Power grid facility health monitoring method and device, computer equipment and storage medium

Abstract

This application relates to a method, apparatus, computer equipment, and storage medium for monitoring the health of power grid facilities. The method includes: acquiring monitoring images of power grid facilities and environmental data of the power grid facilities corresponding to a target power grid; extracting key monitoring feature points from the monitoring images of the power grid facilities based on the environmental data, obtaining a set of monitoring feature points for the power grid facilities; finding paired monitoring feature points corresponding to each monitoring feature point in the set of monitoring feature points for the power grid facilities, obtaining combinations of paired feature points for each power grid facility; performing geometric transformation analysis on the combinations of paired feature points for each power grid facility, obtaining deformation analysis results of the power grid facilities in the target power grid; and using the deformation analysis results to generate power grid facility health assessment data. This method enables the monitoring effect of power grid facilities to meet the current development needs of the power grid.

Description

Technical Field

[0001] This application relates to the field of smart grid technology, and in particular to an infrastructure health monitoring method, device, computer equipment, storage medium and computer program product based on an adaptive image registration algorithm. Background Technology

[0002] With the development of smart grid technology, power grid facility monitoring technology has emerged. This technology is a process that uses advanced sensors, image processing, and data analysis techniques to monitor and evaluate the operating status and health of power grid equipment in real time. The process aims to promptly detect potential faults, improve the reliability and efficiency of power grid operation, and thus ensure the stability and security of power supply. It enables intelligent and refined management and maintenance of power grid equipment.

[0003] Traditional technologies for monitoring power grid facilities suffer from incomplete data acquisition and lagging data analysis. For example, they typically rely on single-type sensors and scattered data collection points, resulting in fragmented monitoring data that makes it difficult to form a comprehensive health status assessment. Furthermore, data processing and analysis are often delayed, failing to achieve real-time monitoring and early warning, thus preventing the timely detection and handling of potential faults. Therefore, using traditional power grid monitoring methods results in monitoring power grid facilities that fails to meet the current development needs of the power grid. Summary of the Invention

[0004] Therefore, it is necessary to provide a method, device, computer equipment, computer-readable storage medium, and computer program product for monitoring the health of power grid facilities, which can enable the monitoring effect of power grid facilities to meet the current development needs of the power grid.

[0005] In a first aspect, this application provides a method for monitoring the health of power grid facilities. The method includes:

[0006] Acquire monitoring images of power grid facilities and environmental data of the power grid facilities corresponding to the target power grid;

[0007] Based on the power grid facility environmental data, key monitoring feature points are extracted from the power grid facility monitoring images to obtain a set of power grid facility monitoring feature points; the key monitoring feature points are the points in the power grid facility monitoring images that exhibit common physical characteristics under different physical conditions.

[0008] Find the paired monitoring feature points corresponding to each power grid facility monitoring feature point in the power grid facility monitoring feature point set to obtain the paired feature point combination of each power grid facility;

[0009] Geometric transformation analysis is performed on the paired feature points of each of the power grid facilities to obtain the deformation analysis results of the power grid facilities of the target power grid; the deformation analysis results of the power grid facilities are used to generate power grid facility health assessment data, and the power grid facility health assessment data are used to monitor the condition of the power grid facilities of the target power grid.

[0010] Secondly, this application also provides a power grid facility health monitoring device. The device includes:

[0011] The power grid data acquisition module is used to acquire monitoring images of power grid facilities and environmental data of power grid facilities corresponding to the target power grid.

[0012] The feature data extraction module is used to extract key monitoring feature points from the power grid facility monitoring images based on the power grid facility environmental data, and obtain a set of power grid facility monitoring feature points; the key monitoring feature points are the points in the power grid facility monitoring images that exhibit common physical characteristics under different physical conditions;

[0013] The feature data matching module is used to find the paired monitoring feature points corresponding to each power grid facility monitoring feature point in the power grid facility monitoring feature point set, and to obtain the paired feature point combination of each power grid facility.

[0014] The power grid health monitoring module is used to perform geometric transformation analysis on the paired feature points of each power grid facility to obtain the deformation analysis results of the power grid facilities of the target power grid; the deformation analysis results of the power grid facilities are used to generate power grid facility health assessment data, and the power grid facility health assessment data are used to monitor the condition of the power grid facilities of the target power grid.

[0015] Thirdly, this application also provides a computer device. The computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to perform the following steps:

[0016] Acquire monitoring images of power grid facilities and environmental data of the power grid facilities corresponding to the target power grid;

[0017] Based on the power grid facility environmental data, key monitoring feature points are extracted from the power grid facility monitoring images to obtain a set of power grid facility monitoring feature points; the key monitoring feature points are the points in the power grid facility monitoring images that exhibit common physical characteristics under different physical conditions.

[0018] Find the paired monitoring feature points corresponding to each power grid facility monitoring feature point in the power grid facility monitoring feature point set to obtain the paired feature point combination of each power grid facility;

[0019] Geometric transformation analysis is performed on the paired feature points of each of the power grid facilities to obtain the deformation analysis results of the power grid facilities of the target power grid; the deformation analysis results of the power grid facilities are used to generate power grid facility health assessment data, and the power grid facility health assessment data are used to monitor the condition of the power grid facilities of the target power grid.

