the duration of the inspection period, the smaller the selected coefficient. In this way, the shorter the duration of the inspection period, the closer the inspection time is to the start time of the fault, so that the drone can find the fault in time, reserve time for maintenance, and reduce the probability of power transmission failure due to component failure in the transmission line.

[0080] It is worth noting that by calculating the duration between the lower limit moment and the upper limit moment during the inspection, the system can dynamically select the appropriate coefficient according to the duration. When the duration is long, a larger first coefficient is selected; when the duration is short, a smaller second coefficient is selected. This dynamic matching strategy ensures that the selection of the inspection time is more refined and can better adapt to the fault development of different components. In the case of a short duration during the inspection, the system selects a smaller second coefficient so that the inspection time is closer to the fault start time. This strategy shortens the delay time of the drone inspection, allowing the drone to reach the fault site for inspection more quickly, thereby improving the timeliness of fault discovery. By discovering the fault in advance, the system reserves more sufficient time for subsequent maintenance work. This helps to reduce the transmission line downtime caused by untimely fault handling and improve the stability and reliability of the power system. In the case of a long duration during the inspection, the system selects a larger first coefficient and appropriately extends the interval between the inspection time and the fault start time. This helps to avoid excessive inspection of the same component, thereby saving the flight time and energy consumption of the drone and improving resource utilization efficiency. By dynamically adjusting the inspection time, the system can reasonably control the inspection cost while ensuring the inspection effect. The intelligent inspection time selection strategy reduces the need for manual inspection planning, reduces the risk of human error, and improves operation and maintenance efficiency. Once the system detects signs of faults, it can quickly generate an optimized inspection plan and control the drone to conduct inspections. This provides timely and accurate information support for operation and maintenance personnel, helping them to make decisions quickly and take effective measures. By promptly detecting and handling potential faults, the system can significantly reduce the probability of transmission line failures caused by component failure. This helps to improve the stability and reliability of the entire power system. Reducing the duration and scope of power outages caused by faults will directly improve users' electricity experience and enhance their satisfaction and trust in the power system.

[0081] In some embodiments, S202, the components are grouped according to their positions and inspection times, and an inspection plan for the drone is generated according to the inspection times and shooting postures of the grouped components, specifically including: S501. Obtain inspection times of various components located in the same inspection area according to the positions of various components, and sort the inspection times of various components located in the same inspection area in order from earliest to latest.

[0082] The components are divided into multiple space groups according to their positions, so that the components in each space group are located in the same area. For each space group, the inspection time of each component in the space group is obtained. In each space group, the components are sorted in descending order according to their inspection time.

[0083] For example, the components on the transmission line are divided into three groups, the first group corresponds to area A, the second group corresponds to area B, and the third group corresponds to area C. For the components in each area, the inspection time of the components is sorted from the earliest to the latest.

[0084] S502, grouping the inspection times of various components according to the time sorting result, adjusting the inspection times of the components according to the positions of the various components in the same group, and generating an inspection plan for the drone according to the adjusted inspection times and shooting postures.

[0085] After obtaining the time sorting result, the inspection time of each component is grouped according to the time sorting result, so that the difference between the inspection time of each component in the same time group is less than the preset time difference.

[0086] After the components are grouped spatially and then temporally, the paths of the components in the same group are planned to generate the inspection order of each component. The inspection time of the components is then adjusted according to the inspection order of each component and the distance between each component. The inspection plan of the drone is generated based on the adjusted inspection time and shooting posture.

[0087] In the above technical solution, after calculating the inspection time of each component, the components are spatially grouped according to their positions, and then temporally grouped according to the inspection time of the components in the spatial groups. The inspection time of the components is adjusted according to the positions of the components in the same group. This can not only reduce the flight path of the drone, but also increase the probability of the drone detecting a fault during inspection.

