Power line defect report evaluation method and system based on unmanned aerial vehicle image

Abstract

The invention discloses a power line defect report evaluation method and system based on an unmanned aerial vehicle image, and relates to the technical field of defect report evaluation, and the method comprises the following steps: dividing a power line into a plurality of to-be-detected regions, and obtaining a power line defect image, preprocessing the power line defect image to obtain defect surface state data and surrounding environment influence data, constructing a maintenance prediction model, and outputting through the maintenance prediction model to obtain a maintainable to-be-detected area table; obtaining influence data of power line defects in the to-be-detected area on power transmission, and constructing a defect influence evaluation model based on a neural network; and performing comparative analysis on the output of the defect influence evaluation model based on a preset risk value to obtain a defect priority processing table, applying the defect priority processing table to the maintainable to-be-detected area table, and performing priority processing on the to-be-detected area. Through combined application of the defect priority processing table and the maintainable to-be-detected area table, the problem of maintenance priority of the line defect area is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of defect report evaluation, and more specifically, to a power line defect report evaluation and system based on drone images. Background Art

[0002] Power lines are key infrastructure in the power transmission and distribution system, ensuring the stable transmission and supply of electricity. With the advancement of urbanization and the continuous development of industrialization, the load of power lines has increased year by year, and the maintenance and management tasks have become increasingly complex. Therefore, how to efficiently and accurately evaluate the health status of power lines, discover potential defects in a timely manner and perform repairs has become an important issue facing the power industry.

[0003] In order to improve the efficiency and safety of power line inspections, drone technology has been widely used in the field of power line inspections in recent years. Drones have the advantages of strong maneuverability, wide field of vision, and access to dangerous areas, and can efficiently complete the inspection tasks of power lines. By carrying high-definition cameras, infrared thermal imagers and other sensors, drones can obtain high-quality image data, provide a comprehensive view of power lines, and collect a large amount of information in a short period of time. Combined with modern computer vision technology and artificial intelligence algorithms, image data captured by drones can be automatically analyzed to identify various defects in power lines.

[0004] For example, the invention patent with announcement number CN109297978A announces a binocular imaging-based UAV inspection and defect intelligent diagnosis system for power lines, including a UAV platform, which is attached to the UAV body in an orthophoto manner, and is used to collect images of the area where the power lines are located, and transmit the images to a ground-mounted mobile graphics workstation via high-definition wireless image transmission; a UAV flight controller, which is used to control the UAV in real time and display the flight status of the UAV in real time; a ground-mounted mobile workstation, which is used to receive binocular image data transmitted from the UAV platform via a high-definition wireless image transmission radio, process the image data and save the processing results to the power supply bureau's patrol management center; the power supply bureau's patrol management center is used to store and process data for UAV power inspection and typical defect image intelligent diagnosis system. The present invention can perform real-time inspections on power lines and intelligently diagnose defects.

[0005] For example, the invention patent with announcement number: CN114155449A discloses a method and device for automatically generating a line inspection report including defect identification information. By acquiring images collected by drones when inspecting power lines, a spherical range with a preset radius is set based on the voltage level of the power lines and the coordinate information of the pole towers in the power lines as a reference point. The system pre-enters basic information of the power lines, including line names, voltage levels, and pole tower coordinate information. The images are grouped according to the spherical range based on the coordinate information of the images, defects are identified on the grouped images, the identified defects are marked, and defect identification information is filled in based on the basic information. A defect report is generated based on the defect identification information and the basic information, which greatly reduces manual participation and significantly improves the accuracy of identification.

[0006] The above disclosed technical solutions have at least the following technical problems: When defect processing and evaluation of power lines is performed through defect reports, wasteful inspections are performed on areas that do not meet maintenance conditions, optimizing the allocation of maintenance resources, and high-risk areas are not prioritized and identified, which reduces the effectiveness of maintenance work and increases potential risks. In view of the above problems, the present invention proposes a solution. Summary of the invention

[0007] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides a power line defect report evaluation and system based on drone images, which solves the problem of priority and accuracy of repair of power line defect areas by combining a defect priority processing table and a repairable test area table.

[0008] To achieve the above object, the present invention provides the following technical solutions: A power line defect report assessment based on drone images includes the following steps: dividing a power line into a number of areas to be tested, acquiring power line defect images through drones, and preprocessing the power line defect images to obtain defect surface state data and surrounding environment impact data; building a maintenance prediction model based on the defect surface state data and surrounding environment impact data, and obtaining a repairable area to be tested table through the output of the maintenance prediction model; obtaining the impact data of power line defects on power transmission in the area to be tested, and building a defect impact assessment model based on a neural network; comparing and analyzing the output of the defect impact assessment model based on a preset risk value to obtain a defect priority processing table, applying the defect priority processing table to the repairable area to be tested table, and giving priority to the area to be tested.

[0009] In a preferred embodiment, the power line defect image is preprocessed to obtain defect surface status data and surrounding environment impact data, and the specific steps are as follows: obtain the location information of important working areas of the power line, divide the drone working area according to the location information, obtain the power line defect image, perform preprocessing operations on the power line defect image and enhance the image clarity based on histogram equalization; obtain the geometric features of the defect area and the environmental information around the power line by extracting information from the processed image; obtain historical maintenance records, and perform correlation analysis on the geometric features of the defect area and the environmental information around the power line based on the historical maintenance records to obtain the defect surface status data and surrounding environment impact data.

