Intelligent inspection method and system for power transmission line

Abstract

The invention discloses a power transmission line intelligent inspection method and system, and relates to the technical field of intelligent inspection, a to-be-inspected power transmission line is recorded as a first power transmission line, and each power transmission line inspected by an unmanned aerial vehicle is recorded as a historical power transmission line set; acquiring state data of the first power transmission line, acquiring state data of the historical power transmission line set, and performing matching to determine the historical power transmission line set similar to the state of the first power transmission line as a reference power transmission line. Extracting a damaged area to calculate an appearance damage coefficient, evaluating a fault hidden danger in combination with a temperature anomaly coefficient, and judging whether the first power transmission line has a fault risk or not, thereby ensuring higher pertinence and accuracy of the inspection of the power transmission line which is not inspected by the unmanned aerial vehicle, and improving the maintenance efficiency and safety of the power transmission line; in addition, whether the power transmission line has a fault can be intelligently predicted according to an inspection result, so that the inspection efficiency is relatively high and the fault is timely processed.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent inspection, and particularly relates to an intelligent inspection method and system for transmission lines. Background Art

[0002] The intelligent inspection of transmission lines is mainly realized through drones, robots, intelligent sensors and artificial intelligence technologies. Among them, it is more convenient to use drones to inspect transmission lines because drones can conduct real-time monitoring of transmission lines at high altitudes, carry devices such as high-definition cameras and infrared sensors, quickly identify whether there are damages on the surface of the lines and the temperature of the transmission lines, timely identify potential faults or hidden dangers, and transmit them to the monitoring center through a wireless network; combined with artificial intelligence algorithms, analyze and process the collected data, predict potential problems and provide maintenance suggestions, so as to improve the inspection efficiency, reduce labor costs, and ensure the stable operation of the power system.

[0003] However, for transmission lines that have not been inspected by drones, the existing technologies cannot intelligently predict whether there are faults in the transmission lines according to the inspection results, resulting in low inspection efficiency and lagging fault handling. Summary of the Invention

[0004] The purpose of the present invention is to solve the above-mentioned problems and provide an intelligent inspection method and system for transmission lines.

[0005] In the first aspect of the implementation of the present invention, an intelligent inspection method for transmission lines is first proposed, including the following steps:

[0006] Obtain the status data of the first transmission line and the historical transmission line set, and match and determine the transmission line in the historical transmission line set with a similarity greater than the first preset threshold to the status data of the first transmission line as the reference transmission line; the first transmission line is the transmission line to be inspected, and the historical transmission line set is the transmission lines that have been inspected by drones;

[0007] Use the same type of drone as used during the inspection of the reference transmission line to inspect the first transmission line, and obtain the high-resolution image and thermal imaging image of the first transmission line;

[0008] Extract the damaged area of the transmission line from the high-resolution image, and calculate the surface damage coefficient of the first transmission line in combination with the preset transmission line damage image set; the surface damage coefficient is used to represent the degree of damage to the surface of the transmission line; the damage image set contains transmission line damage images corresponding to different types of transmission line faults;

[0009] Obtain the temperature anomaly coefficient of the first transmission line based on the thermal imaging image of the first transmission line, and determine whether there are potential faults in the first transmission line according to the appearance damage coefficient and the temperature anomaly coefficient; the temperature anomaly coefficient is used to represent the degree of temperature anomaly of the transmission line.

[0010] Optionally, the status data has multiple data types, specifically including terrain data, tower base data, environmental climate data, and operation data, where:

[0011] The terrain data includes the maximum height difference, the average value, the maximum value, the highest elevation point, and the lowest elevation point of the slope along the line.

[0012] The tower base data includes the distribution density of the tower bases, the percentage of straight towers, and the percentage of corner towers.

[0013] The environmental climate data includes the annual average temperature, the annual average humidity, the annual precipitation, the annual number of thunderstorm days, and the maximum wind speed.

[0014] The operation data includes the rated voltage, the load current, and the transmission power.

[0015] Optionally, the method for determining the reference transmission line set specifically includes:

[0016] Preprocess the status data of the first transmission line and the historical transmission line set.

[0017] Construct a distance matrix for the first transmission line and the target transmission line. Each element in the distance matrix is the difference between the same data types of the first transmission line and the target transmission line; the target transmission line is any transmission line in the historical transmission line set.

[0018] Use the dynamic programming algorithm to calculate the shortest path in the distance matrix to find the best match between the status data sequence of the first transmission line and the status data sequence of the target transmission line.

[0019] Align the status data sequences of the first transmission line and the target transmission line according to the calculated shortest path, and calculate the similarity between the aligned status data sequences of the first transmission line and the target transmission line.

[0020] Take the transmission line with the maximum similarity in the historical transmission line set as the reference transmission line.

[0021] Optionally, extracting the damaged area of the transmission line from the high-resolution image and calculating the appearance damage coefficient of the first transmission line in combination with the preset transmission line damaged image set includes:

[0022] Convert each damaged image in the preset transmission line damaged image set into a damaged image matrix of n×n, and denote the corresponding damaged image matrix as I1.

[0023] For the high-resolution image of the first transmission line, preprocess the high-resolution image, and through image processing and feature extraction algorithms on the preprocessed image, identify and extract the damaged area in the high-resolution image of the first transmission line, and also convert the damaged area into an n×n damaged image matrix. Denote the damaged image matrix of the first transmission line as I2;

[0024] Calculate the correlation coefficient S between the damaged image of the first transmission line and the corresponding damaged image in the preset set of damaged images of transmission lines. The calculation formula is: In the formula, I1(i,j) is the pixel value at the i-th row and j-th column in the damaged image matrix of the first transmission line; I2(i,j) corresponds to the pixel value at the i-th row and j-th column in the damaged image matrix of the damaged image in the set of damaged images of transmission lines;

[0025] Take the maximum correlation coefficient between the damaged image in the preset set of damaged images and the first transmission line as the external damage coefficient of the first transmission line.