[0088] It is worth mentioning that, first, according to the location information of each component, the components located in the same inspection area are divided into the same spatial group. This step ensures that the drone can focus on inspecting a certain area when performing the inspection task, reducing the need for frequent flights between different areas, thereby reducing the flight path and energy consumption of the drone. On the basis of spatial grouping, time grouping is further performed according to the inspection time of each component, and the inspection time within the same group is finely adjusted. By optimizing the flight route and time arrangement of the drone, unnecessary flight paths are further reduced and the inspection cost is reduced. The components in the same inspection area are sorted from first to last according to their inspection time, and time grouping is performed according to the sorting results. This step ensures that the drone can inspect the components in the optimal order, avoiding time waste and reduced efficiency caused by improper inspection order. In the same group, the inspection time is finely adjusted according to factors such as the specific location of the component and the flight speed of the drone. By optimizing the inspection order and time, the drone can cover all components that need to be inspected in the shortest time, improving the inspection efficiency. At the same time, due to the more precise adjustment of the inspection time, the probability of drones conducting inspections at the critical stage of fault development has also increased significantly, thereby improving the fault detection rate. This technical solution can dynamically adjust the inspection plan according to the real-time location and status information of the components. When the position or status of the component changes, the system can respond quickly and regenerate the optimal inspection plan to ensure that the drone can always inspect according to the optimal route and time. Whether it is an emergency inspection of a single faulty component or a comprehensive inspection of the entire transmission line, the technical solution can generate a corresponding inspection plan according to actual needs. This flexibility and adaptability enables the drone inspection system to better meet the inspection needs in different scenarios. By optimizing the drone inspection plan, the system can more reasonably configure inspection resources (such as drones, sensors, etc.) to avoid resource waste and redundancy. This helps to improve the overall operation and maintenance level and reduce operation and maintenance costs. Since drones can detect faults and return relevant information in the shortest time, operation and maintenance personnel can respond quickly and take effective measures to deal with them. This helps to shorten the fault handling time and reduce the power outage time and economic losses caused by faults.

[0089] In some embodiments, S502, adjusting the inspection time of the components according to the positions of the components in the same group, specifically includes: S601. Perform path planning according to the positions of components in the same group, and generate an inspection order for each component.

[0090] The same group refers to a group obtained after spatial grouping and temporal grouping, in which the positions of the components are close and the inspection times of the components are adjacent. The distances between the components in the same group are processed using an existing path planning algorithm to obtain an optimal path. The optimal path includes the inspection order of each component.

[0091] S602. Obtain the distance between components in two adjacent inspection orders according to the inspection order of each component and the position of each component, and calculate the minimum inspection interval between components in two adjacent inspection orders according to the distance between components in two adjacent inspection orders and the flight speed of the drone.

[0092] For example, there are 5 components in the same group. The inspection order of the 5 components is determined by the path planning algorithm as component 3, component 1, component 4, component 2, and component 5. The four groups of component 3 and component 1, component 1 and component 4, component 4 and component 2, and component 2 and component 5 are all components in two adjacent inspection orders. The distance between component 3 and component 1, the distance between component 1 and component 4, the distance between component 4 and component 2, and the distance between component 2 and component 5 are calculated respectively.

[0093] Figure 3 The schematic diagram of the principle of adjusting the inspection time of each component provided in this application is as follows: Figure 3 As shown, according to the distance between component 3 and component 1 and the flight speed of the drone, the minimum inspection interval between component 3 and component 1 is calculated (as shown by the solid dots in the figure). According to the distance between component 1 and component 4 and the flight speed of the drone, the minimum inspection interval between component 1 and component 4 is calculated. According to the distance between component 4 and component 2 and the flight speed of the drone, the minimum inspection interval between component 4 and component 2 is calculated. According to the distance between component 2 and component 5 and the flight speed of the drone, the minimum inspection interval between component 2 and component 5 is calculated.

[0094] S603, calculate the theoretical inspection interval between the inspection moments of components in two adjacent inspection orders, calculate the difference between the theoretical inspection interval and the minimum inspection interval of components in two adjacent inspection orders, and adjust the inspection moment of the components according to the difference and the minimum inspection interval between components in two adjacent inspection orders.