[0020] Fourthly, this application also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program thereon, which, when executed by a processor, performs the following steps:

[0021] Acquire monitoring images of power grid facilities and environmental data of the power grid facilities corresponding to the target power grid;

[0022] Based on the power grid facility environmental data, key monitoring feature points are extracted from the power grid facility monitoring images to obtain a set of power grid facility monitoring feature points; the key monitoring feature points are the points in the power grid facility monitoring images that exhibit common physical characteristics under different physical conditions.

[0023] Find the paired monitoring feature points corresponding to each power grid facility monitoring feature point in the power grid facility monitoring feature point set to obtain the paired feature point combination of each power grid facility;

[0024] Geometric transformation analysis is performed on the paired feature points of each of the power grid facilities to obtain the deformation analysis results of the power grid facilities of the target power grid; the deformation analysis results of the power grid facilities are used to generate power grid facility health assessment data, and the power grid facility health assessment data are used to monitor the condition of the power grid facilities of the target power grid.

[0025] Fifthly, this application also provides a computer program product. The computer program product includes a computer program that, when executed by a processor, performs the following steps:

[0026] Acquire monitoring images of power grid facilities and environmental data of the power grid facilities corresponding to the target power grid;

[0027] Based on the power grid facility environmental data, key monitoring feature points are extracted from the power grid facility monitoring images to obtain a set of power grid facility monitoring feature points; the key monitoring feature points are the points in the power grid facility monitoring images that exhibit common physical characteristics under different physical conditions.

[0028] Find the paired monitoring feature points corresponding to each power grid facility monitoring feature point in the power grid facility monitoring feature point set to obtain the paired feature point combination of each power grid facility;

[0029] Geometric transformation analysis is performed on the paired feature points of each of the power grid facilities to obtain the deformation analysis results of the power grid facilities of the target power grid; the deformation analysis results of the power grid facilities are used to generate power grid facility health assessment data, and the power grid facility health assessment data are used to monitor the condition of the power grid facilities of the target power grid.

[0030] The aforementioned method, apparatus, computer equipment, storage medium, and computer program product for monitoring the health of power grid facilities involve: acquiring monitoring images of power grid facilities and environmental data of power grid facilities corresponding to a target power grid; extracting key monitoring feature points from the monitoring images of power grid facilities based on the environmental data, thus obtaining a set of monitoring feature points; these key feature points are points in the monitoring images of power grid facilities that exhibit common physical characteristics under different physical conditions; finding paired monitoring feature points corresponding to each monitoring feature point in the set of monitoring feature points of power grid facilities, thus obtaining combinations of paired feature points for each power grid facility; performing geometric transformation analysis on the combinations of paired feature points for each power grid facility, thus obtaining the deformation analysis results of the power grid facilities of the target power grid; and using the deformation analysis results to generate power grid facility health assessment data, which is then used to monitor the condition of the power grid facilities of the target power grid.

[0031] By systematically acquiring monitoring images and environmental data of the target power grid facilities, this method extracts key monitoring feature points that reflect common characteristics of power grid facilities under different physical conditions. These feature points are then combined and geometrically transformed for analysis. This multi-dimensional, multi-angle analysis method accurately captures subtle deformations and structural changes that may occur in power grid facilities during operation. By generating detailed health assessment data, it provides a scientific basis for real-time monitoring, fault prediction, and preventative maintenance of power grid facilities. This improves the comprehensiveness and accuracy of power grid facility monitoring, significantly enhances the safety and reliability of power grid operation, reduces the risk of power outages and maintenance costs caused by sudden faults, and ensures that the monitoring effect of power grid facilities meets the current development needs of the power grid. Attached Figure Description

[0032] Figure 1 This is an application environment diagram of a power grid facility health monitoring method in one embodiment;

[0033] Figure 2 This is a flowchart illustrating a method for monitoring the health of power grid facilities in one embodiment;

[0034] Figure 3 This is a flowchart illustrating a method for obtaining the combination of paired feature points of power grid facilities in one embodiment;

[0035] Figure 4 This is a flowchart illustrating the method for combining power grid facility pairing feature points in another embodiment;

[0036] Figure 5 This is a flowchart illustrating a method for obtaining deformation analysis results of power grid facilities in one embodiment;

[0037] Figure 6 This is a flowchart illustrating the method for obtaining the deformation analysis results of power grid facilities in another embodiment;

[0038] Figure 7 This is a flowchart illustrating a method for obtaining a set of monitoring feature points for power grid facilities in one embodiment.

[0039] Figure 8 This is a structural block diagram of a power grid facility health monitoring device in one embodiment;

[0040] Figure 9 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0041] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0042] This application provides a method for monitoring the health of power grid facilities, which can be applied to, for example... Figure 1In the application environment shown, terminal 102 communicates with server 104 via a network. A data storage system can store the data that server 104 needs to process. The data storage system can be integrated onto server 104 or placed on a cloud or other network server. Server 104 obtains monitoring images of power grid facilities and environmental data of the target power grid from terminal 102; based on the environmental data, it extracts key monitoring feature points from the monitoring images of the power grid facilities to obtain a set of monitoring feature points; the key monitoring feature points are points in the monitoring images of the power grid facilities that exhibit common physical characteristics under different physical conditions; it finds paired monitoring feature points corresponding to each monitoring feature point in the set of monitoring feature points of the power grid facilities to obtain combinations of paired feature points for each power grid facility; it performs geometric transformation analysis on the combinations of paired feature points of each power grid facility to obtain the deformation analysis results of the power grid facilities of the target power grid; the deformation analysis results of the power grid facilities are used to generate power grid facility health assessment data, which is used to monitor the condition of the power grid facilities of the target power grid. The terminal 102 can be, but is not limited to, various personal computers, laptops, smartphones, tablets, IoT devices, and portable wearable devices. IoT devices can include smart speakers, smart TVs, smart air conditioners, and smart in-vehicle systems. Portable wearable devices can include smartwatches, smart bracelets, and head-mounted devices. The server 104 can be implemented using a standalone server or a server cluster consisting of multiple servers.