[0026] Optionally, obtain the temperature anomaly coefficient of the first transmission line based on the thermal imaging image of the first transmission line, and determine whether there are potential faults in the first transmission line according to the external damage coefficient and the temperature anomaly coefficient, including:

[0027] Obtain the thermal imaging image of the first transmission line, and extract the temperature value of each pixel point from the thermal imaging image; calculate the temperature difference between the temperature value of each pixel point and the preset reference temperature value, and take the pixel points with a temperature difference greater than the second preset threshold as abnormal pixel points; determine the ratio of the number of abnormal pixel points to the number of all pixel points as the temperature anomaly coefficient of the first transmission line;

[0028] Compare the external damage coefficient of the first transmission line with the preset external damage coefficient threshold. If the external damage coefficient of the first transmission line is not less than the preset external damage coefficient threshold, there are potential faults in the first transmission line;

[0029] Compare the temperature anomaly coefficient of the first transmission line with the preset temperature anomaly coefficient threshold. If the temperature anomaly coefficient of the first transmission line is not less than the preset temperature anomaly coefficient threshold, there are potential faults in the first transmission line;

[0030] When the temperature anomaly coefficient of the first transmission line is less than the preset temperature anomaly coefficient threshold and the external damage coefficient is less than the preset external damage coefficient threshold, perform a weighted sum of the temperature anomaly coefficient and the external damage coefficient of the first transmission line to obtain a fault risk coefficient. Compare the fault risk coefficient with the preset fault risk coefficient threshold. If it is not less than the preset fault risk coefficient threshold, it means that there are potential faults in the first transmission line; otherwise, there are no potential faults.

[0031] In the second aspect of the implementation of the present invention, an intelligent inspection system for transmission lines is proposed. The system includes:

[0032] A matching module: obtains the status data of the first transmission line and the historical transmission line set, and performs matching to determine the transmission line in the historical transmission line set with a similarity greater than the first preset threshold to the status data of the first transmission line as the reference transmission line; the first transmission line is the transmission line to be inspected, and the historical transmission line set is the transmission lines that have been inspected using drones.

[0033] An inspection module: uses the same type of drone as used during the inspection of the reference transmission line to inspect the first transmission line, and obtains high-resolution images and thermal imaging images of the first transmission line.

[0034] A damage module: extracts the damaged area of the transmission line from the high-resolution image, and calculates the surface damage coefficient of the first transmission line in combination with the preset transmission line damage image set; the surface damage coefficient is used to represent the degree of damage to the surface of the transmission line; the damage image set contains transmission line damage images corresponding to different types of transmission line faults.

[0035] A judgment module: obtains the temperature anomaly coefficient of the first transmission line based on the thermal imaging image of the first transmission line, and judges whether there are potential faults in the first transmission line according to the surface damage coefficient and the temperature anomaly coefficient; the temperature anomaly coefficient is used to represent the degree of temperature anomaly of the transmission line.

[0036] Optionally, the status data has multiple data types, specifically including terrain data, tower base data, environmental climate data, and operation data, where:

[0037] The terrain data includes the maximum height difference, the average value, maximum value, highest elevation point, and lowest elevation point of the slope along the line.

[0038] The tower base data includes the distribution density of the tower bases, the percentage of straight towers, and the percentage of corner towers.

[0039] The environmental climate data includes the annual average temperature, annual average humidity, annual precipitation, annual number of thunderstorm days, and maximum wind speed.

[0040] The operation data includes the rated voltage, load current, and transmission power.

[0041] Optionally, the matching module further includes:

[0042] A preprocessing module: preprocesses the status data of the first transmission line and the historical transmission line set.

[0043] Distance matrix construction module: Construct a distance matrix for the first transmission line and the target transmission line. Each element in the distance matrix is the difference between the same data types of the first transmission line and the target transmission line; the target transmission line is any transmission line in the historical transmission line set.

[0044] Optimal matching module: Use the dynamic programming algorithm to calculate the shortest path in the distance matrix to find the optimal match between the state data sequence of the first transmission line and the state data sequence of the target transmission line.

[0045] Alignment module: According to the calculated shortest path, align the state data sequences of the first transmission line and the target transmission line, and calculate the similarity between the aligned state data sequences of the first transmission line and the target transmission line.

[0046] Reference transmission line module: Use the transmission line with the largest similarity in the historical transmission line set as the reference transmission line.

[0047] Optionally, the damage module includes:

[0048] First damaged image matrix module: Convert each damaged image in the preset transmission line damaged image set into an n×n damaged image matrix, and denote the corresponding damaged image matrix as I1.

[0049] Second damaged image matrix module: For the high-resolution image of the first transmission line, preprocess the high-resolution image, and through image processing and feature extraction algorithms for the preprocessed image, identify and extract the damaged area in the high-resolution image of the first transmission line, and also convert the damaged area into an n×n damaged image matrix, and denote the damaged image matrix of the first transmission line as I2.

[0050] Correlation coefficient module: Calculate the correlation coefficient S between the damaged image of the first transmission line and the corresponding damaged image in the preset transmission line damaged image set. The calculation formula is: In the formula, I1(i,j) is the pixel value of the i-th row and j-th column in the damaged image matrix of the first transmission line; I2(i,j) is the pixel value of the i-th row and j-th column in the damaged image matrix corresponding to the damaged image in the transmission line damaged image set.

[0051] Appearance damage coefficient module: Use the maximum correlation coefficient between the damaged images in the preset damaged image set and the first transmission line as the appearance damage coefficient of the first transmission line.

[0052] Optionally, the judgment module includes:

[0053] Temperature anomaly coefficient module: Obtain the thermal imaging image of the first transmission line, and extract the temperature value of each pixel point from the thermal imaging image; calculate the temperature difference between the temperature value of each pixel point and the preset reference temperature value, and use the pixel points with a temperature difference greater than the second preset threshold as abnormal pixel points; determine the ratio of the number of abnormal pixel points to the number of all pixel points as the temperature anomaly coefficient of the first transmission line;

[0054] First judgment module: Compare the surface damage coefficient of the first transmission line with the preset surface damage coefficient threshold. If the surface damage coefficient of the first transmission line is not less than the preset surface damage coefficient threshold, there is a potential fault in the first transmission line;

[0055] Second judgment module: Compare the temperature anomaly coefficient of the first transmission line with the preset temperature anomaly coefficient threshold. If the temperature anomaly coefficient of the first transmission line is not less than the preset temperature anomaly coefficient threshold, there is a potential fault in the first transmission line;

[0056] Third judgment module: When the temperature anomaly coefficient of the first transmission line is less than the preset temperature anomaly coefficient threshold and the surface damage coefficient is less than the preset surface damage coefficient threshold, perform a weighted sum of the temperature anomaly coefficient and the surface damage coefficient of the first transmission line to obtain a fault risk coefficient, and compare the fault risk coefficient with the preset fault risk coefficient threshold. If it is not less than the preset fault risk coefficient threshold, it means that there is a potential fault in the first transmission line; otherwise, there is no potential fault.