[0095] Among them, continue to refer to Figure 3, calculate the time difference between the inspection time of component 3 and the inspection time of component 1, and obtain the theoretical inspection interval between component 3 and component 1 (as shown by the hollow point in the figure). Calculate the time difference between the inspection time of component 1 and the inspection time of component 4, and obtain the theoretical inspection interval between component 1 and component 4. Calculate the time difference between the inspection time of component 4 and the inspection time of component 2, and obtain the theoretical inspection interval between component 4 and component 2. Calculate the time difference between the inspection time of component 2 and the inspection time of component 5, and obtain the theoretical inspection interval between component 2 and component 5.

[0096] Calculate the difference between the theoretical inspection interval and the minimum inspection interval of component 3 and component 1, calculate the difference between the theoretical inspection interval and the minimum inspection interval of component 1 and component 4, calculate the difference between the theoretical inspection interval and the minimum inspection interval of component 4 and component 2, calculate the difference between the theoretical inspection interval and the minimum inspection interval of component 2 and component 5, and obtain the differences between the 4 theoretical inspection intervals and the minimum inspection intervals.

[0097] In some examples, adjusting the inspection time of the component according to the difference and the minimum inspection interval between components in two adjacent inspection orders specifically includes: Multiple difference values ​​are sorted from small to large. If the inspection orders of components corresponding to multiple adjacent difference values ​​are adjacent, and the order of multiple adjacent difference values ​​is less than a preset order value, then any component is selected from the components corresponding to the multiple adjacent difference values ​​as a reference component, and the inspection times of other components are corrected according to the inspection time of the reference component, the inspection order of the components and the minimum inspection interval between the components.

[0098] It is worth noting that by sorting multiple differences (i.e., the difference between the theoretical inspection interval and the minimum inspection interval) and selecting adjacent components with smaller differences as reference components, the system can adjust the inspection path and time more accurately. This strategy ensures that the drone can cover all target components with a shorter path and less time during the inspection process, thereby significantly reducing the inspection cost. By correcting the inspection time, the system can avoid unnecessary flights of the drone during the inspection process, such as excessive circling or turning back, further reducing energy consumption and flight costs. Adjusting the inspection time based on the reference component can ensure that the drone reaches each component at the best time, avoiding inefficient inspections caused by time delays. This precise time management improves inspection efficiency and enables drones to complete inspection tasks in a shorter time. By optimizing the inspection order and time, the drone can reach the target component in time at the critical stage of fault development, thereby improving the fault detection rate. This is crucial for the stable operation of the power system, because timely detection of faults can greatly reduce the power outage time and economic losses caused by faults. This strategy allows the system to dynamically adjust the inspection plan according to real-time conditions. When the position of components, the flight speed of drones, or the inspection requirements change, the system can quickly recalculate the difference, select reference components, and adjust the inspection time to ensure the flexibility and adaptability of the inspection plan. Whether it is for complex and changing inspection environments or for different types of faulty components, this strategy can generate the optimal inspection plan based on actual needs. This adaptability enables the drone inspection system to better meet the inspection needs in various scenarios. By optimizing the inspection path and time, the system can more reasonably configure drone resources to avoid resource waste and redundancy. This helps to improve resource utilization efficiency and reduce overall operation and maintenance costs. Due to the optimization of the inspection plan, operation and maintenance personnel can obtain inspection results more quickly and take corresponding measures. This improves operation and maintenance efficiency and reduces power outage time and economic losses caused by untimely fault handling. By optimizing the inspection path and time, the system can reduce the risk of drone failure during the inspection process. For example, avoid drones flying for a long time in bad weather or complex environments, thereby reducing the risk of inspection interruption due to mechanical failure or signal loss. By improving inspection efficiency and fault detection rate, the system can provide users with more stable and reliable power supply services. This will help improve user satisfaction and trust, and further consolidate the power companies' position in the market.

[0099] For example: the difference between component 2 and component 5 is the smallest, the difference between component 4 and component 2 is the second, the difference between component 3 and component 1 is the third, and the difference between component 1 and component 4 is the fourth. The difference between component 2 and component 5 is adjacent to the difference between component 4 and component 2. The inspection order of the five components is component 3, component 1, component 4, component 2, and component 5. The inspection order corresponding to component 4, component 2, and component 5 is continuous, and the difference between component 2 and component 5 is the smallest, and the difference between component 4 and component 2 is the second, and the rankings are relatively small. Then arbitrarily select component 2 from component 4, component 2, and component 5 as the reference component.

