A method and system for risk management of a power transmission corridor

Abstract

The application discloses a kind of risk processing method and system of power transmission channel, applied to power transmission channel field, method includes: obtaining the initial image data set of target power transmission channel;Extract the line segment information in initial image data set, obtain initial power transmission conductor information;Initial power transmission conductor information is analyzed to structure, with analysis result screening second power transmission conductor information is obtained;Parabolic fitting is carried out to the conductor profile formed by second power transmission conductor information, and based on the inversion result that fitting obtains the power transmission conductor label information of target power transmission channel;Based on power transmission conductor label information, initial image data set is marked and handled, and the marked image data set is obtained;Marked image data set is identified and handled, to generate the risk early warning signal of target power transmission channel with processing result.The risk processing method and system of power transmission channel provided by the application improve the accuracy and efficiency of hidden danger detection, significantly enhance the security and stability of power system.

Description

Technical Field

[0001] This invention relates to the field of power transmission channel technology, and in particular to a method and system for handling risks in power transmission channels. Background Technology

[0002] Power transmission channels overcome geographical barriers, efficiently delivering abundant clean energy from the west to load centers in the east. They are key infrastructure for ensuring energy security, optimizing resource allocation, and promoting green development.

[0003] However, safety hazards caused by conductor sag or tree branch encroachment in transmission channels are a prominent challenge in power system operation and maintenance. Excessive conductor sag can cause contact with the ground or objects it crosses, leading to corona discharge and flashover tripping under humid or windy conditions, potentially resulting in line breaks or wildfires. Currently, addressing these hazards primarily relies on regular manual inspections, with workers visually observing through binoculars or climbing towers for checks. Once an anomaly is detected, workers then organize power outages or live-line work to tighten conductors or clear tree obstructions. This method has significant drawbacks. In low-light conditions such as at night, during rain, or in forested areas, the contrast between conductors and the background is extremely low, and details of tree branches are blurred. It is difficult for the naked eye and conventional optical equipment to accurately identify conductor morphology and vegetation encroachment, leaving many hazards in blind spots. This passive, delayed, and high-risk manual handling mode is not only inefficient but also fails to meet the urgent needs of modern power grids for proactive early warning and intelligent control. Summary of the Invention

[0004] This invention provides a risk management method and system for power transmission channels to solve the technical problem of low efficiency caused by relying on manual inspection in the prior art, thereby improving the efficiency of risk identification.

[0005] To address the aforementioned technical problems, this invention provides a risk management method and system for power transmission channels, the method comprising: Obtain the initial image dataset of the target power transmission channel; Extract line segment information from the initial image dataset to obtain initial power transmission conductor information; Structural analysis is performed on the initial transmission line information, and the analysis results are used to filter and obtain the second transmission line information. The structural analysis process is configured to analyze the similar principal directions and anisotropy of the line segment information at different spatial scales. Based on the catenary mechanics principle, the conductor profile formed by the second transmission conductor information is parabolic fitted, and the transmission conductor label information of the target transmission channel is obtained by inversion based on the fitting result. The initial image dataset is annotated based on the transmission line label information to obtain an annotated image dataset. The labeled image dataset is processed to generate a risk warning signal for the target power transmission channel. The risk warning signal is at least transmitted to the terminal of the inspection personnel to indicate that there is a high-risk power transmission channel.

[0006] Preferably, the step of extracting line segment information from the initial image dataset to obtain initial power transmission conductor information includes: The initial image dataset is preprocessed to obtain the image dataset to be detected; Edge detection algorithms are used to identify the image dataset to be detected and determine edge feature information; The edge feature information is fitted to determine the initial transmission line information.

[0007] Preferably, the structural analysis of the initial transmission line information, and the filtering of the analysis results to obtain the second transmission line information, wherein the structural analysis process is configured to analyze the similar principal directions and anisotropy of the line segment information at different spatial scales, including: The initial transmission line information is analyzed at different spatial scales to obtain a multi-scale principal direction matrix; The multi-scale principal direction matrix is ​​analyzed and processed to obtain the principal direction consistency score; The initial transmission line information was analyzed at different spatial scales to obtain a multi-scale anisotropy index; The initial transmission line information is filtered based on the main direction consistency score and the multi-scale anisotropy index to determine the second transmission line information.

[0008] Preferably, the step of performing parabolic fitting on the conductor profile formed by the information of the second transmission conductor based on the catenary mechanics principle includes: Extract the conductor contour point set from the second transmission conductor information to obtain discrete contour point coordinates; A catenary mechanical model constraint is established on the coordinates of the discrete contour points to obtain the catenary theoretical equations. Based on the theoretical equation of the catenary, the error optimization equation is determined by optimizing the deviation between the coordinates of the discrete contour points and the fitted curve. Solve the error optimization equation to obtain the initial fitting result; The initial fitting result is verified to determine the fitting result.