[0010] In a preferred embodiment, the defect surface state data includes a defect surface state influence coefficient, and the defect surface state influence coefficient is specifically obtained by the following method: obtaining historical maintenance records and defect surface state information of power line defects, wherein the defect surface state information includes the area of ​​the defect, the depth of the defect and the degree of corrosion on the defect surface; performing regression analysis on the historical maintenance records to obtain an area maintenance influence factor and a depth maintenance influence factor; calculating the area maintenance severity based on the area of ​​the defect and the area maintenance influence factor; calculating the depth maintenance severity based on the depth of the defect and the depth maintenance influence factor; and calculating the defect surface state influence coefficient based on the area maintenance severity and the depth maintenance severity combined with the degree of corrosion on the defect surface based on a preset defect surface state influence formula.

[0011] In a preferred embodiment, a maintenance prediction model is constructed according to the defect surface state data and the surrounding environment impact data, and a table of maintainable areas to be tested is obtained through the output of the maintenance prediction model, specifically: the newly acquired defect surface state data and the surrounding environment impact data are used as input items, and the maintainability prediction coefficient is used as an output item to construct a maintenance prediction model; the maintainability prediction coefficients of all areas to be tested are obtained, and the maintainability prediction coefficients of all areas to be tested are determined; if the maintainability prediction coefficient is greater than or equal to 1, the current area is marked as a maintainable area; if the maintainability prediction coefficient is less than 1, the current area is an unmaintainable area and is not marked; all marked power line defect areas are obtained, and all marked power line defect areas are sorted and counted to obtain a table of maintainable areas to be tested.

[0012] In a preferred embodiment, the data on the impact of power line defects on power transmission in the test area include power transmission impact data and protection device response data, the power transmission impact data include a power transmission power impact coefficient and a transmission capacity impact coefficient, and the power transmission power impact coefficient is specifically obtained by the following method: obtaining the resistance, line length, normal impedance and additional impedance caused by the power line defects of the power line; obtaining the impact factor of the defect on the impedance by performing regression analysis on the additional impedance change caused by the change in line length in the power line defect; calculating the total impedance based on the impedance and additional impedance of the normal line; calculating the phase angle based on the phase angle formula by combining the total impedance with the resistance of the power line; obtaining the power transmission power impact coefficient by combining the power supply end voltage and current and the load end voltage and current with the phase angle based on a preset power transmission power impact formula.

[0013] In a preferred embodiment, the protection device response data includes a protection device response coefficient, and the protection device response coefficient is specifically obtained by: obtaining the fault detection time and the fault clearing time by monitoring the time from the occurrence of the fault to the protection device detecting the fault and the time from the occurrence of the fault to the fault removal; obtaining the time required for the system to return to stability after the fault is removed, that is, the system recovery time, and the probability that the protection device performs correct action; obtaining the fault severity coefficient by performing data analysis based on the degree of influence of historical protection data on the fault; and calculating the protection device response coefficient based on the preset protection device response formula according to the fault detection time and the fault clearing time combined with the fault severity coefficient.

[0014] In a preferred embodiment, the output of the defect impact assessment model is compared and analyzed based on a preset risk value to obtain a defect priority processing table, and the defect priority processing table is applied to the table of repairable areas to be tested to give priority to the areas to be tested. The specific steps are as follows: obtain the output of the defect impact assessment model of all areas to be tested, and compare and analyze the output of the defect impact assessment model with the preset risk value; if the output of the defect impact assessment model is greater than the preset risk value, mark the risk level of the power line defects in the current area as high; if the output of the defect impact assessment model is equal to the preset risk value, mark the risk level of the power line defects in the current area as medium; if the output of the defect impact assessment model is less than the preset risk value, mark the risk level of the power line defects in the current area as low; for each risk The output of the defect impact assessment model within the degree is based on bubble sort to obtain a high-risk priority table, a medium-risk priority table and a low-risk priority table; the high-risk priority table, the medium-risk priority table and the low-risk priority table are sorted in the order of high, medium and low to obtain the final defect priority processing table; the defect priority processing table is applied to the repairable area table, and the repairable areas to be tested in the repairable area table are processed in order of high, medium and low risk areas based on the defect priority processing table; and the high-risk areas are repaired first according to the sorting order of the areas in the high-risk priority table, and then the high-risk areas are repaired first according to the sorting order of the areas in the medium-risk priority table, and finally the high-risk areas are repaired first according to the sorting order of the areas in the low-risk priority table.

[0015] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages: 1. By combining the defect priority table and the repairable area table, a priority-based decision-making system is formed. This helps to reasonably prioritize the complex power line maintenance work, improve the maintenance efficiency, and ensure the stable operation of the power system.

[0016] 2. By combining the power transmission power influence coefficient, transmission capacity influence coefficient and protection device response coefficient, the neural network model is used to accurately evaluate the impact of power line defects on power transmission, thereby optimizing power line fault detection, maintenance priority and resource allocation, improving the stability, safety and efficiency of the power system, reducing the risk of faults, reducing power outage time and impact range, and realizing intelligent and precise power system operation and maintenance management. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 A schematic diagram of a power line defect report evaluation structure based on drone images provided in an embodiment of the present application.