[0043] In one embodiment, such as Figure 2 As shown, a method for monitoring the health of power grid facilities is provided, which can be applied to... Figure 1 Taking the server in the example, the following steps are included:

[0044] Step 202: Obtain monitoring images of power grid facilities and environmental data of power grid facilities corresponding to the target power grid.

[0045] Among these, power grid facility monitoring images can be images obtained by capturing images of power grid facilities using high-resolution camera equipment. These images are used to record and analyze the appearance, structure, and operating status of the power grid facilities. These images can help identify potential faults such as physical damage, corrosion, and deformation of the equipment.

[0046] The environmental data for power grid facilities can be information related to the surrounding environment collected through sensor networks, including temperature, humidity, wind speed, vibration, and air pressure. This data is used to monitor and assess the environmental conditions of the power grid facilities and to analyze their operational status and health condition.

[0047] Specifically, high-resolution camera equipment and sensor networks are deployed in key facility areas of the target power grid (designated or random locations) to periodically or continuously collect monitoring images and environmental data of the power grid facilities. These images and data are then transmitted wirelessly or via wired connections to a central data processing system. Within this system, image processing algorithms are used to preprocess the monitoring images, removing noise and interference, while simultaneously processing the environmental sensor data to ensure temporal and spatial correspondence between the monitoring images and the environmental data.

[0048] Step 204: Based on the power grid facility environmental data, extract the key monitoring feature points from the power grid facility monitoring images to obtain a set of power grid facility monitoring feature points.

[0049] Among them, the key monitoring feature points can be feature points that do not fluctuate much under different environmental and physical conditions and can represent the monitoring images of power grid facilities.

[0050] Among them, the set of monitoring feature points for power grid facilities can be a set of key feature points that remain stable under different physical conditions. These feature points reflect the common physical characteristics of power grid facilities under various environmental conditions.

[0051] Specifically, since the power grid facility environmental data and the corresponding power grid facility monitoring images are synchronized in time and space—that is, each power grid facility monitoring image corresponds one-to-one with its environmental data—image processing algorithms are used to analyze the power grid facility monitoring images under the constraints of the power grid facility environmental data (i.e., the target power grid under different environmental conditions). This allows for the extraction of key monitoring feature points, such as edges and corners, that remain stable under different physical conditions (or / and different environmental conditions). These feature points reflect the common physical characteristics of power grid facilities under various environments. After screening and verification, a set of power grid facility monitoring feature points is formed.

[0052] Step 206: Find the paired monitoring feature points corresponding to each power grid facility monitoring feature point in the power grid facility monitoring feature point set, and obtain the paired feature point combination of each power grid facility.

[0053] Among them, the paired monitoring feature points can be the feature points that correspond to each other in different power grid facility monitoring images, which are found by feature matching algorithms in the set of power grid facility monitoring feature points.

[0054] Among them, the pairing of feature points of power grid facilities can refer to a set of corresponding feature point pairs found by feature matching algorithms in different monitoring images. These sets of feature point pairs have similar physical characteristics under various environmental conditions.

[0055] Specifically, feature descriptors are calculated for each power grid facility monitoring feature point in the set of monitoring feature points to generate its unique feature vector. Then, feature matching algorithms (such as nearest neighbor matching or FLANN) are used to find the best-matching paired monitoring feature points in different monitoring images to form paired feature points. Through these paired feature points, the paired feature point combination of power grid facilities can be obtained.

[0056] Step 208: Perform geometric transformation analysis on the paired feature points of each power grid facility to obtain the deformation analysis results of the power grid facilities of the target power grid.

[0057] Among them, geometric transformation analysis can be used to calculate the combination of paired feature points of power grid facilities by applying mathematical transformations (such as affine transformation, perspective transformation, etc.) to reveal the changes in displacement, rotation, scaling and other properties of power grid facilities.

[0058] Among them, the deformation analysis results of power grid facilities can be assessment data on the deformation of power grid facilities at different time points and / or under different physical conditions obtained through geometric transformation analysis.

[0059] Specifically, geometric transformation analysis methods, such as affine or perspective transformation, are applied to the paired feature point combinations of each group of power grid facilities to calculate the deformation parameters of the target power grid facilities at different time points and / or under different physical conditions. These deformation parameters are then used to generate deformation analysis results for the target power grid facilities. After a comprehensive evaluation combining environmental data and these analysis results, the results are converted into health assessment data, reflecting the operating status and potential faults of the power grid facilities. Ultimately, this health assessment data is used to monitor the condition of the power grid facilities in real time, providing timely warnings and guiding maintenance work, thereby improving the safety and reliability of the power grid.

[0060] In the aforementioned method for monitoring the health of power grid facilities, the following steps are taken: acquiring monitoring images of power grid facilities and environmental data of the target power grid; extracting key monitoring feature points from the monitoring images based on the environmental data to obtain a set of monitoring feature points; these key feature points are points in the monitoring images that exhibit common physical characteristics under different physical conditions; finding paired monitoring feature points corresponding to each monitoring feature point in the set to obtain combinations of paired feature points; performing geometric transformation analysis on the combinations of paired feature points to obtain deformation analysis results of the target power grid facilities; and using these deformation analysis results to generate power grid facility health assessment data, which is then used to monitor the condition of the power grid facilities in the target power grid.