[0009] Preferably, the step of performing recognition processing on the labeled image dataset to generate a risk warning signal for the target power transmission channel includes: The labeled image dataset is compared with the constructed normal state benchmark library to detect feature differences and obtain abnormal feature regions. The abnormal feature regions are analyzed and processed to determine the risk type; The risk types are aggregated and analyzed to generate the risk warning signal.

[0010] Another aspect of the present invention provides a risk management system for power transmission channels, comprising: The acquisition module is used to acquire the initial image dataset of the target power transmission channel; The extraction module is used to extract line segment information from the initial image dataset to obtain initial power transmission conductor information; An analysis module is used to perform structural analysis on the initial transmission line information and filter the analysis results to obtain the second transmission line information. The structural analysis process is configured to analyze the similar principal directions and anisotropy of the line segment information at different spatial scales. The fitting module is used to perform parabolic fitting on the conductor profile formed by the second transmission conductor information based on the catenary mechanics principle, and to obtain the transmission conductor label information of the target transmission channel based on the fitting result. The annotation module is used to annotate the initial image dataset based on the transmission line label information to obtain an annotated image dataset; The identification module is used to perform identification processing on the labeled image dataset to generate a risk warning signal for the target power transmission channel. The risk warning signal is at least used to transmit to the terminal of the inspection personnel to indicate that there is a high-risk power transmission channel.

[0011] Preferably, the extraction module includes: The preprocessing unit is used to preprocess the initial image dataset to obtain the image dataset to be detected; The recognition unit is used to identify the image dataset to be detected using an edge detection algorithm and determine edge feature information; The processing unit is used to perform fitting processing on the edge feature information to determine the initial transmission conductor information.

[0012] Preferably, the analysis module includes: The main direction unit is used to analyze the initial transmission line information at different spatial scales to obtain a multi-scale main direction matrix; The consistency unit is used to analyze and process the multi-scale principal direction matrix to obtain the principal direction consistency score; Anisotropic units are used to analyze the initial transmission line information at different spatial scales to obtain multi-scale anisotropic indices. The filtering unit is used to filter the initial transmission line information based on the main direction consistency score and the multi-scale anisotropy index to determine the second transmission line information.

[0013] Preferably, the step of performing parabolic fitting on the conductor profile formed by the information of the second transmission conductor based on the catenary mechanics principle includes: The extraction unit is used to extract the conductor contour point set from the second transmission conductor information to obtain discrete contour point coordinates; The constraint unit is used to establish constraints on the catenary mechanical model for the coordinates of the discrete contour points, thereby obtaining the catenary theoretical equations. The optimization unit is used to determine the error optimization equation based on the catenary theoretical equation by optimizing the deviation between the coordinates of the discrete contour points and the fitted curve. The solving unit is used to solve the error optimization equation and obtain the initial fitting result; A verification unit is used to verify the initial fitting result and determine the fitting result.

[0014] Preferably, the step of performing recognition processing on the labeled image dataset to generate a risk warning signal for the target power transmission channel includes: The difference detection unit is used to perform feature difference detection between the labeled image dataset and the constructed normal state benchmark library to obtain abnormal feature regions; The risk unit is used to analyze and process the abnormal feature region to determine the risk type; The early warning signal unit is used to perform aggregate analysis on the risk types and generate the risk early warning signal.

[0015] Compared with the prior art, the beneficial effects of the present invention are at least one of the following: This invention effectively solves the problem of low efficiency caused by manual inspection in the prior art through a series of automated processing steps, thereby improving the efficiency of risk identification. First, an initial image dataset of the target transmission channel is acquired, and line segment information is extracted to obtain initial transmission conductor information. Next, structural analysis and filtering of this information yield more accurate secondary transmission conductor information. Then, based on the catenary mechanics principle, parabolic fitting is performed on the conductor profile to derive the transmission conductor label information, which is then used to annotate the initial image dataset, generating an annotated image dataset. Finally, through the recognition and processing of the annotated image dataset, a risk warning signal for the target transmission channel is generated, promptly notifying inspection personnel to pay attention to high-risk areas. This series of automated processes avoids the visual blind spots of manual inspection in low-light environments, greatly improving the accuracy and efficiency of hazard detection, reducing the risks caused by delayed response, and meeting the needs of modern power grids for proactive early warning and intelligent prevention and control. Therefore, this solution not only improves work efficiency but also significantly enhances the safety and stability of the power system. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating a risk management method for power transmission channels in one embodiment of the present invention. Figure 2 This is a schematic diagram of the structure of a risk management system for a power transmission channel in one embodiment of the present invention; Figure label: The module consists of: 11. Acquisition module; 12. Extraction module; 13. Analysis module; 14. Fitting module; 15. Labeling module; and 16. Recognition module. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The purpose of providing these embodiments is to make the disclosure of the present invention more thorough and comprehensive. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0018] In the description of this application, the terms "first," "second," "third," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Therefore, a feature defined with "first," "second," "third," etc., may explicitly or implicitly include one or more of that feature. In the description of this application, unless otherwise stated, "a plurality of" means two or more.