[0018] Figure 2 A schematic diagram of the structure of a power line defect reporting system based on drone images provided in an embodiment of the present application. DETAILED DESCRIPTION

[0019] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0020] Embodiment 1, Figure 1 A schematic diagram of a power line defect report evaluation structure based on drone images provided in an embodiment of the present application includes the following steps: S1, divide the power line into several test areas, obtain the power line defect image through the drone, and pre-process the power line defect image to obtain the defect surface state data and surrounding environment impact data.

[0021] In this example, power lines usually cover a wide geographical area, including different types of geographical environments such as mountains, plains, and cities. To divide these lines into several areas to be tested, it is first necessary to regionalize them according to the actual geographical distribution and environmental characteristics of the power lines. Current power line monitoring identifies and analyzes defects by scanning the power lines, but due to the processing performance of drones, it may take a lot of time and effort to process the pictures, which may increase the potential risks in other areas and affect the progress and efficiency of maintenance work. Among them, the drone is equipped with a high-definition camera or infrared imager to capture images of the power lines and their surroundings, and a suitable flight platform such as a quadcopter or a fixed-wing drone is selected to ensure that it can approach the power lines and obtain the required image data. A flight path is formulated so that the drone can fly near the power lines according to the predetermined trajectory and capture high-quality images. It should be ensured that the drone can cover all target areas to avoid image blind spots. Set a suitable flight altitude and speed to obtain clear images to avoid difficulties in analysis due to image blur or distortion.

[0022] The specific steps of preprocessing the power line defect image to obtain defect surface state data and surrounding environment impact data are as follows: Acquire location information of important working areas of power lines, divide the working areas of the drone according to the location information, acquire power line defect images, perform preprocessing operations on the power line defect images and enhance the image clarity based on histogram equalization; By extracting information from the processed images, the geometric features of the defect area and the environmental information around the power line are obtained; Historical maintenance records are obtained, and correlation analysis is performed on the geometric features of the defect area and the environmental information around the power line based on the historical maintenance records to obtain defect surface state data and surrounding environment impact data.

[0023] It should be noted that the image preprocessing mainly includes denoising, image correction and image enhancement. Since the image may contain noise, especially in complex environments, the image needs to be denoised first, such as using a filtering algorithm. The distorted image is corrected to ensure that the geometric deformation of the image is repaired. Image enhancement can improve image quality and visibility of details, especially subtle defects on power lines.

[0024] S2, constructing a maintenance prediction model according to the defect surface state data and the surrounding environment impact data, and obtaining a table of repairable areas to be tested through the output of the maintenance prediction model.

[0025] In this example, the defect surface state data includes a defect surface state influence coefficient, and the surrounding environment influence data includes a surrounding environment influence coefficient.

[0026] The defect surface state influence coefficient refers to the influence data of the shape, depth and area of ​​the defect on the maintainability of the power line. By analyzing the defect shape, the most critical defect areas can be effectively identified to ensure that maintenance resources are focused on these areas first. This not only reduces unnecessary inspection and maintenance work, but also speeds up the entire maintenance process and improves work efficiency. In defect analysis, the defect surface state influence coefficient can help determine the severity of the defect and the difficulty of maintenance, avoid excessive maintenance, thereby reducing maintenance costs and unnecessary maintenance expenses.

[0027] The defect surface state influence coefficient is specifically obtained as follows: Obtaining historical maintenance records of power line defects and defect surface status information, wherein the defect surface status information includes the area of ​​the defect, the depth of the defect, and the degree of corrosion of the defect surface; Regression analysis of historical maintenance records was performed to obtain the area maintenance influencing factors and depth maintenance influencing factors; The area repair severity is calculated based on the defect area and area repair impact factor; The severity of deep repair is calculated based on the depth of the defect and the deep repair influencing factor The defect surface state influence coefficient is calculated based on the preset defect surface state influence formula according to the area repair severity and the depth repair severity combined with the corrosion degree of the defect surface.

[0028] The specific calculation formula for the area maintenance severity is as follows:

[0029] The specific calculation formula of the deep maintenance severity is as follows:

[0030] The specific calculation formula of the defect surface state influence coefficient is as follows:

[0031] In the formula, is the defect surface state influence coefficient, is the corrosion degree of the defect surface, is the corrosion maintenance rate, is the area maintenance severity, For the The area of ​​defects, where =1,2,3,...,R, R is an integer, N is the total number of defects, is the area maintenance influencing factor, is the area causing the fracture, For deep maintenance severity, is the influencing factor of deep maintenance, For the The depth of the defect, The depth that causes the fracture.

[0032] It should be noted that the area maintenance influencing factor and the depth maintenance influencing factor can be understood as the degree of influence of the area and depth of power line defects on maintenance work. By obtaining historical data, the influence of area, depth and corrosion on maintainability can be obtained based on previous maintenance records. The larger the area, depth and corrosion, the more difficult the maintenance effect and the greater the engineering workload.

[0033] The environmental impact coefficient is a quantitative assessment of external environmental conditions to help predict the complexity of maintenance. In the power line defect maintainability analysis, the various factors of the current area's surrounding environment affect the repair of power line defects. These factors may affect the stability, reliability and maintenance efficiency of power facilities, and thus determine the difficulty of maintenance, response time and maintenance cost when a fault occurs.