[0057] Advantages of the present invention:

[0058] The present invention proposes a method and system for intelligent inspection of transmission lines. For transmission lines that have not been inspected by drones, it can perform intelligent matching based on transmission lines that have been inspected by drones and select a suitable drone to inspect the transmission lines that have not been inspected by drones. This process can utilize the drone types and experience in historical inspection data, ensuring higher pertinence and accuracy in the inspection of transmission lines that have not been inspected by drones, thereby improving the maintenance efficiency and safety of transmission lines; in addition, it can intelligently predict whether there is a fault in the transmission line based on the inspection results, resulting in higher inspection efficiency and timely fault handling. Description of the drawings

[0059] The following further describes the present invention with reference to the drawings.

[0060] Figure 1 It is a flowchart of a method for intelligent inspection of transmission lines;

[0061] Figure 2 It is a framework diagram of a system for intelligent inspection of transmission lines. Detailed implementation manners

[0062] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without making creative efforts shall fall within the protection scope of the present invention.

[0063] All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without making creative efforts shall fall within the protection scope of the present invention.

[0064] The embodiment of the present invention provides an intelligent inspection method for transmission lines. Refer to Figure 1 , Figure 1 which is a flowchart of an intelligent inspection method for transmission lines provided by the embodiment of the present invention. The method includes the following steps:

[0065] Obtain the status data of the first transmission line and the historical transmission line set, and perform matching to determine the transmission line in the historical transmission line set with a similarity greater than the first preset threshold to the status data of the first transmission line as the reference transmission line; the first transmission line is the transmission line to be inspected, and the historical transmission line set is the transmission line that has been inspected by an unmanned aerial vehicle.

[0066] Use the same type of unmanned aerial vehicle as that used during the inspection of the reference transmission line to inspect the first transmission line, and obtain the high-resolution image and thermal imaging image of the first transmission line.

[0067] Extract the damaged area of the transmission line from the high-resolution image, and calculate the external damage coefficient of the first transmission line in combination with the preset damaged image set of the transmission line; the external damage coefficient is used to represent the degree of damage to the external surface of the transmission line; the damaged image set includes the damaged images of the transmission line corresponding to different types of transmission line faults.

[0068] Obtain the temperature anomaly coefficient of the first transmission line according to the thermal imaging image of the first transmission line, and judge whether there are potential faults in the first transmission line according to the external damage coefficient and the temperature anomaly coefficient; the temperature anomaly coefficient is used to represent the degree of temperature anomaly of the transmission line.

[0069] Based on an intelligent inspection method for transmission lines provided by an embodiment of the present invention, through the above method, for transmission lines that have not been inspected by drones, it is possible to perform intelligent matching based on transmission lines that have been inspected by drones, and select a suitable drone to inspect the transmission lines that have not been inspected by drones. This process can utilize the drone types and experience in historical inspection data, ensuring that the inspection of transmission lines that have not been inspected by drones has higher pertinence and accuracy, thereby improving the maintenance efficiency and safety of transmission lines; in addition, it can intelligently predict whether there are faults in the transmission lines based on the inspection results, resulting in higher inspection efficiency and timely fault handling.

[0070] In one embodiment, obtaining the status data of the first transmission line and obtaining the status data of the historical transmission line set includes:

[0071] The status data includes the length of the transmission line, terrain data, tower base data, environmental climate data, and operation data;

[0072] The terrain data includes the maximum height difference, the average value, the maximum value, the highest elevation point, and the lowest elevation point of the slope along the line;

[0073] The tower base data includes the distribution density of the tower bases, the percentage of straight towers, and the percentage of corner towers;

[0074] The environmental climate data includes the annual average temperature, the annual average humidity, the annual precipitation, the annual number of thunderstorm days, and the maximum wind speed;

[0075] The operation data includes the rated voltage, the load current, and the transmission power.

[0076] It should be noted that the main purpose of matching the historical transmission line set similar to the first transmission line based on these data is to improve the inspection efficiency and the accuracy of the results. The status data of the transmission line directly affects the potential risks of line operation and the key points of inspection. For example, the maximum height difference, slope, and elevation in the terrain data affect the flight path planning and endurance of drones; the tower base distribution density and tower shape ratio determine the site layout and shooting frequency of the inspection; environmental climate data such as the number of thunderstorm days and wind speed can help estimate the wear degree of equipment affected by the environment, while operation data such as rated voltage and load current reflect the power carrying pressure and fault probability of the line. By matching similar lines, it is possible to use existing inspection experience, analysis models, and fault data to perform more targeted inspections on the first transmission line, avoid waste of resources, quickly identify high-risk areas, and form a more scientific and efficient inspection plan. In addition, based on the historical data of similar lines, it is also possible to better predict the trend of potential hidden dangers, reduce the later maintenance cost, and improve the safety and reliability of power grid operation.

[0077] It should be noted that the status data of the first transmission line is generally obtained by real-time collection and retrieval of existing databases. Specifically, terrain data can be obtained through geographic information systems (GIS) and satellite remote sensing technology; tower base data comes from the records of transmission line design drawings or line management systems; environmental climate data is obtained through historical and real-time monitoring data of meteorological departments; operation data can be obtained in real-time from the power grid dispatching center or monitoring sensors on the transmission line. For the historical transmission line set, since it has been inspected by drones, its data mainly comes from high-resolution images, thermal imaging images taken during drone inspections, and operation data recorded by intelligent sensors installed on the transmission line. These data are analyzed and processed and stored in the power grid management system to form a comprehensive inspection history record library. This method ensures the real-time and accuracy of the first transmission line, and at the same time uses the historical data of the historical transmission line set to provide a reliable basis for the status matching of the first transmission line.

[0078] In one embodiment, the method for determining the reference transmission line set specifically includes:

[0079] Preprocess the status data of the first transmission line and the historical transmission line set;

[0080] Construct a distance matrix for the first transmission line and the target transmission line. Each element in the distance matrix is the difference between the same data types of the first transmission line and the target transmission line; the target transmission line is any transmission line in the historical transmission line set;

[0081] Use the dynamic programming algorithm to calculate the shortest path in the distance matrix to find the best match between the status data sequence of the first transmission line and the status data sequence of the target transmission line;

[0082] According to the calculated shortest path, align the status data sequences of the first transmission line and the target transmission line, and calculate the similarity between the aligned status data sequences of the first transmission line and the target transmission line;

[0083] Take the transmission line with the maximum similarity in the historical transmission line set as the reference transmission line.

[0084] It should be noted that for preprocessing the status data: Before performing the matching, it is necessary to preprocess the status data of the first transmission line and the historical transmission line set. This may include data cleaning, normalization processing (such as unit unification, outlier processing), and filling in missing data. Ensure that the data of different lines can be compared on the same scale, thereby improving the accuracy of the matching.