[0019] In the description of this application, it should be noted that, unless otherwise expressly specified and limited, the terms "installation," "connection," and "joint" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection between two components. The term "and / or" as used herein includes any and all combinations of one or more of the related listed items. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0020] In the description of this application, it should be noted that, unless otherwise defined, all technical and scientific terms used in this invention have the same meaning as commonly understood by one of ordinary skill in the art. The terminology used in this specification is for the purpose of describing specific embodiments only and is not intended to limit the invention. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0021] Power transmission channels are critical infrastructure for ensuring energy security and green development. However, sagging conductors or tree branches encroaching on the power line can easily cause serious hazards such as power outages, power trips, and even wildfires. Currently, inspections mainly rely on manual checks, which are limited by low-light conditions such as nighttime, rainy days, or forest areas. These methods suffer from difficulties in identification, delayed response, low efficiency, and high safety risks, making it difficult to meet the needs of modern power grids for proactive early warning and intelligent control.

[0022] One embodiment of the present invention provides a risk management method for power transmission channels. For details, please refer to [link / reference needed]. Figure 1 , Figure 1 The diagram shown is a flowchart illustrating a risk management method for power transmission channels according to one embodiment of the present invention. The method includes: S1. Obtain the initial image dataset of the target power transmission channel; S2. Extract line segment information from the initial image dataset to obtain initial power transmission conductor information; S3. Perform structural analysis on the initial transmission line information, and use the analysis results to filter and obtain the second transmission line information. The structural analysis process is configured to analyze the similar principal directions and anisotropy of line segment information at different spatial scales. S4. Based on the catenary mechanics principle, parabolic fitting is performed on the conductor profile formed by the second transmission conductor information, and the transmission conductor label information of the target transmission channel is obtained by inversion based on the fitting result. S5. Based on the transmission line label information, the initial image dataset is annotated to obtain an annotated image dataset; S6. The labeled image dataset is processed to generate a risk warning signal for the target power transmission channel. The risk warning signal is at least transmitted to the terminal of the inspection personnel to indicate the power transmission channel with high risk.

[0023] Obtain the initial image dataset of the target transmission line. Current technologies for transmission line inspection rely heavily on manual on-site inspections or simple image comparisons, resulting in low efficiency and high missed detection rates. The initial image dataset is fundamental for subsequent extraction of transmission conductor information, analysis of structural features, and risk warning. Only by acquiring comprehensive and accurate image data can a reliable basis be provided for the entire transmission line risk identification process. The initial image dataset is a collection of visualized data on the equipment and environment along the target transmission line, specifically covering images of scenes such as transmission conductors, towers, insulators, surrounding vegetation and buildings. It includes both distant images of the overall conductor alignment and close-up images showing the detailed state of the conductors. Employing mature inspection methods from existing technologies, drones equipped with high-definition cameras fly along a pre-set route along the power transmission corridor, capturing images at a controlled altitude that clearly captures conductor details and covers the surrounding environment. The camera resolution is set to 4K to ensure image clarity. The frequency of shooting is increased in critical sections of the transmission corridor, such as crossing sections and mountainous areas. Furthermore, ground-based fixed monitoring equipment and image acquisition modules on inspection robots are used to supplement coverage of blind spots that drones cannot reach. During the acquisition process, relevant information such as shooting time and geographical location is recorded simultaneously. All acquired raw images are then filtered and deduplicated, removing blurry, overexposed, or severely obscured images to form a complete initial image dataset. Compared to existing technologies that rely solely on manual inspection or partial image acquisition, this method achieves comprehensive, blind-spot-free image coverage of the power transmission corridor, obtaining more comprehensive and clearer image data. This provides high-quality raw material for subsequent line segment information extraction and conductor structure analysis, reducing identification errors caused by insufficient data. It also improves inspection efficiency, shortens the data acquisition cycle, and allows more time for risk warning, helping inspection personnel to promptly identify potential hazards in the power transmission corridor and ensuring the safe and stable transmission of electricity.