[0034] The specific method for obtaining the surrounding environment impact coefficient is as follows: Acquire first data of the current power line defect area, wherein the first data includes vegetation density, weather information, terrain slope, soil adhesion and population density; Extracting features based on the weather information to obtain wind speed, precipitation and temperature; The terrain data is calculated based on soil adhesion, vegetation density and terrain slope; The surrounding environment influence coefficient is calculated based on a preset surrounding environment influence formula according to the first data combined with the terrain data.

[0035] The specific calculation formula of the surrounding environment impact coefficient is as follows:

[0036] In the formula, is the surrounding environment impact coefficient, is the wind speed, is the temperature, is the population density, is the precipitation, For terrain data, , , , , is the weight coefficient.

[0037] It should be noted that , , , , The weight coefficients are the weight coefficient of wind speed, the weight coefficient of temperature, the weight coefficient of population density, the weight coefficient of precipitation, and the weight coefficient of terrain data, which are obtained through historical data respectively.

[0038] The maintenance prediction model is constructed according to the defect surface state data and the surrounding environment impact data, and the table of repairable areas to be tested is obtained through the output of the maintenance prediction model: The newly acquired defect surface state data and surrounding environment impact data are used as input items, and the maintainability prediction coefficient is used as the output item to build a maintenance prediction model; Obtaining maintainability prediction coefficients of all areas to be tested, and determining the maintainability prediction coefficients of all areas to be tested; If the maintainability prediction coefficient is greater than or equal to 1, the current area is marked as a maintainable area; If the maintainability prediction coefficient is less than 1, the current area is a non-maintainable area and is not marked; All marked power line defect areas are obtained, and all marked power line defect areas are sorted and counted to obtain a repairable area table.

[0039] The specific calculation formula of the maintenance prediction model is as follows:

[0040] In the formula, is the defect surface state influence coefficient, is the surrounding environment impact coefficient, is the weight coefficient of the defect surface state influence coefficient, is the weight coefficient of the surrounding environment impact coefficient, is the standard defect surface state influence coefficient, is the standard ambient environment impact coefficient.

[0041] It should be noted that the defect surface state influence coefficient of the maintainability prediction coefficient is affected by the surrounding environment influence coefficient. The defect surface state influence coefficient and the surrounding environment influence coefficient are getting larger and larger. When the output of the maintenance prediction model is greater than or equal to 1, it means that the area is a maintainable area.

[0042] S3, obtaining the impact data of power line defects on power transmission in the test area, and building a defect impact assessment model based on a neural network.

[0043] In this example, the impact data of power line defects on power transmission in the test area refers to the power transmission impact data and protection device response data obtained through the power line system. The power transmission impact data includes a power transmission power impact coefficient and a transmission capacity impact coefficient, and the protection device response data includes a protection device response coefficient.

[0044] The power transmission power influence coefficient refers to the fact that power line defects usually cause changes in line resistance or reactance, which directly affects the magnitude of the current and the transmission of power. Analyzing the power transmission power influence coefficient has the following advantages in improving the priority and accuracy of repairing power line defect areas: Accurately locate defective areas: The power transmission power impact coefficient can reflect the load conditions and power transmission capacity of different line sections. By analyzing the performance of power lines under different load and power conditions, it is possible to accurately locate areas where faults or defects may exist. For example, in areas with large or fluctuating transmission power, the probability of faults may be higher. Discovering potential problems in these areas in advance can reduce the possibility of faults.

[0045] Optimize maintenance sequence: According to the power transmission power impact coefficient, the lines can be prioritized. For important load sections and areas with greater power transmission impact, higher maintenance priorities can be set. In this way, when maintenance resources are limited, the focus can be placed on areas that have a greater impact on the safety and stability of the power grid, thereby optimizing maintenance decisions and improving maintenance efficiency.

[0046] Improve the reliability of the power system: Accurate analysis of the power transmission impact factor helps identify which areas are critical to the overall stability of the system, and then improve the overall reliability and risk resistance of the power grid by repairing these areas first. Timely repair of high-risk areas can reduce the possibility of large-scale power outages or fault expansion, thereby improving the operational stability of the power system.

[0047] Improve the utilization efficiency of maintenance resources: In power system maintenance, limited maintenance resources (such as manpower, material resources and time) need to be efficiently allocated. Through the analysis of the power transmission power influence coefficient, maintenance resources can be allocated more scientifically, ensuring that more resources are invested in line sections with higher fault risks and greater impacts, avoiding resource waste and ineffective repairs.

[0048] Supporting smart grid and automated operation and maintenance: In the context of smart grid and automated operation and maintenance, power transmission power impact coefficient analysis helps to combine real-time monitoring data, evaluate line conditions in real time, and automatically generate maintenance and repair suggestions. This data-driven approach can make maintenance work more intelligent and systematic, thereby improving the operation and maintenance efficiency of the entire power grid.

[0049] Improve preventive maintenance capabilities: Power impact factor analysis can not only help identify existing problems, but also perform preventive maintenance through long-term monitoring of power lines. By establishing a long-term database of power transmission and line health status, the trend of potential faults can be predicted, so that repairs and optimizations can be carried out in advance to avoid sudden faults.

[0050] In short, by analyzing the power transmission power influence coefficient, we can not only improve the priority and accuracy of repairing defective areas of power lines, but also improve the operating efficiency of the power grid, reduce fault losses, and enhance the stability of the power system, providing strong guarantees for the healthy operation of the power system.