[0061] By systematically acquiring monitoring images and environmental data of the target power grid facilities, this method extracts key monitoring feature points that reflect common characteristics of power grid facilities under different physical conditions. These feature points are then combined and geometrically transformed for analysis. This multi-dimensional, multi-angle analysis method accurately captures subtle deformations and structural changes that may occur in power grid facilities during operation. By generating detailed health assessment data, it provides a scientific basis for real-time monitoring, fault prediction, and preventative maintenance of power grid facilities. This improves the comprehensiveness and accuracy of power grid facility monitoring, significantly enhances the safety and reliability of power grid operation, reduces the risk of power outages and maintenance costs caused by sudden faults, and ensures that the monitoring effect of power grid facilities meets the current development needs of the power grid.

[0062] In one embodiment, such as Figure 3 As shown, the process involves finding paired monitoring feature points corresponding to each power grid facility monitoring feature point in the power grid facility monitoring feature point set, resulting in combinations of paired feature points for each power grid facility, including:

[0063] Step 302: Using a high-dimensional spatial proximity search algorithm, find the nearest paired monitoring feature points to each power grid facility monitoring feature point, and obtain the initial facility paired feature point combination.

[0064] Among these, high-dimensional proximity search algorithms can be used to find the most similar points in a high-dimensional feature space. By calculating the distance between feature points in high-dimensional space (such as KD-trees or Ball trees), the nearest pairing point with the target feature point can be found efficiently. This method is widely used in image processing and pattern recognition, enabling fast and accurate matching of feature points in large amounts of data, improving the accuracy and efficiency of data analysis.

[0065] The initial facility pairing feature point combination can be a feature point pair formed after each feature point finds its nearest paired feature point.

[0066] Specifically, high-dimensional feature descriptions are performed on each power grid facility monitoring feature point to generate feature vectors. Then, a high-dimensional space proximity search algorithm (such as KD tree or Ball tree) is used to find feature points that are paired with each power grid facility monitoring feature point in the feature vector space. Furthermore, by calculating the distance between each power grid facility monitoring feature point and the feature vectors of each paired feature point, the feature point with the smallest distance is selected as the paired monitoring feature point for each power grid facility monitoring feature point, forming each initial facility paired feature point combination, ensuring that the most similar feature point pair is found in the high-dimensional feature space.

[0067] Step 304: Perform multiple ratio tests on each initial facility pairing feature point combination to obtain each test facility pairing feature point combination.

[0068] Multiple ratio testing is a technique used to verify and screen feature point pairs. It calculates the similarity ratio of each initial pair of feature points and compares it with other potential matches to ensure the uniqueness and reliability of the selected feature point pairs. This test eliminates pairs that do not meet the similarity threshold and uniqueness criteria, resulting in more accurate and stable feature point pairings and improving the precision of subsequent analysis and evaluation.

[0069] Among them, the test facility paired feature point combination can refer to the feature point pairing results after multiple ratio tests and screening. These feature point pairs have been verified by strict similarity and uniqueness to ensure the accuracy and reliability of the pairing.

[0070] Specifically, for each initial facility pair feature point combination, multiple ratio tests are performed. By calculating the similarity ratio of each pair of initial facility pair feature points, the feature point pairs with the highest similarity are selected. Simultaneously, each initial facility pair feature point combination is compared with the ratios of other potential matching feature points to ensure the uniqueness and reliability of the selected feature point pairs. After multiple ratio tests, initial facility pair feature point combinations that do not meet the similarity threshold and uniqueness criteria are eliminated, resulting in more accurate and stable test facility pair feature point combinations.

[0071] Step 306: Use graph theory optimization algorithm to optimize the pairing feature point combination of each test facility to obtain the pairing feature point combination of each power grid facility.

[0072] Among them, graph theory optimization algorithms are algorithms based on graph theory, used to optimize the relationships between nodes and edges in a graph model. By calculating the weights of nodes and edges, they seek the globally optimal solution. These algorithms (such as minimum spanning tree, maximum matching algorithm, or graph cut algorithm) can effectively eliminate unreliable or redundant connections, ensuring the accuracy and consistency of the selected feature point pairings.

[0073] Specifically, a graph model is established based on the combination of paired feature points of various test facilities, where paired feature points are nodes and paired relationships are edges. Then, graph theory optimization algorithms (such as minimum spanning tree or maximum matching algorithms) are used to optimize the graph model, eliminating unreliable or redundant paired feature points to ensure global consistency and optimality of paired feature points. The final combination of paired feature points for each power grid facility is obtained through the optimized graph model.

[0074] In this embodiment, a comprehensive application of a high-dimensional spatial proximity search algorithm, initial facility pairing feature point combinations, multiple ratio tests, and graph theory optimization algorithms accurately and efficiently matches monitoring feature points of power grid facilities. This series of steps ensures the accuracy and reliability of feature point matching, thereby significantly improving the accuracy and stability of power grid facility deformation analysis. The final combination of paired feature points can more accurately reflect the actual deformation of power grid facilities under different environmental conditions, providing a solid data foundation and scientific basis for the health monitoring and fault prediction of power grid facilities, and improving the safety and efficiency of power grid operation and maintenance.