[0024] Furthermore, line segment information is extracted from the initial image dataset to obtain the initial transmission line information. The initial image dataset is preprocessed to obtain the image dataset to be detected. An edge detection algorithm is used to identify edge features in the image dataset to determine edge feature information. The edge feature information is then fitted to determine the initial transmission line information. Directly identifying transmission line information using algorithms is easily affected by interference factors such as tower and vegetation shadows in the image, resulting in low accuracy. Transmission lines appear as continuous linear structures in the image, and line segment information is the core visual feature of the transmission line. Extracting line segment information can remove interference and focus on the essential features of the transmission line, laying the foundation for accurate acquisition of the initial transmission line information. The initial transmission line information is preliminary data containing key parameters such as the position, direction, and length of the transmission line in the image. The image dataset to be detected is a collection of images that have been preprocessed from the initial images to remove interference and improve quality. Edge feature information is the contour line data formed by the gray-level difference between objects and the background in the image. Edge detection algorithms are commonly used tools in existing technologies for extracting image edges and can capture gray-level abrupt changes in object contours. First, the initial image dataset is preprocessed. Color images are converted to grayscale using existing grayscale conversion techniques. Noise points are removed using a Gaussian filtering algorithm, and contrast enhancement is performed to improve the distinction between the conductor and the background, resulting in the image dataset to be detected. Then, a mature edge detection algorithm is used, setting appropriate high and low thresholds to capture all edges with abrupt grayscale changes, identifying edge features including conductor edges, tower edges, and vegetation edges. Next, linear fitting is performed on the edge features, using the least squares method to remove short, scattered non-conductor edge segments, retaining continuous segments that conform to the conductor's direction, ultimately determining the initial transmission conductor information. Compared to existing methods that directly identify conductor information, extracting line segment information effectively filters out non-conductor interference elements, improving the accuracy and completeness of the initial transmission conductor information. This avoids missed or false detections due to complex backgrounds, providing high-quality data support for subsequent structural analysis and parabolic fitting. Furthermore, relying on mature edge detection and fitting algorithms reduces recognition difficulty, improves processing efficiency, and ensures the smooth progress of the entire transmission channel risk warning process.

[0025] Furthermore, structural analysis is performed on the initial transmission line information. The results are then used to filter and obtain the second transmission line information. The structural analysis process involves analyzing the similar principal directions and anisotropy of line segments at different spatial scales. The initial transmission line information is analyzed at different spatial scales to obtain a multi-scale principal direction matrix. This matrix is ​​then processed to obtain a principal direction consistency score. Similarly, the initial transmission line information is analyzed at different spatial scales to obtain a multi-scale anisotropy index. Based on the principal direction consistency score and the multi-scale anisotropy index, the second transmission line information is determined. Because the initial transmission line information still contains interfering segments such as tower edges and vegetation branches, which are easily confused with the transmission line segments at a single spatial scale, and transmission lines, as continuously extending linear structures, exhibit consistent principal directions and significant anisotropy at different spatial scales, structural analysis can accurately remove these interferences, yielding pure second transmission line information and providing reliable data for subsequent fitting and inversion. The second transmission line information consists of core data that, after structural analysis and screening, eliminates interfering segments and retains only the relevant features of the actual transmission line. Spatial scale refers to the different pixel range windows set when analyzing the image, such as analysis units of different sizes like 10×10 pixels, 20×20 pixels, and 30×30 pixels. The principal direction is the main orientation of the line segment in space; the multi-scale principal direction matrix is ​​a collection recording the principal direction data of each line segment at different spatial scales. The principal direction consistency score is a quantitative value that measures the degree of overlap of the principal directions of line segments at different spatial scales; a higher score indicates a more unified principal direction. The anisotropy index reflects the uniformity of the distribution of line segments in different directions; the anisotropy index of linear structures like transmission lines is significantly higher than that of randomly distributed interfering segments. The specific approach involves first employing a mature multi-scale analysis method, using spatial scale windows of 10×10 pixels, 20×20 pixels, and 30×30 pixels respectively, to detect the direction of each line segment in the initial transmission conductor information, recording the principal direction data at each scale, and constructing a multi-scale principal direction matrix. Then, a cosine similarity algorithm is used to calculate the matching degree of the principal directions at different scales, converting the matching degree into a principal direction consistency score of 0 to 1. For example, if the principal directions are completely consistent across all three scales, a score of 1 is obtained; if they are consistent across two scales, a score of 0.7 is obtained. Simultaneously, at each spatial scale, the multi-scale anisotropy index is obtained by calculating the variance ratio of the line segment distribution. Linearly extending line segments have a large variance ratio and an index close to 1, while disordered line segments have a small variance ratio and an index close to 0. Finally, a screening threshold is set: line segments with a principal direction consistency score higher than 0.6 and an anisotropy index higher than 0.8 are determined as valid conductor line segments, while the rest are discarded as interference line segments. The valid line segments are then integrated to obtain the second transmission conductor information.Compared to existing technologies that rely solely on single-feature screening, this method combines multi-scale principal direction consistency and anisotropy index to more accurately distinguish between conductors and interfering segments, significantly improving the purity of the second transmission conductor information and avoiding errors caused by interference during subsequent fitting. Furthermore, relying on mature multi-scale analysis and similarity calculation algorithms ensures the efficiency and stability of the screening process, providing a solid foundation for the accurate inversion of transmission conductor label information and thereby enhancing the reliability of risk warnings for the entire transmission channel.