[0100] The inspection order of the five components is component 3, component 1, component 4, component 2, and component 5.

[0101] The inspection time of component 2 is subtracted from the minimum inspection interval between component 4 and component 2 to obtain the corrected inspection time of component 4.

[0102] The corrected inspection time of component 1 is obtained by subtracting the minimum inspection interval between component 1 and component 4 from the corrected inspection time of component 4.

[0103] The corrected inspection time of component 3 is obtained by subtracting the minimum inspection interval between component 3 and component 1 from the corrected inspection time of component 1.

[0104] The inspection time of component 2 is added to the minimum inspection interval between component 2 and component 5 to obtain the corrected inspection time of component 5.

[0105] In the above technical solution, the paths of the components in the same group are planned to determine the inspection order of each component. The minimum inspection interval between the components under the inspection order is determined according to the flight speed of the drone, and the theoretical inspection interval between the components under the inspection order is determined according to the inspection time of each component. The difference between the minimum inspection interval and the theoretical inspection interval is calculated, and the inspection time of each component is corrected according to the difference. In this way, the inspection time requirement can be met while inspecting with a shorter inspection path, which reduces the inspection cost and increases the probability of faults during inspection.

[0106] It is worth noting that by planning the path of each component in the same group, the system can generate the optimal inspection order to ensure that the drone covers all target components with the shortest path during the inspection process. This step significantly reduces the flight distance and time of the drone, and reduces the energy consumption and flight cost during the inspection process. According to the flight speed of the drone and the distance between adjacent components, the system can calculate the minimum inspection interval between each component under the inspection order. This step ensures that the drone can reach each component on time during the inspection process, avoiding inspection delays caused by improper flight speed or path selection. By calculating the difference between the theoretical inspection interval and the minimum inspection interval, and adjusting the inspection time according to these differences, the system can ensure that the drone inspects with the optimal inspection path and order while meeting the inspection time requirements. This adjustment strategy not only ensures the timeliness of the inspection, but also improves the fault detection rate. In addition, the technical solution can dynamically adjust the inspection plan according to factors such as the actual flight speed of the drone, the location of the components, and the inspection time requirements. When any factor changes, the system can respond quickly and regenerate the optimal inspection order and time to ensure the flexibility and adaptability of the inspection plan. Whether it is for complex and changeable inspection environments or for different types of faulty components, the technical solution can generate corresponding inspection plans according to actual needs. This adaptability enables the drone inspection system to better meet the inspection needs in various scenarios. By optimizing the inspection path and time, the system can more reasonably configure drone resources to avoid resource waste and redundancy. This helps to improve resource utilization efficiency and reduce overall operation and maintenance costs. The intelligent inspection time adjustment strategy reduces the need for manual formulation and adjustment of inspection plans, reduces the risk of human error, and improves the degree of automation of inspection work. By optimizing the inspection path and time, the system can ensure that the drone operates in the optimal state during the inspection process, thereby reducing the risk of failure caused by improper flight paths or delayed inspection times. By improving inspection efficiency and fault detection rate, the system can provide users with more stable and reliable power supply services and improve user satisfaction and trust.

[0107] Some embodiments of the present application also provide a smart line dynamic scheduling optimization method based on an AI algorithm, which specifically includes the following steps: S701. The management system obtains the temperature, vibration data, and voltage and current of each node on the transmission line detected by sensors at multiple times.

[0108] S702. The management system inputs the temperature, voltage, current and vibration data of multiple nodes at multiple times into the simulation software, so that the simulation software outputs the status data of each component on the transmission line at multiple times.

[0109] S703. The management system inputs the status data of each component on the transmission line at multiple times into the fault prediction model for processing to obtain fault prediction data of each component.

[0110] S704. The management system generates an inspection plan for the drone based on the fault prediction data of each component on the transmission line, and controls the drone to inspect the transmission line according to the inspection plan.

[0111] S705. The management system obtains the inspection image of the transmission line taken by the drone, performs target detection on the inspection image using the target detection model, and obtains the target detection result.