[0085] Construct a distance matrix: A 4×4 distance matrix is constructed between two lines. The elements in the distance matrix are the differences between the same data types in the state data, specifically including 16 types of state data such as the maximum height difference, the average value, the maximum value, the highest elevation point, the lowest elevation point, the distribution density of tower bases, the percentage of straight towers, the percentage of corner towers, the annual average temperature, the annual average humidity, the annual precipitation, the annual number of thunderstorm days, the maximum wind speed, the rated voltage, the load current, and the transmission power. This is used to quantify the differences between the first transmission line and each transmission line in the historical transmission line set at each state data point. Each matrix element represents the distance (e.g., Euclidean distance or Manhattan distance) between two lines in a certain state data dimension. In this way, the differences between the two lines in terms of terrain, climate, tower base distribution, operating data, etc. can be comprehensively reflected.

[0086] Calculate the shortest path using the dynamic programming algorithm: The dynamic programming algorithm finds the best matching path between the state data sequence of the first transmission line and the state data sequence of the historical transmission line set by minimizing the distance matrix. This algorithm efficiently matches by gradually comparing the distances between different state data points to find the matching order with the strongest similarity in multiple dimensions;

[0087] When finding the shortest path using the dynamic programming algorithm, first construct a two-dimensional "cost matrix". Each element of the matrix represents the cumulative minimum "distance" or "cost" from a certain state data point of the first transmission line to a certain state data point of the historical transmission line set. The core idea of dynamic programming is to start from the upper left corner of the matrix and gradually calculate the value of each position until the lower right corner. The calculation of each position depends on the values on the left, above, and upper left, and the minimum path and cost are selected. Finally, the value in the lower right corner is the minimum matching cost of the state data sequences of the two lines, and the best matching alignment scheme can be found by backtracking the path;

[0088] Align the state data sequences: According to the shortest path obtained by dynamic programming, align the state data of the first transmission line and the historical transmission line set. This alignment process enables the time series and spatial series of the state data of the two lines to be directly compared in the same framework, facilitating subsequent similarity calculations;

[0089] The meaning of data alignment is to synchronize the status data of two transmission lines in the best-matching way so that the corresponding data points can correspond as much as possible. Specifically, after calculating the shortest path through dynamic programming, the status data of the two lines are adjusted according to this path so that they can correspond in time or space, reducing the deviation caused by time delay or different data characteristics. The relationship between the shortest path and similarity is that the shortest path represents the minimum difference between the status data of the two lines. By calculating the shortest path, the matching method with the smallest difference after alignment of the status sequences of the two lines can be found, thereby obtaining the similarity between the two lines. Generally, the shorter the shortest path, the more similar the status data of the two lines are, and vice versa, indicating a larger difference.

[0090] Calculate similarity: Once the status data alignment is completed, the similarity between the two lines can be calculated. The similarity calculation method can adopt weighted average, cosine similarity, etc., comprehensively considering the differences in each data dimension. Generally, the higher the similarity calculated on the aligned data sequence, the more similar the status of the two lines is.

[0091] Determine the reference transmission line: Finally, select the historical transmission line set corresponding to the calculated maximum similarity as the reference transmission line. This reference transmission line represents the line that is most similar to the first transmission line in terms of status data and can be used to provide a reference for the inspection strategy of the first transmission line. Through this matching process, the accuracy of inspection can be greatly improved, and potential fault hazards can be avoided.

[0092] In one implementation method, through this series of steps, a more accurate scientific reference and guidance can be provided for new inspection tasks, thereby improving the inspection efficiency and maintenance management level of transmission lines.

[0093] In one embodiment, the type of unmanned aerial vehicle (UAV) used during the inspection of the reference transmission line is taken as the target UAV type, and the UAV corresponding to the target UAV type is used to inspect the first transmission line to obtain high-resolution images and thermal imaging images of the first transmission line.

[0094] Determine the target UAV type: By taking the reference transmission line as the line with the most similar status to the first transmission line, the historical inspection data of the reference transmission line can be used to select the most suitable UAV type. The inspection data of the reference transmission line already record the types of UAVs used, such as fixed-wing UAVs, multi-rotor UAVs, or hybrid UAVs, etc. According to the inspection experience of the reference transmission line, select a UAV type that is most suitable for the environment and characteristics of the first transmission line.

[0095] Select a suitable UAV for inspection: Based on the geographical location, environmental climate, terrain features, etc. of the first transmission line, match the type of UAV used during historical inspections. These factors determine the flight ability, endurance, wind resistance, etc. of the UAV. For example, for complex mountainous areas or long-distance transmission lines, a fixed-wing UAV may be selected; while for relatively flat or variable terrains, multi-rotor UAVs are suitable for local high-precision inspections.

[0096] Conduct inspections and obtain image data: After selecting the appropriate target UAV, use it to inspect the first transmission line to obtain high-resolution images and thermal imaging images. High-resolution images help identify damages, abrasions, or any other obvious physical problems on the appearance of the line, while thermal imaging images can detect temperature anomalies on the line, helping to discover potential overheating problems or current abnormalities.

[0097] Data transmission and analysis: After the inspection, the UAV transmits the obtained high-resolution images and thermal imaging images to the monitoring center or data processing system via a wireless network. Through artificial intelligence algorithms and image recognition technologies, the image data is analyzed in real time, and areas with anomalies, such as line damages, conductor heating, insulation damage, etc., are automatically marked, further assisting engineers in determining whether there are potential fault hazards.

[0098] In one implementation method, through this process, with the help of the UAV type and experience in historical inspection data, it can be ensured that the inspection of the first transmission line has higher pertinence and accuracy, thereby improving the maintenance efficiency and safety of the transmission line.

[0099] In one embodiment, extracting the damaged area of the transmission line from the high-resolution image and calculating the external damage coefficient of the first transmission line in combination with a preset set of damaged images of the transmission line includes:

[0100] Convert each damaged image in the preset set of damaged images of the transmission line into an n×n damaged image matrix, and denote the corresponding damaged image matrix as I1;

[0101] For the high-resolution image of the first transmission line, preprocess the high-resolution image, and through image processing and feature extraction algorithms on the preprocessed image, identify and extract the damaged area in the high-resolution image of the first transmission line, and also convert the damaged area into an n×n damaged image matrix, and denote the damaged image matrix of the first transmission line as I2;

[0102] Calculate the correlation coefficient S between the damaged image of the first transmission line and the corresponding damaged image in the preset set of damaged images of the transmission line. The calculation formula is: Wherein, I1(i,j) is the pixel value of the i-th row and j-th column in the damaged image matrix of the first transmission line; I2(i,j) corresponds to the pixel value of the i-th row and j-th column in the damaged image matrix in the set of damaged images of the transmission line;

[0103] Take the maximum correlation coefficient between the damaged images in the preset set of damaged images and the first transmission line as the external damage coefficient of the first transmission line.