[0026] Furthermore, based on the catenary mechanics principle, a parabolic fit is performed on the conductor profile formed by the information of the second transmission conductor, and the transmission conductor label information of the target transmission channel is obtained by inversion based on the fitting results. The conductor profile point set is extracted from the second transmission conductor information to obtain discrete profile point coordinates; a catenary mechanics model constraint is established on the discrete profile point coordinates to obtain the catenary theoretical equation; based on the catenary theoretical equation, the error optimization equation is determined by optimizing the deviation between the discrete profile point coordinates and the fitted curve; the error optimization equation is solved to obtain the initial fitting result; the initial fitting result is verified to determine the final fitting result. Because the conductor profile in the second transmission conductor information is composed of discrete points, it cannot directly reflect the complete shape and key parameters of the conductor. However, the actual suspension shape of the transmission conductor under its own weight conforms to the catenary mechanics principle. Parabolic fitting can simplify complex calculations and accurately approximate the actual shape. Based on the fitting results, core parameters can be obtained by inversion, providing a crucial basis for subsequent risk warning. The transmission conductor label information contains annotation data including core parameters such as conductor sag, span, and suspension point height. The catenary mechanics principle describes the physical law of the suspension shape of a flexible object under its own weight. When there is no external force interference, the conductor will naturally form a catenary shape. Parabolic fitting is a mathematical method that uses the parabolic equation to approximate the catenary shape. Discrete contour point coordinates are the position data of key pixels on the edge of the conductor extracted from the second transmission conductor information. First, continuously distributed edge pixels of the transmission line are selected from the second transmission line information, and the x and y coordinates of each point are recorded to obtain discrete contour point coordinates. Then, based on the mature catenary mechanical model in the existing technology, combined with known parameters such as the material density and weight per unit length of the transmission line, constraints are established on the discrete contour point coordinates to derive the catenary theoretical equation describing the ideal suspension shape of the transmission line. Subsequently, based on the catenary theoretical equation, the least squares method is used to optimize the deviation between the discrete contour point coordinates and the fitted curve to construct an error optimization equation, which aims to minimize the sum of squared deviations. Then, the error optimization equation is solved using the gradient descent algorithm to obtain the initial fitting result that initially fits the discrete points. Finally, the initial fitting result is compared with the theoretical value of the catenary theoretical equation to calculate the fitting error. If the error is lower than a preset threshold, it is determined as the final fitting result. If it exceeds the threshold, the optimization parameters are readjusted and the calculation is repeated. Finally, the transmission line label information such as the sag and span of the transmission line is obtained based on the final fitting result. Compared to existing technologies that rely solely on image features to infer conductor parameters, parabolic fitting combined with the principles of catenary mechanics more closely resembles the true physical shape of the conductor, resulting in more accurate inverted label information. This avoids parameter errors caused by image distortion or interference, providing reliable conductor parameter support for subsequent labeled image datasets and risk identification. Furthermore, by leveraging mature mathematical optimization algorithms and mechanical models, it improves fitting efficiency and stability, helping to accurately determine whether conductors have abnormal sag or other safety hazards, thus ensuring the accuracy of risk warnings for power transmission channels.

[0027] Preferably, the initial image dataset is annotated based on the transmission line label information to obtain an annotated image dataset. Because the initial image dataset only contains raw visual information and lacks a clear correlation with key parameters of the transmission lines, it cannot be directly used for subsequent risk identification. Annotation processing binds the transmission line label information to the image pixel locations, giving the image data semantic interpretability and laying the foundation for accurate risk identification. The annotated image dataset is a visual data set overlaid with transmission line label information, specifically including the location bounding box of the transmission line in the image, sag numerical annotations, and span range markings, which can intuitively present the correspondence between key parameters of the transmission lines and the image scene. The transmission line label information, including the inverted core parameters such as conductor sag, span, and suspension point height, is the core basis for annotation. This study utilizes mature image annotation tools to first import an initial image dataset and transmission line label information. Using the tool's coordinate matching function, key parameters of the transmission lines in the label information are aligned with the actual pixel positions of the transmission lines in the images. For example, based on the sag parameter, the position of the lowest point of the transmission line and its corresponding sag value are marked in the image; based on the span parameter, the image area between the two suspension points of the transmission line is selected and the span length is marked. Then, following a unified annotation standard, the transmission lines in each image are classified and labeled to distinguish different transmission line types such as phase lines and ground lines. Simultaneously, information such as the distance between the transmission lines and surrounding vegetation and buildings is marked. Finally, the annotation results are reviewed, and images with annotation errors and parameter misalignments are removed, resulting in a complete annotated image dataset. Compared to the unannotated raw images in existing technologies, the annotated image dataset transforms abstract transmission line parameters into intuitive markers in the image, allowing subsequent recognition algorithms to quickly locate transmission line areas and obtain key parameters, improving the efficiency and accuracy of risk identification, avoiding missed detections of potential hazards due to blurred image information. Furthermore, the annotated dataset can be used to train more accurate risk identification models, continuously optimizing the performance of the transmission channel risk warning system and ensuring power transmission safety.