[0112] Among them, the drone will transmit the inspection images it has taken to the drone control platform, and the drone control platform will return the inspection images to the management system. The management system runs a target detection model, and the inspection images are input into the target detection model, so that the target detection model can extract the target objects in the inspection images and obtain the target detection results.

[0113] The target detection model may be a convolutional neural network. By training the convolutional neural network, the target detection model can identify images of components to be analyzed from inspection images.

[0114] S706: The management system extracts an image of the target object from the inspection image according to the target detection result, and inputs the image of the target object into a defect recognition model for fault recognition.

[0115] The target detection result includes the type of target object and the position of the target object. The image of the target object is extracted from the inspection image according to the position of the target object. The fault recognition model corresponding to the target object is determined according to the type of the target object. The fault detection model can also be a convolutional neural network. The convolutional neural network is trained through different training data, and the trained model can identify multiple types of faults.

[0116] If the target object is an insulator, the image of the insulator is input into the crack detection model to obtain the crack detection result.

[0117] If the target object is a hardware fitting, the image of the hardware fitting is input into the hot spot detection model to obtain the hot spot detection result; and the image of the hardware fitting is input into the distance detection model to obtain the movement displacement of the hardware fitting to realize the falling-off fault identification.

[0118] In the above technical solution, when the drone takes inspection images, in order to ensure the safety of the drone, it is usually far away from the photographed object. Compared with other objects in the inspection image, in order to more accurately identify defects, the target object in the inspection image is first extracted using the target detection model, and then the fault recognition model is used to identify the fault type of the extracted image to improve the accuracy of fault recognition.

[0119] It is worth noting that when drones take inspection images, the images may contain a large number of non-target objects due to the long distance. Through the target detection model, the target objects (such as insulators, hardware, etc.) in the image can be accurately extracted to avoid the interference of non-target objects, thereby improving the accuracy of fault identification. After extracting the target objects, the defect recognition model can perform more detailed fault type identification for these objects. This detailed analysis process helps to find more hidden or minor fault signs and improves the sensitivity of fault identification. In the traditional inspection process, operation and maintenance personnel need to view and analyze the inspection images taken by drones one by one, which is time-consuming and labor-intensive. However, through the target detection and defect recognition model, most of the fault identification work can be completed automatically, greatly reducing the need for manual intervention and improving the inspection efficiency. Through the target detection model, the target objects in the inspection image can be quickly located, and then the defect recognition model can identify the faults of the target objects. This process-based processing method helps operation and maintenance personnel quickly locate the fault point and take corresponding measures to repair it. UAV inspection reduces the risk of operation and maintenance personnel working at high altitude or in harsh environments, while the application of target detection and defect recognition models further reduces the risk of operation and maintenance personnel in the fault identification process and enhances the safety of the inspection process. Through the automated fault identification process, it is possible to reduce misjudgments and missed judgments caused by human factors, ensuring the accuracy and reliability of the inspection results. The application of target detection and defect recognition models realizes the intelligent processing of inspection images and improves the intelligence level of operation and maintenance work. This intelligent processing method helps operation and maintenance personnel to formulate inspection plans and maintenance strategies more scientifically. By processing and analyzing inspection images, a large amount of fault data and information can be obtained. These data can provide decision support for operation and maintenance personnel, helping them to more accurately grasp the operating status and maintenance needs of transmission lines.

[0120] The embodiment of the present application provides a smart line dynamic scheduling optimization system based on an AI algorithm, including: An acquisition module, used to acquire temperature, voltage, current and vibration data of multiple nodes on the transmission line at multiple times detected by sensors; A processing module, used for inputting the temperature, voltage, current and vibration data of multiple nodes at multiple times into simulation software, so that the simulation software outputs the status data of each component on the transmission line at multiple times; The processing module is used to input the state data of each component on the transmission line at multiple times into the fault prediction model for processing to obtain the fault prediction data of each component on the transmission line; The processing module is used to generate an inspection plan for the drone according to the fault prediction data of each component on the transmission line, and control the drone to inspect the transmission line according to the inspection plan of the drone.