[0051] The specific method for obtaining the power transmission power influence coefficient is as follows: Obtain the resistance, line length, normal impedance, and additional impedance caused by power line defects of the power line; The influence factor of the defect on the impedance is obtained by performing regression analysis on the additional impedance change caused by the change of line length in the power line defect; The total impedance is calculated based on the impedance of the normal line and the additional impedance; The phase angle is calculated based on the phase angle formula by combining the total impedance with the resistance of the power line; The power transmission power influence coefficient is calculated based on the preset power transmission power influence formula by obtaining the power supply terminal voltage and current and the load terminal voltage and current in combination with the phase angle.

[0052] The specific calculation formula of the phase angle is as follows:

[0053] The specific calculation formula of the power transmission power influence coefficient is as follows:

[0054] In the formula, is the power transmission power influence coefficient, is the power supply voltage, is the power supply current, is the load terminal voltage, is the load current, is the phase angle, is the resistance of the power line, is the impedance of the normal line, is the additional impedance, is the line length, is the influence factor of the defect on the impedance.

[0055] It should be noted that in impedance circuit analysis, calculating the phase angle is very important because it helps us understand the phase relationship between current and voltage. The phase angle reflects the relative time difference between voltage and current, or the delay or advance of the voltage waveform relative to the current waveform. When a line defect occurs, the faulty part increases the change in the total impedance of the line. Power line defects affect power transmission, resulting in a direct reduction in transmission power.

[0056] The transmission capacity impact coefficient is an important indicator to measure the impact of power line defects on transmission capacity. It is usually expressed as the ratio of the actual transmission capacity after the fault to the capacity under normal circumstances. Through the analysis of the transmission capacity impact coefficient, power companies can achieve more accurate and efficient maintenance scheduling and improve the safety and reliability of the power system.

[0057] Analyzing the transmission capacity impact coefficient has the following advantages in improving the priority and accuracy of repairing defective areas of power lines: Improve maintenance priority and optimize resource allocation: The transmission capacity impact coefficient reflects the load conditions and failure probability of power lines in different areas. By quantifying the transmission capacity and failure probability of the line, it can provide a more scientific basis for maintenance work and help dispatchers prioritize those areas that may affect large-scale power supply or cause serious economic losses. For example, when the line load in a certain area is close to or exceeds its maximum transmission capacity, the risk of failure is greater. At this time, maintenance and inspection can be prioritized to avoid catastrophic failures.

[0058] Reduce power outage duration and impact range: For power systems, the analysis of transmission capacity impact coefficients can effectively predict which areas of line faults will lead to a wider range of power outages or a longer power outage. By optimizing the maintenance sequence and strategy, the power outage time can be shortened, the risk of fault spread can be reduced, and the negative impact on residents' lives and industrial production can be reduced. For example, by analyzing the transmission capacity and load of each line, resources can be invested first in those lines with higher loads and transmission capacity close to saturation, thereby reducing the impact of power outages on the social economy.

[0059] Improve the accuracy of maintenance: Through the transmission capacity impact coefficient, the bottlenecks and fault-prone areas of the line can be accurately identified, thus helping maintenance personnel to more accurately identify the problem areas. This targeted analysis avoids ineffective inspections and unnecessary maintenance, saving manpower and material resources.

[0060] Improve emergency response capabilities: When the power system encounters sudden failures or natural disasters, accurate maintenance priority analysis can help dispatchers make quick decisions to ensure that power can be restored to the faulty area as soon as possible, reducing the impact of catastrophic power outages. The transmission capacity impact factor helps to prioritize the repair of high-load, high-risk lines in emergency situations to ensure that the system quickly recovers and stabilizes.

[0061] For example, if a line is loaded beyond its designed capacity, the likelihood of failure increases, and identifying and repairing it early can reduce future major failures.

[0062] The transmission capacity impact coefficient is specifically obtained in the following manner: Obtain the loss current, loss voltage and capacitor compensation capacity of a single power grid during the transmission of defective power lines; Obtaining a loss capacity of a single power grid based on the loss current and the loss voltage; Acquire the grid layout of the power lines, and extract features of the grid layout to obtain the grid density and the standard capacity of the power lines; Obtaining actual loss capacity of power lines based on the loss capacity of a single power grid according to the power grid density; The transmission capacity impact coefficient is obtained by comparing and analyzing the actual loss capacity of the power line with the standard capacity of the power line.

[0063] The transmission capacity impact coefficient is specifically calculated as follows:

[0064] In the formula, is the transmission capacity impact coefficient, is the loss current of the jth grid, is the loss voltage of the jth power grid, where j=1,2,3,...,R, R is an integer, is the capacitor compensation capacity, It is the standard capacity of power lines.

[0065] It should be noted that the capacity loss of the power grid wires is not caused by a single wire, but by the simultaneous capacity loss of multiple wires distributed on the power grid.

[0066] The protection device response factor refers to the fact that power line defects have a significant impact on the stability of the power system. In particular, when a fault occurs, the response time of the protection system and the fault clearing time are key factors in determining whether the power system can operate stably. In order to ensure the stability of the system, the protection device must quickly identify and clear the fault. There is a complex relationship between the stability of the system and the fault clearing time, load and power demand.