[0028] Preferably, the labeled image dataset is processed to generate a risk warning signal for the target transmission channel. This risk warning signal is transmitted to the terminals of inspection personnel, indicating transmission channels with high risks. The labeled image dataset is compared with a pre-built normal state benchmark library for feature difference detection to obtain abnormal feature regions. These abnormal feature regions are analyzed to determine the risk type. The risk types are then aggregated to generate a risk warning signal. While the labeled image dataset contains conductor parameters and image information, safety hazards must be identified through processing. Current technologies for traditional transmission channel risk screening rely on manual inspections, resulting in low efficiency, high missed detection rates, and inability to respond in real time. Automatic identification and processing, however, can quickly locate risks and generate warning signals, allowing inspection personnel to take timely action and ensuring the safety of the transmission channel. The labeled image dataset is a collection of images with labels such as conductor sag and span superimposed on them. The normal state benchmark library is a standard feature library built based on a large number of risk-free transmission channel images, which includes benchmark parameters such as normal conductor shape and safe distance from surrounding objects. Abnormal feature regions are image regions in the labeled images that do not match the features in the benchmark library. Risk type is the category of safety hazard determined based on abnormal features. Risk warning signal is instruction information containing risk location, type and level, which can be transmitted wirelessly to the mobile phones and tablets of inspection personnel. First, mature feature extraction algorithms from existing technologies are used to extract features such as conductor morphology, distance to surrounding objects, and conductor damage marks from the labeled image dataset. These features are then compared with corresponding features in the normal state benchmark library. By calculating feature similarity, regions with similarity below a preset threshold are selected as abnormal feature regions, such as areas where conductor sag exceeds the benchmark range and areas where the distance to trees is less than the safety standard. Next, image classification models from deep learning are used to analyze abnormal feature regions and determine the risk type. Common types include abnormal sag, conductor damage, excessive vegetation, and foreign object entanglement. Subsequently, multiple risk types of the same transmission channel are aggregated and analyzed, and levels are classified according to the scope and severity of the risk impact. For example, excessive vegetation less than 1 meter away from the conductor is classified as a high-risk risk, and slight conductor wear is classified as a general risk. Finally, a risk warning signal containing the risk location, type, and level is generated and transmitted to the inspection personnel's terminal in real time. Compared to manual inspections in existing technologies, this technology enables automated and real-time risk identification, significantly improving inspection efficiency, reducing labor costs and the probability of missed inspections. At the same time, the early warning signals can accurately indicate high-risk areas, allowing inspection personnel to carry out targeted measures, avoiding blind inspections, shortening the time for handling hidden dangers, reducing the probability of accidents such as power line tripping and short circuits, and ensuring the stability and reliability of power transmission.

[0029] One embodiment of the present invention provides a risk management system for power transmission channels. For details, please refer to [link / reference needed]. Figure 2 , Figure 2The diagram shown illustrates the structure of a risk management system for a power transmission channel according to one embodiment of the present invention. The system includes: The acquisition module is used to acquire the initial image dataset of the target power transmission channel; The extraction module is used to extract line segment information from the initial image dataset to obtain the initial power transmission conductor information; The analysis module is used to perform structural analysis on the initial transmission line information and to filter the analysis results to obtain the second transmission line information. The structural analysis process is configured to analyze the similar principal directions and anisotropy of line segment information at different spatial scales. The fitting module is used to perform parabolic fitting on the conductor profile formed by the information of the second transmission conductor based on the catenary mechanics principle, and to obtain the transmission conductor label information of the target transmission channel based on the fitting results. The annotation module is used to annotate the initial image dataset based on the transmission line label information to obtain an annotated image dataset. The identification module is used to identify and process the labeled image dataset to generate a risk warning signal for the target power transmission channel. The risk warning signal is at least transmitted to the terminal of the inspection personnel to indicate that there is a high risk in the power transmission channel.