[0121] The electronic device provided in this embodiment includes: at least one processor and a memory. Optionally, the device also includes a communication component. The processor, the memory and the communication component are connected via a bus.

[0122] In a specific implementation process, at least one processor executes computer-executable instructions stored in a memory, so that at least one processor executes the above method.

[0123] The specific implementation process of the processor can be found in the above method embodiment, and its implementation principle and technical effect are similar, so this embodiment will not be repeated here.

[0124] In the above embodiments, it should be understood that the processor can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), etc. A general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in the invention can be directly implemented as a hardware processor, or can be implemented by a combination of hardware and software modules in the processor.

[0125] The memory may include a high-speed memory (Random Access Memory, RAM), and may also include a non-volatile memory (NVM), such as at least one disk storage.

[0126] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, the bus in the drawings of this application is not limited to only one bus or one type of bus.

[0127] The present application also provides a computer program product, including a computer program, which implements the above method when executed by a processor.

[0128] The present application also provides a computer-readable storage medium, in which computer-executable instructions are stored. When a processor executes the computer-executable instructions, the above method is implemented.

[0129] The above-mentioned readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk. The readable storage medium can be any available medium that can be accessed by a general or special-purpose computer.

[0130] An exemplary readable storage medium is coupled to a processor so that the processor can read information from the readable storage medium and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can be located in an application specific integrated circuit (Application Specific Integrated Circuits, referred to as: ASIC). Of course, the processor and the readable storage medium can also exist in the device as discrete components.

[0131] The division of units is only a logical function division, and there may be other divisions in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interface, device or unit, which can be electrical, mechanical or other forms.

[0132] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0133] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0134] If the function is implemented 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 solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods of each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0135] Those skilled in the art can understand that all or part of the steps of implementing the above-mentioned method embodiments can be completed by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, the steps of the above-mentioned method embodiments are executed; and the aforementioned storage medium includes: ROM, RAM, disk or optical disk and other media that can store program codes.

[0136] Finally, it should be noted that those skilled in the art will readily conceive of other embodiments of the present invention after considering the specification and practicing the invention disclosed herein. The present invention is intended to cover any variations, uses or adaptations of the present invention, which follow the general principles of the present invention and include common knowledge or customary technical means in the art not disclosed by the present invention, are not limited to the precise structure described above and shown in the drawings, and may be modified and changed in various ways without departing from the scope thereof. The scope of the present invention is limited only by the appended claims.

Claims

1. A smart line dynamic scheduling optimization method based on AI algorithm, characterized in that: Multiple nodes of the transmission line are provided with sensors, and the method is applied to a management system, and the method comprises: Acquire temperature, voltage, current and vibration data of multiple nodes on the transmission line at multiple times detected by the sensor; Input the temperature, voltage, current and vibration data of multiple nodes at multiple times into simulation software, so that the simulation software outputs the status data of each component on the transmission line at multiple times; Inputting the state data of each component on the transmission line at multiple times into the fault prediction model for processing to obtain the fault prediction data of each component on the transmission line; An inspection plan for drones is generated based on fault prediction data of various components on the transmission line, and drones are dispatched to inspect the transmission line based on the inspection plan.

2. The method for intelligent line dynamic scheduling optimization based on AI algorithm according to claim 1 is characterized in that: The fault prediction data includes fault location, fault start time, fault type and fault expansion rate; Accordingly, generating an inspection plan of the drone according to the fault prediction data of each component on the transmission line, and controlling the drone to inspect the transmission line according to the inspection plan of the drone specifically includes: For each component, generating an inspection time of the component according to the fault start time, the fault type and the fault expansion rate; Determine the shooting posture of the component according to the fault position; The components are grouped according to their positions and the inspection times of the components, and an inspection plan for the drone is generated according to the inspection times and shooting postures of the grouped components.

3. The method for intelligent line dynamic scheduling optimization based on AI algorithm according to claim 2 is characterized in that: The inspection time of the drone is generated according to the fault start time, the fault type and the fault expansion rate, specifically including: Determine a failure time when a component fails according to the fault start time, the fault type and the fault expansion rate; Calculate the difference between the fault start time and the preset lower limit buffer time as the lower limit time of the inspection period; calculate the sum of the failure time and the preset upper limit buffer time as the upper limit time of the inspection period; At least one moment is selected during the inspection period as the inspection moment of the UAV.