[0104] It should be noted that the value range of the calculation result of S is [0,1].

[0105] It should be noted that for image preprocessing: Process the acquired images, including operations such as denoising, enhancing contrast, and adjusting brightness, to improve the image quality for subsequent processing.

[0106] Feature extraction algorithm: Use computer vision techniques (such as edge detection, region growing algorithm, image segmentation, etc.) to extract the damaged areas in the transmission line images. Common algorithms include Canny edge detection, Hough transform, deep learning models (such as convolutional neural network CNN), etc., which can automatically identify the damaged areas.

[0107] Image matrix conversion: Convert the extracted damaged areas into image matrices. Each pixel value represents the gray value or color value of that area in the image. This can convert the image information into a digital matrix for subsequent calculation and matching.

[0108] It should be noted that obtaining the high-resolution image and thermal imaging image of the first transmission line is mainly achieved through drone inspection. The specific steps are as follows: Drone inspection: The drone is equipped with a high-resolution camera and an infrared thermal imager and flies above the transmission line for shooting. High-resolution camera: Used to take clear images of the appearance of the transmission line, including details of structures such as wires and tower bases, to identify damaged areas. Infrared thermal imager: Used to detect the temperature distribution of the line, especially to assist in discovering possible electrical faults by identifying temperature anomalies in the line. Image acquisition and transmission: The high-resolution images and thermal imaging images taken by the drone are transmitted to the ground monitoring center in real time for subsequent image processing and analysis;

[0109] The preset power transmission line damage image set is usually obtained through long-term drone inspections, manual inspections, and the accumulation of historical fault data. These image sets contain different types of power transmission line fault samples, such as common problems like insulator damage, wire breakage, tower base loosening, corrosion, icing, etc. To ensure the comprehensiveness of the data, the image sets cover the power transmission line damage situations under various climate conditions, terrains, and different seasons, thus forming a representative database. During drone inspections, images of various angles of the line are taken through high-resolution cameras, and different types of damage features are extracted through image processing algorithms. These images are labeled to ensure the accuracy of each damaged area and are classified and archived according to their severity. After years of accumulation, these image sets provide valuable reference data for the condition assessment and fault prediction of power transmission lines, enabling the system to accurately identify and predict potential fault hazards; the historical damaged images are labeled through manual annotation or machine learning methods so that they can be compared with the images of the first power transmission line and the similarity can be calculated.

[0110] It should be noted that the external damage coefficient of the first power transmission line is used to measure the similarity between the external damage of the first power transmission line and the damaged images in the preset power transmission line damage image set; if the external damage coefficient is larger, it means that the external damage of the first power transmission line is more similar to the damaged images in the preset power transmission line damage image set, and the coincidence degree is higher. At this time, it indicates that the first power transmission line may have a fault. Therefore, the possibility of predicting the fault of the external damage coefficient of the first power transmission line is greater.

[0111] In one implementation method, the benefits of analyzing the external damage coefficient for predicting the fault of the external damage coefficient of the first power transmission line are as follows: Analyzing the external damage coefficient has important benefits for predicting the fault of the external damage coefficient of the first power transmission line. First, by comparing the high-resolution images of the first power transmission line with the damaged images of historical fault lines, potential damaged areas can be accurately identified, and the damage degree of these areas can be quantified to form an external damage coefficient for evaluation. This quantitative index not only helps to accurately judge the current damage condition of the line but also can predict in advance whether there are possible fault hazards based on historical data comparison. Second, the calculation of the external damage coefficient can capture some subtle but key damage features by comparing with the damaged images in the preset power transmission line damage image set, thereby helping to predict the possible faults that the power transmission line may occur at a certain future moment. This prediction ability can provide strong decision-making support for maintenance personnel, help carry out maintenance in advance, optimize maintenance strategies, and effectively reduce the occurrence frequency of sudden faults, thus improving the safety and stability of power transmission lines.

[0112] In one embodiment, obtaining the temperature anomaly coefficient of the first transmission line based on the thermal imaging image of the first transmission line, and determining whether there are potential fault hazards in the first transmission line according to the appearance damage coefficient and the temperature anomaly coefficient includes:

[0113] Obtain the thermal imaging image of the first transmission line, and extract the temperature value of each pixel point from the thermal imaging image; calculate the temperature difference between the temperature value of each pixel point and the preset reference temperature value, and take the pixel points with the temperature difference greater than the second preset threshold as abnormal pixel points; determine the ratio of the number of abnormal pixel points to the number of all pixel points as the temperature anomaly coefficient of the first transmission line;

[0114] Calculate the difference between the temperature value of each pixel point and the preset reference temperature value as the temperature anomaly value of the corresponding pixel point; calculate the average value of the temperature anomaly values of all pixel points to obtain the temperature anomaly coefficient of the first transmission line;

[0115] Compare the appearance damage coefficient of the first transmission line with the preset appearance damage coefficient threshold. If the appearance damage coefficient of the first transmission line is not less than the preset appearance damage coefficient threshold, there are potential fault hazards in the first transmission line;

[0116] Compare the temperature anomaly coefficient of the first transmission line with the preset temperature anomaly coefficient threshold. If the temperature anomaly coefficient of the first transmission line is not less than the preset temperature anomaly coefficient threshold, there are potential fault hazards in the first transmission line;

[0117] When the temperature anomaly coefficient of the first transmission line is less than the preset temperature anomaly coefficient threshold and the appearance damage coefficient is less than the preset appearance damage coefficient threshold, perform a weighted sum of the temperature anomaly coefficient and the appearance damage coefficient of the first transmission line to obtain a fault risk coefficient, and compare the fault risk coefficient with the preset fault risk coefficient threshold. If it is not less than the preset fault risk coefficient threshold, it means that there are potential fault hazards in the first transmission line; otherwise, there are none.

[0118] It should be noted that the thermal imaging image records the surface temperature distribution of the transmission line and can show temperature anomalies in different regions; perform preprocessing operations such as noise removal and image enhancement on the thermal imaging image to improve the image quality and reduce interference information. Usually, each pixel in the image corresponds to a certain temperature value, and these temperature values will be used as the basis for subsequent calculations. The preset reference temperature value is determined according to historical data or preset standards for the normal temperature of the transmission line corresponding to the pixel temperature value of the thermal imaging image;

[0119] It should be noted that the preset appearance damage coefficient threshold, the preset temperature anomaly coefficient threshold, and the preset fault risk coefficient threshold are all set by professionals according to the actual situation, and are not specifically limited and elaborated here.