[0067] Analyzing the protection device response coefficient can improve the priority and accuracy of power line defect area maintenance, which has the following advantages: Priority allocation: By analyzing the response coefficient of the protection device, it is possible to provide priority judgment for the maintenance of the fault area. For example, when multiple faults occur, areas with lower response coefficients may require more urgent repairs because these areas may be caused by overload or equipment aging, which may trigger more chain reactions. Through intelligent scheduling and priority maintenance strategies, maintenance resources can be effectively allocated to avoid waste of maintenance personnel and equipment resources, reduce maintenance time and costs, and ensure that the most critical fault areas are handled in a timely manner.

[0068] Improve the stability and safety of the power system: Analyzing the response coefficient of the protection device of each line can provide a reference for the setting of the protection device. By reasonably adjusting the setting parameters of the protection device, it can be ensured that it can respond in time when a fault occurs, and cut off the faulty part according to priority to protect the power system from excessive interference. Improve the overall stability and safety of the power system, reduce the probability of power supply interruption due to faults, and effectively prevent large-scale power outages at the system level.

[0069] Improve the accuracy of fault location: The response coefficient of the protection device refers to the sensitivity or speed of the protection device's response when a power line fault occurs. By monitoring and analyzing the response coefficients of protection devices in different areas, the location of the fault can be identified more accurately. For example, in the event of a short circuit, grounding or overload, a protection device with a higher response coefficient can be triggered more quickly, thereby helping to determine the location of the fault point. The accuracy of fault location is improved, avoiding delays in fault handling caused by delayed response or misjudgment in traditional methods. Through accurate positioning, maintenance time can be shortened, the scope of power outages can be reduced, and the recovery efficiency of the system can be improved.

[0070] Reduce operation and maintenance costs and response time: Improved fault detection accuracy and analysis of response coefficients can monitor and diagnose in real time during power system operation, reducing maintenance delays or misjudgments caused by traditional inspections or inaccurate manual judgments. Especially in some more complex power line systems, fast and accurate fault location can save a lot of maintenance costs. It reduces the number of manual inspections and maintenance, saves maintenance personnel's scheduling time and transportation costs. At the same time, through real-time monitoring and automated analysis, the time for power system recovery can be significantly shortened and operational efficiency can be improved.

[0071] The specific method for obtaining the protection device response coefficient is as follows: By monitoring the time from when a fault occurs to when the protection device detects the fault and the time from when a fault occurs to when the fault is cleared, the fault detection time and fault clearing time are obtained; Obtain the time required for the system to return to stability after the fault is removed, that is, the system recovery time, and the probability that the protection device performs correct action; The fault severity coefficient is obtained by analyzing the impact of the fault based on historical protection data; The protection device response coefficient is calculated based on a preset protection device response formula according to the fault detection time and the fault clearing time combined with the fault severity coefficient.

[0072] The specific calculation formula of the protection device response coefficient is as follows:

[0073] In the formula, is the protection device response coefficient, is the fault detection time, is the fault clearing time, is the system recovery time, , , They are the proportions of fault detection time, fault clearing time, and system recovery time, respectively. The probability that the protective device performs correct action, is the fault severity factor.

[0074] It should be noted that the fault detection time reflects the time required for the protection device to issue an alarm from the occurrence of a fault. Generally, a shorter detection time can improve the response speed of the system, but may require higher-performance equipment and more complex algorithms. Fault clearing time refers to the time required for the protection device to cut off the fault line after fault detection. This time includes the action time of the circuit breaker and communication delay. Reducing the fault clearing time is crucial to quickly restore system stability. System recovery time The time required for the system to return to stability, usually the time to restore system balance after the protection device removes the fault. For large-scale power systems, rapid recovery is the key to avoiding chain reactions and large-scale power outages. Reliability of protection devices. Fault severity refers to the scope and extent of the impact of the fault. The more severe the fault, the larger the impact range, which may cause more equipment damage or longer power outages. Therefore, the weight of the severity of the fault in the response factor should be appropriately increased. The reliability of the protection device and the fault severity factor indicate the effectiveness of the protection system in different environments. When the reliability is high and the fault severity is low, the response factor is small, indicating that the system can handle faults more quickly and effectively.

[0075] The specific steps of constructing the defect impact assessment model based on the neural network are as follows: Obtain historical power transmission power influence coefficient, transmission capacity influence coefficient and protection device response coefficient; Preprocess the historical data and divide the processed data into training set and test set; The model is trained using the training set and an initial model is built based on the neural network. The initial model is adjusted using the test set to ultimately obtain a defect impact assessment model.

[0076] The specific calculation formula of the defect impact assessment model is as follows:

[0077] In the formula, is the defect impact assessment coefficient, is the power transmission power influence coefficient, is the transmission capacity impact coefficient, is the protection device response coefficient, is the weight coefficient of the power transmission power influence coefficient, is the weight coefficient of the transmission capacity impact coefficient, It is the weight coefficient of the protection device response coefficient.

[0078] It should be noted that the defect impact assessment model is used to evaluate the influence of the protection device response coefficient, the power transmission power influence coefficient and the transmission capacity influence coefficient on the power line defects. The larger the protection device response coefficient and the power transmission power influence coefficient, the larger the defect impact assessment coefficient is, and the greater the degree of impact of the defect on the power line.