[0075] In one embodiment, such as Figure 4 As shown, multiple ratio tests were conducted on each initial facility pairing feature point combination to obtain each test facility pairing feature point combination, including:

[0076] Step 402: Use a high-dimensional space second nearest neighbor distance search algorithm to find paired monitoring feature points that are second nearest to each power grid facility monitoring feature point, and obtain each second nearest neighbor distance.

[0077] Among them, the high-dimensional space second nearest neighbor distance search algorithm can be a method to find the second nearest feature point to the target feature point in a high-dimensional feature space. Using such algorithms (e.g., KD-trees or Ball trees), the second nearest paired feature points can be found quickly and efficiently in the feature vector space, and their distances can be calculated.

[0078] The second nearest neighbor distance can be the distance between the second nearest feature point to the target feature point in a high-dimensional feature space.

[0079] Specifically, a high-dimensional feature vector is generated for each power grid facility monitoring feature point. Then, a high-dimensional space second nearest neighbor distance search algorithm (such as KD-tree or Ball tree) is used to find the second nearest neighbor feature point in the feature vector space. By calculating the distance between the power grid facility monitoring feature point and the corresponding second nearest neighbor feature point, the second nearest neighbor distance for each power grid facility monitoring feature point is obtained.

[0080] Step 404: Calculate the ratio of the nearest neighbor distance to the corresponding second nearest neighbor distance for each initial facility pairing feature point combination.

[0081] The nearest neighbor distance can be the distance between a feature point and its nearest neighbor in a high-dimensional feature space. This distance reflects the similarity between the two feature points in the feature space.

[0082] The nearest neighbor ratio can be the ratio of the nearest neighbor distance to the second nearest neighbor distance of a feature point. This ratio is used to measure the relative similarity of feature point pairings.

[0083] Specifically, the nearest neighbor distance for each pair of initial facility pairing feature points is calculated, which is the distance between the two nearest neighbor feature points in the initial facility pairing feature point combination. Then, the nearest neighbor distance for each pair of initial facility pairing feature point combinations is divided by its next nearest neighbor distance to obtain the ratio of each nearest neighbor distance.

[0084] Step 406: Determine the pairing of feature points for each test facility based on the nearest neighbor distance ratio.

[0085] Specifically, each proximity distance ratio is compared with a preset threshold, and initial facility pairing feature point combinations with proximity distance ratios lower than the threshold are selected, indicating that these initial facility pairing feature point combinations have higher similarity and reliability. Test facility pairing feature point combinations are determined based on the selection results.

[0086] In this embodiment, a high-dimensional spatial second nearest neighbor distance search algorithm is used to find the second nearest neighbor paired feature points for each power grid facility monitoring feature point. The ratio of the nearest neighbor distance to the second nearest neighbor distance is calculated, and the most reliable combination of test facility paired feature points is selected based on these ratios. This not only improves the accuracy of feature point matching but also effectively eliminates mismatches and unstable paired points. Through this series of steps, the final combination of test facility paired feature points is more stable and accurate, providing more reliable data support for subsequent deformation analysis and power grid facility health assessment, thereby improving the accuracy and reliability of power grid facility monitoring.

[0087] In one embodiment, such as Figure 5 As shown, geometric transformation analysis is performed on the paired feature points of each power grid facility to obtain the deformation analysis results of the target power grid facilities, including:

[0088] Step 502: Construct a geometric transformation model based on the combination of paired feature points of each power grid facility.

[0089] Among them, the geometric transformation model can be used to describe and calculate the spatial relationship between two coordinate systems, mapping one set of points to another set of points through specific transformation rules (such as translation, rotation, scaling and perspective transformation).

[0090] Specifically, by using the selected combinations of paired feature points of various power grid facilities, an appropriate geometric transformation model (such as affine transformation, perspective transformation, etc.) is selected; then, based on the coordinates of the paired feature points in each combination of paired feature points of power grid facilities, the geometric transformation parameters are calculated, and the parameters are solved and the model is fitted using optimization algorithms such as the least squares method to ensure that the transformation model can accurately describe the spatial relationship between feature points, thus constructing a geometric transformation model.

[0091] Step 504: Calculate the positional deformation of the combination of paired feature points for each power grid facility based on the geometric transformation model.

[0092] Among them, the positional deformation can be the positional difference of feature points between different coordinate systems calculated by the geometric transformation model.

[0093] Specifically, the coordinates of the feature points of each paired feature point combination of various power grid facilities are substituted into the geometric transformation model to calculate the coordinates of the transformed feature points. Then, the coordinates of the feature points before and after the transformation are compared to calculate the positional difference between the two, i.e., the positional deformation, which can be used to assess the degree of deformation of power grid facilities under different conditions.

[0094] Step 506: Based on the deformation at each location, obtain the deformation analysis results of the power grid facilities.

[0095] Specifically, the calculated location deformation variables of all paired feature point combinations of power grid facilities are summarized, and these location deformation variables are statistically analyzed to identify overall deformation patterns and anomalous changes. Then, deformation analysis algorithms (such as deviation analysis or strain calculation) are used to conduct in-depth analysis of the location deformation variables, quantifying information such as the degree and distribution of deformation of power grid facilities as the results of power grid facility deformation analysis.

[0096] In this embodiment, a geometric transformation model is constructed to calculate the positional deformation of paired feature points of power grid facilities, and deformation analysis results of the power grid facilities are derived based on these deformations. This process can accurately capture the minute displacements and deformations of power grid facilities under different environmental conditions, thereby providing detailed deformation data. Through this precise analysis method, potential problems and abnormal changes in power grid facilities can be detected in a timely manner, providing a scientific basis for the health monitoring and maintenance of power grid facilities, improving the safety and stability of power grid operation, and reducing losses and maintenance costs caused by faults.