[0120] It should be noted that the formula for obtaining the fault risk coefficient by weighted summation of the temperature anomaly coefficient and the appearance damage coefficient of the first transmission line is: Gz = α×Gh + β×S, where α and β respectively represent the preset proportionality coefficients of the temperature anomaly coefficient and the appearance damage coefficient, and both α and β are greater than 0; α and β are set according to the actual situation. For example, the expert weighting method is adopted, that is, experts in the relevant field are invited to determine the preset proportionality coefficients of each index through professional opinion surveys and comprehensive evaluations.

[0121] It should be noted that when the temperature anomaly coefficient of the first transmission line is less than the preset temperature anomaly coefficient threshold and the appearance damage coefficient is less than the preset appearance damage coefficient threshold, although individually, the damage or temperature anomaly may not be sufficient to directly determine a fault, the combination of the two may indicate early deterioration or potential problems in some areas of the transmission line; because the combination of the two may reflect potential faults in the transmission line. The appearance damage of the transmission line may lead to a decrease in insulation performance, and a slight temperature anomaly may imply an increase in local resistance or uneven current distribution; for example, slight damage may cause local arcing or partial discharge, and such phenomena are often accompanied by a slight temperature rise; if not dealt with for a long time, it may further develop into a serious fault.

[0122] In one implementation, through the comprehensive evaluation of combining the appearance damage coefficient, the temperature anomaly coefficient, and the fault risk coefficient, it is possible to more comprehensively and accurately determine whether there are potential fault hazards in the first transmission line. The appearance damage coefficient reflects the degree of surface damage of the transmission line, while the temperature anomaly coefficient captures the possible overheating problems that may occur during the operation of the line. By performing weighted summation on these data, the mechanical state and temperature state of the line can be comprehensively considered, so as to make a more accurate risk assessment under different fault manifestations. By comparing with the preset thresholds, false alarms and missed alarms can be effectively avoided, ineffective inspections can be reduced, the inspection efficiency can be improved, potential faults can be detected in advance, and actual problems can be avoided due to deviations in single indicators. In addition, this multi-index evaluation method can provide a more comprehensive fault warning, help the operation and maintenance personnel take measures in a timely manner, prevent faults from occurring, ensure the safe and stable operation of the power system, and thus improve the accuracy and scientific nature of equipment management and maintenance.

[0123] Both the first preset threshold and the second preset threshold are set by technical personnel according to actual needs and are not limited here.

[0124] Based on the same inventive concept, the embodiment of the present invention also provides a transmission line intelligent inspection system. Refer to Figure 2 , Figure 2 is a framework diagram of a transmission line intelligent inspection system provided by the embodiment of the present invention. The system includes:

[0125] Matching module: Obtain the status data of the first transmission line and the historical transmission line set, and perform matching to determine the transmission lines in the historical transmission line set with a similarity greater than the first preset threshold to the status data of the first transmission line as the reference transmission lines; the first transmission line is the transmission line to be inspected, and the historical transmission line set is the transmission lines that have been inspected using drones.

[0126] Inspection module: Use the same type of drone as used during the inspection of the reference transmission line to inspect the first transmission line, and obtain the high-resolution image and thermal imaging image of the first transmission line.

[0127] Damage module: Extract the damaged areas of the transmission line from the high-resolution image, and calculate the external damage coefficient of the first transmission line in combination with the preset transmission line damage image set; the external damage coefficient is used to represent the degree of damage to the external surface of the transmission line; the damage image set contains transmission line damage images corresponding to different types of transmission line faults.

[0128] Judgment module: Obtain the temperature anomaly coefficient of the first transmission line based on the thermal imaging image of the first transmission line, and judge whether there are potential faults in the first transmission line according to the external damage coefficient and the temperature anomaly coefficient; the temperature anomaly coefficient is used to represent the degree of temperature anomaly of the transmission line.

[0129] Based on an embodiment of the present invention, a transmission line intelligent inspection system can, through the above method, for transmission lines that have not been inspected by drones, intelligently match according to the transmission lines that have been inspected by drones, and select a suitable drone to inspect the transmission lines that have not been inspected by drones. This process can utilize the drone types and experience in historical inspection data, ensuring higher pertinence and accuracy in the inspection of transmission lines that have not been inspected by drones, thereby improving the maintenance efficiency and safety of transmission lines; in addition, it can intelligently predict whether there are faults in the transmission line based on the inspection results, resulting in higher inspection efficiency and timely fault handling.

[0130] In one embodiment, the status data has multiple data types, specifically including terrain data, tower base data, environmental climate data, and operation data, where:

[0131] The terrain data includes the maximum height difference, the average value, maximum value, highest elevation point, and lowest elevation point of the slope along the line.

[0132] The tower base data includes the distribution density of the tower bases, the percentage of straight towers, and the percentage of corner towers.

[0133] The environmental climate data includes the annual average temperature, annual average humidity, annual precipitation, annual number of thunderstorm days, and maximum wind speed.

[0134] The operation data includes the rated voltage, load current, and transmission power.

[0135] In one embodiment, the matching module further includes:

[0136] A preprocessing module: preprocess the status data of the first transmission line and the status data of the historical transmission line set;

[0137] A distance matrix construction module: construct a distance matrix for the first transmission line and the target transmission line, where each element in the distance matrix is the difference between the same data types of the first transmission line and the target transmission line; the target transmission line is any transmission line in the historical transmission line set;

[0138] An optimal matching module: use the dynamic programming algorithm to calculate the shortest path in the distance matrix to find the optimal matching between the status data sequence of the first transmission line and the status data sequence of the target transmission line;

[0139] An alignment module: align the status data sequence of the first transmission line and the status data sequence of the target transmission line according to the calculated shortest path, and calculate the similarity between the aligned status data sequence of the first transmission line and the status data sequence of the target transmission line;

[0140] A reference transmission line module: use the transmission line with the largest similarity in the historical transmission line set as the reference transmission line.