[0030] Preferably, the extraction module includes: The preprocessing unit is used to preprocess the initial image dataset to obtain the image dataset to be detected; The recognition unit is used to identify the image dataset to be detected using an edge detection algorithm and determine edge feature information; The processing unit is used to fit the edge feature information and determine the initial transmission line information.

[0031] Preferably, the analysis module includes: The principal direction element is used to analyze the initial transmission line information at different spatial scales to obtain a multi-scale principal direction matrix; The consistency unit is used to analyze and process the multi-scale principal direction matrix to obtain the principal direction consistency score; Anisotropic elements are used to analyze initial transmission line information on a spatial scale to obtain multi-scale anisotropic indices. The filtering unit is used to filter the initial transmission conductor information based on the main direction consistency score and the multi-scale anisotropy index to determine the second transmission conductor information.

[0032] Preferably, based on the catenary mechanics principle, parabolic fitting is performed on the conductor profile formed by the information of the second transmission conductor, including: The extraction unit is used to extract the conductor contour point set from the second transmission conductor information to obtain the discrete contour point coordinates. Constraint elements are used to establish constraints on the coordinates of discrete contour points to create a catenary mechanical model constraint, thereby obtaining the catenary theoretical equations. The optimization unit is used to determine the error optimization equation based on the catenary theory equation by optimizing the deviation between the coordinates of discrete contour points and the fitted curve. The solving unit is used to solve the error optimization equation and obtain the initial fitting result; The validation unit is used to validate the initial fitting results and determine the final fitting result.

[0033] Preferably, the labeled image dataset is subjected to recognition processing to generate a risk warning signal for the target power transmission channel, including: The difference detection unit is used to perform feature difference detection between the labeled image dataset and the constructed normal state benchmark library to obtain abnormal feature regions; Risk units are used to analyze and process regions with abnormal characteristics to determine the type of risk. The early warning signal unit is used to perform aggregate analysis on risk types and generate risk early warning signals.

[0034] 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 program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0035] Accordingly, embodiments of the present invention provide a computer-readable storage medium comprising a stored computer program, wherein, when the computer program is executed, it controls the device containing the computer-readable storage medium to perform steps in the risk handling method for power transmission channels as described in the above embodiments, for example... Figure 1 Steps S1 to S6 as described above.

[0036] This invention offers several significant advantages from a technical implementation perspective. Firstly, it represents a qualitative leap in automation. By automatically acquiring and analyzing transmission line image datasets, it reduces reliance on manual inspections, especially in harsh or inaccessible environments. This automated approach greatly enhances the safety and feasibility of operations. Secondly, regarding accuracy, the method combining structural analysis with the principles of catenary mechanics for conductor contour fitting ensures highly accurate identification results, effectively avoiding misjudgments or omissions that may occur during manual inspection.

[0037] Furthermore, this invention possesses strong adaptability, capable of coping with various complex meteorological conditions and geographical environments. It can operate stably and reliably in low-light conditions such as nighttime, rainy weather, or forest cover, providing reliable hazard detection capabilities. This not only compensates for the ineffectiveness of traditional methods under specific conditions but also provides year-round monitoring and support for the operation and maintenance of power systems.

[0038] Finally, by generating and transmitting risk warning signals in real time, this invention enables inspection personnel to receive high-risk notifications immediately and take swift action, thereby significantly shortening response time and reducing the probability of potential accidents. In summary, this solution, with its high degree of automation, accuracy, adaptability, and real-time response capabilities, provides a novel solution for improving the safety management and operation and maintenance efficiency of power transmission channels.

[0039] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

Claims

1. A method for handling risks in power transmission channels, characterized in that, include: Obtain the initial image dataset of the target power transmission channel; Extract line segment information from the initial image dataset to obtain initial power transmission conductor information; Structural analysis is performed on the initial transmission line information, and the analysis results are used to filter and obtain the second transmission line information. The structural analysis process is configured to analyze the similar principal directions and anisotropy of the line segment information at different spatial scales. Based on the catenary mechanics principle, the conductor profile formed by the second transmission conductor information is parabolically fitted, and the transmission conductor label information of the target transmission channel is obtained by inversion based on the fitting result. The transmission conductor label information includes conductor sag, span and suspension point height. The initial image dataset is annotated based on the transmission line label information to obtain an annotated image dataset. The labeled image dataset is processed to generate a risk warning signal for the target power transmission channel. The risk warning signal is at least transmitted to the terminal of the inspection personnel to indicate that there is a high-risk power transmission channel. The parabolic fitting of the conductor profile formed by the information of the second transmission conductor, based on the catenary mechanics principle, includes: Extract the conductor contour point set from the second transmission conductor information to obtain discrete contour point coordinates; A catenary mechanical model constraint is established on the coordinates of the discrete contour points to obtain the catenary theoretical equations. Based on the theoretical equation of the catenary, the error optimization equation is determined by optimizing the deviation between the coordinates of the discrete contour points and the fitted curve. Solve the error optimization equation to obtain the initial fitting result; The initial fitting result is verified to determine the fitting result.