[0097] In one embodiment, such as Figure 6 As shown, based on the geometric transformation model, the positional deformation of the paired feature point combinations of each power grid facility is calculated, including:

[0098] Step 602: Using a geometric transformation model, the source feature points in the combination of paired feature points of each power grid facility are mapped to the mapped feature points to obtain the positions of each mapped feature point.

[0099] The source feature points can be selected feature points in the original image or coordinate system in the geometric transformation model. These points are the starting point for geometric transformation and mapping.

[0100] Among them, the mapped feature points can be feature points obtained in the target image or coordinate system after the coordinates of the source feature points are transformed through a geometric transformation model.

[0101] The mapped feature point position can be the new position obtained in the target image or coordinate system after the coordinates of the source feature point are transformed through a geometric transformation model. This position reflects the specific coordinates of the source feature point in the new coordinate system.

[0102] Specifically, a suitable geometric transformation model (such as affine transformation or perspective transformation) is selected and applied to the combination of paired feature points for each power grid facility. The coordinates of the source feature points of each paired feature point combination are substituted into the formula of the geometric transformation model so that the coordinates of each source feature point are mapped to the corresponding mapped points, and the positions of the corresponding mapped feature points are calculated. Through this mapping process, the source feature points of each paired feature point combination for each power grid facility can be accurately transformed into a new coordinate system, obtaining the positions of all mapped feature points.

[0103] Step 604: Calculate the distance between the position of each mapped feature point and the corresponding target feature point to obtain each initial deformation.

[0104] The target feature point location can be the actual location of the feature point in the power grid facility or image in the target coordinate system or target image (paired monitoring feature points in the combination of paired feature points of power grid facility) during geometric transformation analysis.

[0105] The initial deformation can be a value obtained by calculating the distance between the location of the mapped feature point and the location of the corresponding target feature point. It reflects the displacement of the feature point after the geometric transformation and is used to preliminarily assess the degree of deformation of the object or structure.

[0106] Specifically, the mapped feature point positions of the paired feature point combinations of power grid facilities are compared with the corresponding target feature point positions (the positions of the paired monitoring feature points in the paired feature point combinations of power grid facilities). The Euclidean distance between each pair of mapped feature point positions and the target feature point is calculated, which is the initial deformation, reflecting the displacement of the source feature point after geometric transformation.

[0107] Step 606: Calculate the positional deformation based on each initial deformation.

[0108] Specifically, the average value of each initial deformation is calculated to obtain the positional average deformation, which reflects the average degree of overall displacement and deformation; the maximum value is found among all initial deformations to obtain the positional maximum deformation, which represents the most significant displacement and deformation. The positional average deformation and the positional maximum deformation are used as the positional deformation.

[0109] In this embodiment, a geometric transformation model is used to map source feature points to mapped feature points, and the distance between the mapped feature point location and the target feature point location is calculated to obtain the initial deformation. Then, by analyzing the initial deformation, the average deformation and maximum deformation at each location are calculated. This series of steps enables precise quantification of the overall deformation and local maximum deformation of the power grid facilities, providing comprehensive and detailed deformation analysis results. This method allows for more accurate monitoring of the health status of power grid facilities, timely detection of potential problems and abnormal changes, enhanced reliability and security of power grid operation, and optimized maintenance and fault prevention strategies.

[0110] In one embodiment, such as Figure 7 As shown, based on the power grid facility environmental data, key monitoring feature points are extracted from the power grid facility monitoring images to obtain a set of power grid facility monitoring feature points, including:

[0111] Step 702: Detect the key feature points of the power grid facility monitoring image to obtain an initial set of facility monitoring feature points.

[0112] The initial set of facility monitoring feature points can be a set of key monitoring feature points detected and extracted from images obtained by processing and analyzing power grid facility monitoring images.

[0113] Specifically, the power grid facility monitoring images are preprocessed to remove noise and improve image quality. Then, image processing algorithms (such as edge detection, corner detection, or feature descriptor algorithms) are applied to detect and extract key monitoring feature points in the images. These image processing algorithms identify points with significant features in the images (points in the power grid facility monitoring images that exhibit common physical characteristics under different physical conditions), such as corners, edge points, or other significant regions, forming an initial set of facility monitoring feature points.

[0114] Step 704: Based on the local information entropy of the power grid facility monitoring image, dynamically adjust the weights of each key monitoring feature point in the initial facility monitoring feature point set to obtain the adjusted facility monitoring feature point set.

[0115] Local information entropy, in image processing, is a measure of the complexity or uncertainty of information within a local region of an image. A high information entropy value indicates that the region contains more detail and complex information, while a low information entropy value indicates that the region is relatively simple and contains less information.

[0116] Among them, the set of facility monitoring feature points can be the set of feature points obtained by dynamically adjusting the weights of the initial facility monitoring feature points based on the local information entropy.

[0117] Specifically, the local information entropy of the power grid facility monitoring image is calculated to assess the information complexity of each region. Then, based on these local information entropy values, the weights of each key monitoring feature point in the initial facility monitoring feature point set are adjusted, assigning higher weights to regions with high local information entropy and lower weights to regions with low local information entropy. This dynamic adjustment highlights information-rich feature points in the image, ultimately resulting in an adjusted facility monitoring feature point set.

[0118] Step 706: Based on the power grid facility environment data, optimize each key monitoring feature point in the set of monitoring feature points of the adjustment facility to obtain the set of monitoring feature points of the power grid facility.