[0141] In one embodiment, the damage module includes:

[0142] A first damaged image matrix module: convert each damaged image in the preset transmission line damaged image set into an n×n damaged image matrix, and denote the corresponding damaged image matrix as I1;

[0143] A second damaged image matrix module: for the high-resolution image of the first transmission line, preprocess the high-resolution image, and through image processing and feature extraction algorithms on the preprocessed image, identify and extract the damaged area in the high-resolution image of the first transmission line, and also convert the damaged area into an n×n damaged image matrix, and denote the damaged image matrix of the first transmission line as I2;

[0144] A correlation coefficient module: calculate the correlation coefficient S between the damaged image of the first transmission line and the corresponding damaged image in the preset transmission line damaged image set, and the calculation formula is: where I1(i,j) is the pixel value of the i-th row and j-th column in the damaged image matrix of the first transmission line; I2(i,j) is the pixel value of the i-th row and j-th column in the damaged image matrix of the corresponding damaged image in the transmission line damaged image set;

[0145] Appearance damage coefficient module: Use the maximum correlation coefficient between the damaged images in the preset damaged image set and the first transmission line as the appearance damage coefficient of the first transmission line.

[0146] In one embodiment, the judgment module includes:

[0147] Temperature anomaly coefficient module: Obtain the thermal imaging image of the first transmission line, and extract the temperature value of each pixel point from the thermal imaging image; calculate the temperature difference between the temperature value of each pixel point and the preset reference temperature value, and use the pixel points with a temperature difference greater than the second preset threshold as abnormal pixel points; determine the ratio of the number of abnormal pixel points to the number of all pixel points as the temperature anomaly coefficient of the first transmission line;

[0148] First judgment module: Compare the appearance damage coefficient of the first transmission line with the preset appearance damage coefficient threshold. If the appearance damage coefficient of the first transmission line is not less than the preset appearance damage coefficient threshold, there is a potential fault in the first transmission line;

[0149] Second judgment module: Compare the temperature anomaly coefficient of the first transmission line with the preset temperature anomaly coefficient threshold. If the temperature anomaly coefficient of the first transmission line is not less than the preset temperature anomaly coefficient threshold, there is a potential fault in the first transmission line;

[0150] Third judgment module: When the temperature anomaly coefficient of the first transmission line is less than the preset temperature anomaly coefficient threshold and the appearance damage coefficient is less than the preset appearance damage coefficient threshold, perform a weighted sum of the temperature anomaly coefficient and the appearance damage coefficient of the first transmission line to obtain a fault risk coefficient. Compare the fault risk coefficient with the preset fault risk coefficient threshold. If it is not less than the preset fault risk coefficient threshold, it means there is a potential fault in the first transmission line; otherwise, there is no potential fault.

[0151] The above has described an embodiment of the present invention in detail, but the content is only the preferred embodiment of the present invention and cannot be artificially used to limit the scope of implementation of the present invention. All equivalent changes and improvements made according to the scope of the application of the present invention should still fall within the scope covered by the patent of the present invention.

Claims

1. An intelligent inspection method for transmission lines, characterized in that, It includes the following steps: Obtain the status data of the first transmission line and the historical transmission line set, and perform matching to determine the transmission line in the historical transmission line set with a similarity greater than the first preset threshold to the status data of the first transmission line as the reference transmission line; The first transmission line is the transmission line to be inspected, and the historical transmission line set is the transmission line that has been inspected by an unmanned aerial vehicle; Inspect the first transmission line using the same type of unmanned aerial vehicle as used during the inspection of the reference transmission line to obtain the high-resolution image and thermal imaging image of the first transmission line; Extract the damaged area of the transmission line from the high-resolution image, and calculate the external damage coefficient of the first transmission line in combination with the preset transmission line damaged image set; The external damage coefficient is used to represent the degree of damage to the external surface of the transmission line; the damaged image set contains transmission line damaged images corresponding to different types of transmission line faults; Obtain the temperature anomaly coefficient of the first transmission line based on the thermal imaging image of the first transmission line, and determine whether there are potential faults in the first transmission line based on the external damage coefficient and the temperature anomaly coefficient; The temperature anomaly coefficient is used to represent the degree of temperature anomaly of the transmission line.

2. The intelligent inspection method for a power transmission line according to claim 1, wherein The status data has multiple data types, specifically including terrain data, tower base data, environmental climate data, and operation data, where: The terrain data includes the maximum height difference, the average value, maximum value, highest elevation point, and lowest elevation point of the slope along the line; The tower base data includes the distribution density of the tower bases, the percentage of straight towers, and the percentage of corner towers; The environmental climate data includes the annual average temperature, annual average humidity, annual precipitation, annual number of thunderstorm days, and maximum wind speed; The operation data includes the rated voltage, load current, and transmission power.

3. The intelligent inspection method for transmission lines according to claim 2, characterized in that, The method for determining the reference transmission line set specifically includes: Preprocess the status data of the first transmission line and the historical transmission line set; Construct a distance matrix for the first transmission line and the target transmission line. Each element in the distance matrix is the difference between the same data types of the first transmission line and the target transmission line; the target transmission line is any transmission line in the historical transmission line set; Use the dynamic programming algorithm to calculate the shortest path in the distance matrix to find the best match between the status data sequence of the first transmission line and the status data sequence of the target transmission line; Align the status data sequence of the first transmission line and the status data sequence of the target transmission line according to the calculated shortest path, and calculate the similarity between the aligned status data sequence of the first transmission line and the status data sequence of the target transmission line; Take the transmission line with the maximum similarity in the historical transmission line set as the reference transmission line.

4. The intelligent inspection method for transmission lines according to claim 1, characterized in that Extracting the damaged area of the transmission line from the high-resolution image and calculating the external damage coefficient of the first transmission line in combination with the preset transmission line damaged image set includes: Convert each damaged image in the preset transmission line damaged image set into a damaged image matrix of n×n, and denote the corresponding damaged image matrix as I1; For the high-resolution image of the first transmission line, preprocess the high-resolution image, and through image processing and feature extraction algorithms on the preprocessed image, identify and extract the damaged areas in the high-resolution image of the first transmission line, and also convert the damaged areas into an n×n damaged image matrix. Denote the damaged image matrix of the first transmission line as I2; Calculate the correlation coefficient S between the damaged image of the first transmission line and the corresponding damaged image in the preset damaged image set of the transmission line. The calculation formula is as follows: In the formula, I1(i,j) is the pixel value of the i-th row and j-th column in the damaged image matrix of the first transmission line; I2(i,j) is the pixel value of the i-th row and j-th column in the damaged image matrix corresponding to the damaged image in the damaged image set of the transmission line; Take the maximum correlation coefficient between the damaged images in the preset damaged image set and the first transmission line as the external damage coefficient of the first transmission line.