[0079] S4, comparing and analyzing the output of the defect impact assessment model based on the preset risk value to obtain a defect priority processing table, applying the defect priority processing table to the repairable test area table to prioritize the test areas, and the specific steps are as follows: Obtain the output of the defect impact assessment model for all areas to be tested, and compare and analyze the output of the defect impact assessment model with the preset risk value; If the output of the defect impact assessment model is greater than the preset risk value, the risk level of the power line defect in the current area is marked as high; If the output of the defect impact assessment model is equal to the preset risk value, the risk level of the power line defects in the current area is marked as medium; If the output of the defect impact assessment model is less than the preset risk value, the risk level of the power line defect in the current area is marked as low; Based on the bubble sorting, the output of the defect impact assessment model within each risk level is used to obtain a high risk level priority table, a medium risk level priority table, and a low risk level priority table; Sort the high risk priority table, the medium risk priority table and the low risk priority table in the order of high, medium and low to obtain the final defect priority processing table; Apply the defect priority processing table to the repairable test area table, and process the repairable test areas in the repairable test area table in priority order according to the defect priority processing table, i.e., high, medium, and low risk areas; And repairs are given priority to high-risk areas according to the sorting order of areas in the high-risk priority table, and then repairs are given priority to high-risk areas according to the sorting order of areas in the medium-risk priority table, and finally repairs are given priority to high-risk areas according to the sorting order of areas in the low-risk priority table.

[0080] It should be noted that bubble sort means repeatedly exchanging adjacent elements, gradually "bubbling" the largest element to the front or the smallest element to the back, here it means putting the largest element in the front.

[0081] Embodiment 2, Figure 2A schematic diagram of the structure of a power line defect reporting system based on drone images provided in an embodiment of the present application includes an image surface data acquisition module, a repairable test area screening module, a defect impact assessment model construction module, and a test area priority processing module. There are connections between the modules: The image surface data acquisition module is used to divide the power line into several test areas, obtain the power line defect image through the drone, and pre-process the power line defect image to obtain the defect surface state data and surrounding environment impact data; A repairable area to be tested screening module is used to construct a repair prediction model based on the defect surface state data and the surrounding environment impact data, and obtain a repairable area to be tested table through the output of the repair prediction model; A defect impact assessment model building module is used to obtain the impact data of power line defects on power transmission in the test area and build a defect impact assessment model based on a neural network; The test area priority processing module is used to compare and analyze the output of the defect impact assessment model based on a preset risk value to obtain a defect priority processing table, apply the defect priority processing table to the repairable test area table, and give priority to the test area.

[0082] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0083] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments may be implemented in whole or in part in the form of a computer program product.

[0084] Those of ordinary skill in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0085] In addition, each functional module in each embodiment of the present application may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0086] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0087] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A power line defect report assessment based on drone images, characterized in that: The steps include: The power line is divided into several test areas, and the power line defect images are obtained by drones. The power line defect images are preprocessed to obtain the defect surface state data and surrounding environment impact data; Building a maintenance prediction model based on the defect surface state data and the surrounding environment impact data, and obtaining a table of repairable areas to be tested through the output of the maintenance prediction model; Obtain data on the impact of power line defects on power transmission in the test area, and build a defect impact assessment model based on a neural network; The output of the defect impact assessment model is compared and analyzed based on a preset risk value to obtain a defect priority processing table, and the defect priority processing table is applied to the repairable area to be tested table to give priority to the areas to be tested.

2. The method for evaluating power line defect reports based on drone images according to claim 1 is characterized in that: The specific steps of preprocessing the power line defect image to obtain defect surface state data and surrounding environment impact data are as follows: Acquire location information of important working areas of power lines, divide the working areas of the drone according to the location information, acquire power line defect images, perform preprocessing operations on the power line defect images and enhance the image clarity based on histogram equalization; By extracting information from the processed images, the geometric features of the defect area and the environmental information around the power line are obtained; Historical maintenance records are obtained, and correlation analysis is performed on the geometric features of the defect area and the environmental information around the power line based on the historical maintenance records to obtain defect surface state data and surrounding environment impact data.

3. The method for evaluating power line defect reports based on drone images according to claim 2 is characterized in that: The defect surface state data includes a defect surface state influence coefficient, and the defect surface state influence coefficient is specifically obtained as follows: Obtaining historical maintenance records of power line defects and defect surface status information, wherein the defect surface status information includes the area of ​​the defect, the depth of the defect, and the degree of corrosion of the defect surface; Regression analysis of historical maintenance records was performed to obtain the area maintenance influencing factors and depth maintenance influencing factors; The area repair severity is calculated based on the defect area and area repair impact factor; The severity of deep repair is calculated based on the depth of the defect and the deep repair influencing factor The defect surface state influence coefficient is calculated based on the preset defect surface state influence formula according to the area repair severity and the depth repair severity combined with the corrosion degree of the defect surface.

4. The method of evaluating power line defect reports based on drone images according to claim 1 is characterized in that: The maintenance prediction model is constructed according to the defect surface state data and the surrounding environment impact data, and the table of repairable areas to be tested is obtained through the output of the maintenance prediction model, specifically: The newly acquired defect surface state data and surrounding environment impact data are used as input items, and the maintainability prediction coefficient is used as the output item to build a maintenance prediction model; Obtaining maintainability prediction coefficients of all areas to be tested, and determining the maintainability prediction coefficients of all areas to be tested; If the maintainability prediction coefficient is greater than or equal to 1, the current area is marked as a maintainable area; If the maintainability prediction coefficient is less than 1, the current area is a non-maintainable area and is not marked; All marked power line defect areas are obtained, and all marked power line defect areas are sorted to obtain a repairable area table.