[0119] Specifically, by combining environmental data of power grid facilities (such as temperature, humidity, and wind speed), each key monitoring feature point in the set of monitoring feature points for the facilities is analyzed. For example, the stability and reliability of these key monitoring feature points under different environmental conditions are evaluated using the environmental data of power grid facilities. Then, feature points that are significantly affected by the environment and are unstable are eliminated, while feature points with high environmental stability and reliability are strengthened. Through this optimization process, the final set of monitoring feature points for power grid facilities is obtained.

[0120] In this embodiment, a preliminary set of feature points for power grid facility monitoring is obtained by detecting key feature points in the monitoring images. Then, the weights of these feature points are dynamically adjusted based on local information entropy, highlighting feature points with rich information content, forming an adjusted feature point set. Finally, these feature points are further optimized by combining environmental data of the power grid facilities, eliminating unstable feature points that are significantly affected by the environment, ultimately obtaining an optimized set of power grid facility monitoring feature points. This process ensures the representativeness and stability of the feature points, improves the accuracy and reliability of monitoring, and thus provides high-quality data support for deformation analysis and health assessment of power grid facilities, enhancing the safety of power grid operation and the scientific nature of maintenance.

[0121] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0122] Based on the same inventive concept, this application also provides a power grid facility health monitoring device for implementing the power grid facility health monitoring method described above. The solution provided by this device is similar to the solution described in the above method; therefore, the specific limitations of the one or more power grid facility health monitoring device embodiments provided below can be found in the limitations of the power grid facility health monitoring method described above, and will not be repeated here.

[0123] In one embodiment, such as Figure 8 As shown, a power grid facility health monitoring device is provided, comprising: a power grid data acquisition module 802, a feature data extraction module 804, a feature data matching module 806, and a power grid health monitoring module 808, wherein:

[0124] The power grid data acquisition module 802 is used to acquire monitoring images of power grid facilities and environmental data of power grid facilities corresponding to the target power grid.

[0125] The feature data extraction module 804 is used to extract key monitoring feature points from the power grid facility monitoring images based on the power grid facility environmental data, and obtain a set of power grid facility monitoring feature points; the key monitoring feature points are the points in the power grid facility monitoring images that reflect common physical characteristics under different physical conditions;

[0126] The feature data matching module 806 is used to find the paired monitoring feature points corresponding to each power grid facility monitoring feature point in the power grid facility monitoring feature point set, and to obtain the paired feature point combination of each power grid facility.

[0127] The power grid health monitoring module 808 is used to perform geometric transformation analysis on the paired feature points of each power grid facility to obtain the deformation analysis results of the power grid facilities of the target power grid; the deformation analysis results of the power grid facilities are used to generate power grid facility health assessment data, and the power grid facility health assessment data are used to monitor the condition of the power grid facilities of the target power grid.

[0128] In one embodiment, the feature data matching module 806 is further configured to use a high-dimensional spatial proximity search algorithm to find the nearest paired monitoring feature point to each power grid facility monitoring feature point, thereby obtaining each initial facility paired feature point combination; perform multiple ratio tests on each initial facility paired feature point combination to obtain each test facility paired feature point combination; and use a graph theory optimization algorithm to optimize each test facility paired feature point combination to obtain each power grid facility paired feature point combination.

[0129] In one embodiment, the feature data matching module 806 is further configured to use a high-dimensional space second nearest neighbor distance search algorithm to find the second nearest paired monitoring feature points of each power grid facility monitoring feature point, and obtain each second nearest neighbor distance; calculate the nearest neighbor distance ratio of each initial facility paired feature point combination to the corresponding second nearest neighbor distance; and determine each test facility paired feature point combination based on each nearest neighbor distance ratio.

[0130] In one embodiment, the power grid health monitoring module 808 is further configured to construct a geometric transformation model based on the combination of paired feature points of each power grid facility; calculate the positional deformation of the combination of paired feature points of each power grid facility based on the geometric transformation model; and obtain the deformation analysis results of the power grid facility based on the positional deformation.

[0131] In one embodiment, the power grid health monitoring module 808 is further configured to use a geometric transformation model to map source feature points in the combination of paired feature points of each power grid facility to mapped feature points, thereby obtaining the position of each mapped feature point; calculate the distance between the position of each mapped feature point and the corresponding target feature point position, thereby obtaining each initial deformation variable; and calculate the position deformation variable based on each initial deformation variable; the position deformation variable includes the average position deformation variable and the maximum position deformation variable.

[0132] In one embodiment, the feature data extraction module 804 is further configured to detect key monitoring feature points in the power grid facility monitoring image to obtain an initial set of facility monitoring feature points; dynamically adjust the weights of each key monitoring feature point in the initial set of facility monitoring feature points based on the local information entropy of the power grid facility monitoring image to obtain an adjusted set of facility monitoring feature points; and optimize each key monitoring feature point in the adjusted set of facility monitoring feature points based on the power grid facility environmental data to obtain a power grid facility monitoring feature point set.

[0133] The various modules in the aforementioned power grid facility health monitoring device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0134] In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 9As shown, the computer device includes a processor, memory, and a network interface connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The database stores server data. The network interface communicates with external terminals via a network connection. When the computer program is executed by the processor, it implements a method for monitoring the health of power grid facilities.

[0135] Those skilled in the art will understand that Figure 9 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0136] In one embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above method embodiments.