5. The intelligent inspection method for a transmission line according to claim 1, wherein, Obtain the temperature anomaly coefficient of the first transmission line based on the thermal imaging image of the first transmission line, and judge whether there are potential faults in the first transmission line according to the external damage coefficient and the temperature anomaly coefficient, including: Obtain the thermal imaging image of the first transmission line, and extract the temperature value of each pixel point from the thermal imaging image; calculate the temperature difference between the temperature value of each pixel point and the preset reference temperature value, and take the pixel points with the temperature difference greater than the second preset threshold as abnormal pixel points; determine the ratio of the number of abnormal pixel points to the number of all pixel points as the temperature anomaly coefficient of the first transmission line; Compare the external damage coefficient of the first transmission line with the preset external damage coefficient threshold. If the external damage coefficient of the first transmission line is not less than the preset external damage coefficient threshold, there are potential faults in the first transmission line; Compare the temperature anomaly coefficient of the first transmission line with the preset temperature anomaly coefficient threshold. If the temperature anomaly coefficient of the first transmission line is not less than the preset temperature anomaly coefficient threshold, there are potential faults in the first transmission line; When the temperature anomaly coefficient of the first transmission line is less than the preset temperature anomaly coefficient threshold and the external damage coefficient is less than the preset external damage coefficient threshold, perform a weighted sum of the temperature anomaly coefficient and the external damage coefficient of the first transmission line to obtain a fault risk coefficient. Compare the fault risk coefficient with the preset fault risk coefficient threshold. If it is not less than the preset fault risk coefficient threshold, it means there are potential faults in the first transmission line; otherwise, there are no potential faults.

6. An intelligent inspection system for transmission lines, characterized in that, The system includes: Matching module: Obtain the status data of the first transmission line and the historical transmission line set, and perform matching to determine the transmission line in the historical transmission line set with a similarity greater than the first preset threshold to the status data of the first transmission line as the reference transmission line; the first transmission line is the transmission line to be inspected, and the historical transmission line set is the transmission lines that have been inspected by drones; Inspection module: Use the same type of drone as used when inspecting the reference transmission line to inspect the first transmission line to obtain the high-resolution image and thermal imaging image of the first transmission line; Damage module: Extract the damaged areas of the transmission line from the high-resolution image, and calculate the external damage coefficient of the first transmission line in combination with the preset transmission line damaged image set; the external damage coefficient is used to represent the degree of damage to the external surface of the transmission line; the damaged image set contains transmission line damaged images corresponding to different types of transmission line faults; Judgment module: Obtain the temperature anomaly coefficient of the first transmission line based on the thermal imaging image of the first transmission line, and judge whether there are potential faults in the first transmission line according to the appearance damage coefficient and the temperature anomaly coefficient; the temperature anomaly coefficient is used to represent the degree of temperature anomaly of the transmission line.

7. An intelligent inspection system for transmission lines according to claim 6, characterized in that, The state data has multiple data types, specifically including terrain data, tower base data, environmental climate data, and operation data, where: The terrain data includes the maximum height difference, the average value, maximum value, highest elevation point, and lowest elevation point of the slope along the line. The tower base data includes the distribution density of the tower bases, the percentage of straight towers, and the percentage of corner towers. The environmental climate data includes the annual average temperature, annual average humidity, annual precipitation, annual number of thunderstorm days, and maximum wind speed. The operation data includes the rated voltage, load current, and transmission power.

8. An intelligent inspection system for transmission lines according to claim 7, characterized in that, The matching module further includes: Preprocessing module: Preprocess the state data of the first transmission line and the historical transmission line set. Distance matrix construction module: Construct a distance matrix for the first transmission line and the target transmission line. Each element in the distance matrix is the difference between the same data types of the first transmission line and the target transmission line; the target transmission line is any transmission line in the historical transmission line set. Optimal matching module: Use the dynamic programming algorithm to calculate the shortest path in the distance matrix to find the best match between the state data sequence of the first transmission line and the state data sequence of the target transmission line. Alignment module: Align the state data sequences of the first transmission line and the target transmission line according to the calculated shortest path, and calculate the similarity between the aligned state data sequences of the first transmission line and the target transmission line. Reference transmission line module: Use the transmission line with the maximum similarity in the historical transmission line set as the reference transmission line.

9. An intelligent inspection system for transmission lines according to claim 6, characterized in that, The damage module includes: First damage image matrix module: Convert each damage image in the preset transmission line damage image set into an n×n damage image matrix, and denote the corresponding damage image matrix as I1. Second damage image matrix module: For the high-resolution image of the first transmission line, preprocess the high-resolution image, and use image processing and feature extraction algorithms on the preprocessed image to identify and extract the damaged area in the high-resolution image of the first transmission line, and also convert the damaged area into an n×n damage image matrix. Denote the damage image matrix of the first transmission line as I2. Correlation coefficient module: Calculate the correlation coefficient S between the damaged image of the first transmission line and the corresponding damaged image in the preset set of damaged images of transmission lines. The calculation formula is as follows: In the formula, I1(i,j) is the pixel value of the i-th row and j-th column in the damaged image matrix of the first transmission line; I2(i,j) is the pixel value of the i-th row and j-th column in the damaged image matrix of the corresponding damaged image in the set of damaged images of transmission lines; Appearance damage coefficient module: Use the maximum correlation coefficient between the damage images in the preset damage image set and the first transmission line as the appearance damage coefficient of the first transmission line.

10. The intelligent inspection system for transmission lines according to claim 6, characterized in that, The judgment module includes: Temperature anomaly coefficient module: Obtain the thermal imaging image of the first transmission line, and extract the temperature value of each pixel point from the thermal imaging image; calculate the temperature difference between the temperature value of each pixel point and the preset reference temperature value, and regard the pixel points with a temperature difference greater than the second preset threshold as abnormal pixel points; determine the ratio of the number of abnormal pixel points to the number of all pixel points as the temperature anomaly coefficient of the first transmission line; First judgment module: Compare the surface damage coefficient of the first transmission line with the preset surface damage coefficient threshold. If the surface damage coefficient of the first transmission line is not less than the preset surface damage coefficient threshold, there is a potential fault in the first transmission line; Second judgment module: Compare the temperature anomaly coefficient of the first transmission line with the preset temperature anomaly coefficient threshold. If the temperature anomaly coefficient of the first transmission line is not less than the preset temperature anomaly coefficient threshold, there is a potential fault in the first transmission line; Third judgment module: When the temperature anomaly coefficient of the first transmission line is less than the preset temperature anomaly coefficient threshold and the surface damage coefficient is less than the preset surface damage coefficient threshold, perform a weighted sum of the temperature anomaly coefficient and the surface damage coefficient of the first transmission line to obtain a fault risk coefficient, and compare the fault risk coefficient with the preset fault risk coefficient threshold. If it is not less than the preset fault risk coefficient threshold, it indicates that there is a potential fault in the first transmission line; otherwise, there is no potential fault.

Images