5. The method for evaluating power line defect reports based on drone images according to claim 1 is characterized in that: The data on the influence of the power line defects on the power transmission in the test area include the power transmission influence data and the protection device response data. The power transmission influence data include the power transmission power influence coefficient and the transmission capacity influence coefficient. The specific acquisition method of the power transmission power influence coefficient is as follows: Obtain the resistance, line length, normal impedance, and additional impedance caused by power line defects of the power line; The influence factor of the defect on the impedance is obtained by performing regression analysis on the additional impedance change caused by the change of line length in the power line defect; The total impedance is calculated based on the impedance of the normal line and the additional impedance; The phase angle is calculated based on the phase angle formula by combining the total impedance with the resistance of the power line; The power transmission power influence coefficient is calculated based on the preset power transmission power influence formula by obtaining the power supply terminal voltage and current and the load terminal voltage and current in combination with the phase angle.

6. The method of evaluating power line defect reports based on drone images according to claim 5 is characterized in that: The protection device response data includes a protection device response coefficient. The specific acquisition method of the protection device response coefficient is as follows: By monitoring the time from when a fault occurs to when the protection device detects the fault and the time from when a fault occurs to when the fault is cleared, the fault detection time and fault clearing time are obtained; Obtain the time required for the system to return to stability after the fault is removed, that is, the system recovery time, and the probability that the protection device performs correct action; The fault severity coefficient is obtained by analyzing the impact of the fault based on historical protection data; The protection device response coefficient is calculated based on a preset protection device response formula according to the fault detection time and the fault clearing time combined with the fault severity coefficient.

7. The method of evaluating power line defect reports based on drone images according to claim 1 is characterized in that: The output of the defect impact assessment model is compared and analyzed based on a preset risk value to obtain a defect priority processing table, and the defect priority processing table is applied to the repairable test area table to give priority to the test area. The specific steps are as follows: Obtain the output of the defect impact assessment model for all areas to be tested, and compare and analyze the output of the defect impact assessment model with the preset risk value; If the output of the defect impact assessment model is greater than the preset risk value, the risk level of the power line defect in the current area is marked as high; If the output of the defect impact assessment model is equal to the preset risk value, the risk level of the power line defects in the current area is marked as medium; If the output of the defect impact assessment model is less than the preset risk value, the risk level of the power line defect in the current area is marked as low; Based on the bubble sorting, the output of the defect impact assessment model within each risk level is used to obtain a high risk priority list, a medium risk priority list, and a low risk priority list; Sort the high risk priority table, the medium risk priority table and the low risk priority table in the order of high, medium and low to obtain the final defect priority processing table; Apply the defect priority processing table to the repairable test area table, and process the repairable test areas in the repairable test area table in priority order according to the defect priority processing table, i.e., high, medium, and low risk areas; And repair the high-risk defect areas first according to the sorting order of the areas in the high-risk priority table, then repair the medium-risk defect areas first according to the sorting order of the areas in the medium-risk priority table, and finally repair the low-risk defects first according to the sorting order of the areas in the low-risk priority table.

8. The method of evaluating power line defect reports based on drone images according to claim 3 is characterized in that: The specific calculation formula for the area maintenance severity is as follows: The specific calculation formula of the deep maintenance severity is as follows: The specific calculation formula of the defect surface state influence coefficient is as follows: In the formula, is the defect surface state influence coefficient, is the corrosion degree of the defect surface, is the corrosion maintenance rate, is the area maintenance severity, For the The area of ​​defects, where =1,2,3,...,R, R is an integer, N is the total number of defects, is the area maintenance influencing factor, is the area causing the fracture, For deep maintenance severity, is the influencing factor of deep maintenance, For the The depth of the defect, The depth that causes the fracture.

9. The method of evaluating power line defect reports based on drone images according to claim 5 is characterized in that: The specific calculation formula of the phase angle is as follows: The specific calculation formula of the power transmission power influence coefficient is as follows: In the formula, is the power transmission power influence coefficient, is the power supply voltage, is the power supply current, is the load terminal voltage, is the load current, is the phase angle, is the resistance of the power line, is the impedance of the normal line, is the additional impedance, is the line length, is the influence factor of the defect on the impedance.

10. A system for evaluating power line defect reports based on drone images as described in any one of claims 1 to 9, characterized in that: It includes image surface data acquisition module, repairable test area screening module, defect impact assessment model building module and test area priority processing module. There are connections between the modules: The image surface data acquisition module is used to divide the power line into several test areas, obtain the power line defect image through the drone, and pre-process the power line defect image to obtain the defect surface state data and surrounding environment impact data; A repairable area to be tested screening module is used to construct a repair prediction model based on the defect surface state data and the surrounding environment impact data, and obtain a repairable area to be tested table through the output of the repair prediction model; A defect impact assessment model building module is used to obtain the impact data of power line defects on power transmission in the test area and build a defect impact assessment model based on a neural network; The test area priority processing module is used to compare and analyze the output of the defect impact assessment model based on a preset risk value to obtain a defect priority processing table, apply the defect priority processing table to the repairable test area table, and give priority to the test area.

Images