4. The intelligent line dynamic scheduling optimization method based on AI algorithm according to claim 3 is characterized in that: Selecting at least one moment during the inspection period as the inspection moment of the drone specifically includes: Calculating the duration between the lower limit time of the inspection period and the upper limit time of the inspection period; If the duration is greater than a preset duration threshold, the product of the first coefficient and the duration is calculated to obtain a first interval time; the sum of the fault start time and the first interval time is used as the inspection time; If the duration is less than or equal to the preset duration threshold, the product of the second coefficient and the duration is calculated to obtain the second interval time; the sum of the fault start time and the second interval time is used as the inspection time; wherein the first coefficient is greater than the second coefficient.

5. The method for intelligent line dynamic scheduling optimization based on AI algorithm according to claim 3 is characterized in that: The inspection plan of the drone is generated according to the position of each component, the inspection time of each component and the shooting posture of each component, specifically including: Obtaining the inspection time of each component located in the same inspection area according to the position of each component, and sorting the inspection time of each component located in the same inspection area in order from earliest to latest; The inspection times of the various components are grouped according to the time sorting results, the inspection times of the components are adjusted according to the positions of the various components in the same group, and the inspection plan of the drone is generated according to the adjusted inspection times and shooting postures.

6. The intelligent line dynamic scheduling optimization method based on AI algorithm according to any one of claims 1 to 5, characterized in that: Inputting the temperature, voltage, current and vibration data of multiple nodes at multiple times into simulation software, so that the simulation software outputs the status data of each component on the transmission line at multiple times, specifically includes: Input the temperature and vibration data of multiple nodes at multiple times into mechanical vibration simulation software to obtain stress or force of each component of the transmission line at multiple times; Inputting the voltage and current of multiple nodes at multiple times into the power simulation software to obtain the voltage and current of each component of the transmission line at multiple times; Accordingly, the state data of each component on the transmission line at multiple times are input into the fault prediction model for processing to obtain fault prediction data, which specifically includes: Input the stress of the component at multiple times into the fracture failure prediction model to obtain the time when the component starts to crack and the crack growth rate; Input the forces of the component at multiple moments into the loosening fault prediction model to obtain the moment when the component starts to loosen and the loosening rate; The voltage and current of the component at multiple times are input into the power fault prediction model to obtain the time when the component will short-circuit or open-circuit fail.

7. The intelligent line dynamic scheduling optimization method based on AI algorithm according to any one of claims 1 to 5, characterized in that: After generating an inspection plan for the drone according to the fault prediction data of each component on the transmission line, and controlling the drone to inspect the transmission line according to the inspection plan of the drone, the method further includes: Obtaining an inspection image of a power transmission line taken by a drone, performing target detection on the inspection image using a target detection model, and obtaining a target detection result; An image of a target object is extracted from the inspection image according to the target detection result, and the image of the target object is input into a defect recognition model for fault recognition.

8. An intelligent line dynamic scheduling optimization system based on AI algorithm, characterized in that: include: An acquisition module, used to acquire temperature, voltage, current and vibration data of multiple nodes on the transmission line at multiple times detected by sensors; A processing module, used for inputting the temperature, voltage, current and vibration data of multiple nodes at multiple times into simulation software, so that the simulation software outputs the status data of each component on the transmission line at multiple times; The processing module is used to input the state data of each component on the transmission line at multiple times into the fault prediction model for processing to obtain the fault prediction data of each component on the transmission line; The processing module is used to generate an inspection plan for the drone according to the fault prediction data of each component on the transmission line, and dispatch the drone to inspect the transmission line according to the inspection plan of the drone.

9. An electronic device, characterized in that: include: processor; as well as, A memory, configured to store executable instructions of the processor; The processor is configured to perform the method of any one of claims 1 to 7 by executing the executable instructions.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, which are used to implement the method according to any one of claims 1 to 7 when executed by a processor.

Images