2. The risk management method for power transmission channels as described in claim 1, characterized in that, The step of extracting line segment information from the initial image dataset to obtain initial power transmission conductor information includes: The initial image dataset is preprocessed to obtain the image dataset to be detected; Edge detection algorithms are used to identify the image dataset to be detected and determine edge feature information; The edge feature information is fitted to determine the initial transmission line information.

3. The risk management method for power transmission channels as described in claim 1, characterized in that, The initial transmission line information is subjected to structural analysis, and the analysis results are used to filter and obtain the second transmission line information. The structural analysis process is configured to analyze the similar principal directions and anisotropies of the line segment information at different spatial scales, including: The initial transmission line information is analyzed at different spatial scales to obtain a multi-scale principal direction matrix; The multi-scale principal direction matrix is ​​analyzed and processed to obtain the principal direction consistency score; The initial transmission line information was analyzed at different spatial scales to obtain a multi-scale anisotropy index; The initial transmission line information is filtered based on the main direction consistency score and the multi-scale anisotropy index to determine the second transmission line information.

4. The risk management method for power transmission channels as described in claim 1, characterized in that, The step of performing recognition processing on the labeled image dataset to generate a risk warning signal for the target power transmission channel includes: The labeled image dataset is compared with the constructed normal state benchmark library to detect feature differences and obtain abnormal feature regions. The abnormal feature regions are analyzed and processed to determine the risk type; The risk types are aggregated and analyzed to generate the risk warning signals.

5. A risk management system for power transmission channels, characterized in that, include: The acquisition module is used to acquire the initial image dataset of the target power transmission channel; The extraction module is used to extract line segment information from the initial image dataset to obtain initial power transmission conductor information; An analysis module is used to perform structural analysis on the initial transmission line information and filter the analysis results to obtain the second transmission line information. The structural analysis process is configured to analyze the similar principal directions and anisotropy of the line segment information at different spatial scales. The fitting module is used to perform parabolic fitting on the conductor profile formed by the second transmission conductor information based on the catenary mechanics principle, and to obtain the transmission conductor label information of the target transmission channel based on the fitting result. The transmission conductor label information includes conductor sag, span and suspension point height. The annotation module is used to annotate the initial image dataset based on the transmission line label information to obtain an annotated image dataset; The identification module is used to perform identification processing on the labeled image dataset and generate a risk warning signal for the target power transmission channel based on the processing result. The risk warning signal is at least used to transmit to the terminal of the inspection personnel to indicate that there is a power transmission channel with high risk. The parabolic fitting of the conductor profile formed by the information of the second transmission conductor, based on the catenary mechanics principle, includes: The extraction unit is used to extract the conductor contour point set from the second transmission conductor information to obtain discrete contour point coordinates; The constraint unit is used to establish constraints on the catenary mechanical model for the coordinates of the discrete contour points, thereby obtaining the catenary theoretical equations. The optimization unit is used to determine the error optimization equation based on the catenary theoretical equation by optimizing the deviation between the coordinates of the discrete contour points and the fitted curve. The solving unit is used to solve the error optimization equation and obtain the initial fitting result; A verification unit is used to verify the initial fitting result and determine the fitting result.

6. The risk management system for power transmission channels as described in claim 5, characterized in that, The extraction module includes: The preprocessing unit is used to preprocess the initial image dataset to obtain the image dataset to be detected; The recognition unit is used to identify the image dataset to be detected using an edge detection algorithm to determine edge feature information; The processing unit is used to perform fitting processing on the edge feature information to determine the initial transmission conductor information.

7. The risk management system for power transmission channels as described in claim 5, characterized in that, The analysis module includes: The main direction unit is used to analyze the initial transmission line information at different spatial scales to obtain a multi-scale main direction matrix; The consistency unit is used to analyze and process the multi-scale principal direction matrix to obtain the principal direction consistency score; Anisotropic units are used to analyze the initial transmission line information at different spatial scales to obtain multi-scale anisotropic indices. The filtering unit is used to filter the initial transmission line information based on the main direction consistency score and the multi-scale anisotropy index to determine the second transmission line information.

8. The risk management system for power transmission channels as described in claim 5, characterized in that, The step of performing recognition processing on the labeled image dataset to generate a risk warning signal for the target power transmission channel includes: The difference detection unit is used to perform feature difference detection between the labeled image dataset and the constructed normal state benchmark library to obtain abnormal feature regions; The risk unit is used to analyze and process the abnormal feature region to determine the risk type; The early warning signal unit is used to perform aggregate analysis on the risk types and generate the risk early warning signal.

Images