[0137] In one embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0138] In one embodiment, a computer program product or computer program is provided, the computer program product or computer program including computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and executes the computer instructions, causing the computer device to perform the steps in the above method embodiments.

[0139] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0140] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0141] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0142] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for monitoring the health of power grid facilities, characterized in that, The method includes: Acquire monitoring images of power grid facilities and environmental data of the power grid facilities corresponding to the target power grid; The monitoring key feature points of the power grid facility monitoring image are detected to obtain an initial set of facility monitoring feature points; based on the local information entropy of the power grid facility monitoring image, the weights of each monitoring key feature point in the initial set of facility monitoring feature points are dynamically adjusted to obtain an adjusted set of facility monitoring feature points; based on the power grid facility environmental data, each monitoring key feature point in the adjusted set of facility monitoring feature points is optimized to obtain a power grid facility monitoring feature point set; the monitoring key feature points are points in the power grid facility monitoring image that exhibit common physical characteristics under different physical conditions. Find the paired monitoring feature points corresponding to each power grid facility monitoring feature point in the power grid facility monitoring feature point set to obtain the paired feature point combination of each power grid facility; Geometric transformation analysis is performed on the paired feature points of each of the power grid facilities to obtain the deformation analysis results of the power grid facilities of the target power grid; the deformation analysis results of the power grid facilities are used to generate power grid facility health assessment data, and the power grid facility health assessment data are used to monitor the condition of the power grid facilities of the target power grid.

2. The method of claim 1, wherein, The process of finding paired monitoring feature points corresponding to each power grid facility monitoring feature point in the set of power grid facility monitoring feature points, to obtain combinations of paired feature points for each power grid facility, includes: A high-dimensional spatial proximity search algorithm is used to find the nearest paired monitoring feature point to each of the power grid facility monitoring feature points, thus obtaining the initial facility paired feature point combination; Each of the initial facility pairing feature point combinations is subjected to multiple ratio tests to obtain each test facility pairing feature point combination. A graph theory optimization algorithm is used to optimize the pairing feature point combinations of each test facility to obtain the pairing feature point combinations of each power grid facility.

3. The method of claim 2, wherein, The step of performing multiple ratio tests on each of the initial facility pairing feature point combinations to obtain each test facility pairing feature point combination includes: A high-dimensional spatial second nearest neighbor distance search algorithm is used to find paired monitoring feature points that are second nearest to each of the power grid facility monitoring feature points, and to obtain each second nearest neighbor distance; Calculate the ratio of the nearest neighbor distance to the corresponding second nearest neighbor distance for each of the initial facility pairing feature point combinations; Based on the respective proximity distance ratios, the pairing of feature points for each test facility is determined.

4. The method of claim 1, wherein, The step of performing geometric transformation analysis on the paired feature points of each of the power grid facilities to obtain the deformation analysis results of the target power grid facilities includes: A geometric transformation model is constructed based on the combination of paired feature points of each of the aforementioned power grid facilities; Based on the geometric transformation model, calculate the positional deformation of each pair of power grid facility feature point combinations; Based on the deformation variables at each location, the deformation analysis results of the power grid facilities are obtained.

5. The method of claim 4, wherein, The step of calculating the positional deformation of each pair of power grid facility feature point combinations based on the geometric transformation model includes: Using a geometric transformation model, the source feature points in the paired feature point combinations of each power grid facility are mapped to the mapped feature points to obtain the positions of each mapped feature point. Calculate the distance between each mapped feature point position and the corresponding target feature point position to obtain each initial deformation; The positional deformation is calculated based on the initial deformations; the positional deformation includes the average positional deformation and the maximum positional deformation.

6. A power grid facility health monitoring apparatus characterized by, The device includes: The power grid data acquisition module is used to acquire monitoring images of power grid facilities and environmental data of power grid facilities corresponding to the target power grid. The feature data extraction module is used to detect key monitoring feature points in the power grid facility monitoring image to obtain an initial set of facility monitoring feature points; based on the local information entropy of the power grid facility monitoring image, the weights of each key monitoring feature point in the initial set of facility monitoring feature points are dynamically adjusted to obtain an adjusted set of facility monitoring feature points; based on the power grid facility environmental data, each key monitoring feature point in the adjusted set of facility monitoring feature points is optimized to obtain a power grid facility monitoring feature point set; the key monitoring feature points are points in the power grid facility monitoring image that exhibit common physical characteristics under different physical conditions. The feature data matching module is used to find the paired monitoring feature points corresponding to each power grid facility monitoring feature point in the power grid facility monitoring feature point set, and to obtain the paired feature point combination of each power grid facility. The power grid health monitoring module is used to perform geometric transformation analysis on the paired feature points of each power grid facility to obtain the deformation analysis results of the power grid facilities of the target power grid; the deformation analysis results of the power grid facilities are used to generate power grid facility health assessment data, and the power grid facility health assessment data are used to monitor the condition of the power grid facilities of the target power grid.

7. The apparatus according to claim 6, characterized in that, The feature data matching module is further configured to use a high-dimensional spatial proximity search algorithm to find the nearest paired monitoring feature point to each of the power grid facility monitoring feature points, thereby obtaining each initial facility paired feature point combination; to perform multiple ratio tests on each of the initial facility paired feature point combinations, thereby obtaining each test facility paired feature point combination; and to use a graph theory optimization algorithm to optimize each test facility paired feature point combination, thereby obtaining each power grid facility paired feature point combination.

8. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 5.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 5.

10. A computer program product comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 5.

Images