Power transmission channel security maintenance method and device, equipment and storage medium

Abstract

This application discloses a method, apparatus, equipment, and storage medium for the safety maintenance of power transmission channels, relating to the field of power transmission channel safety maintenance technology. The method involves acquiring three-dimensional point cloud data of the power transmission channel; classifying the three-dimensional point cloud data into point cloud types to identify ground points, power line points, tower points, and vegetation points; extracting and integrating features from the ground points, power line points, tower points, and vegetation points respectively to obtain a point cloud statistical feature set; matching the point cloud statistical feature set with a preset procedure knowledge base to determine the operating status and potential risk level of each section of the power transmission channel; and calling a preset set of handling strategies to generate operating procedure strategies based on the operating status and potential risk level of each section, and applying them to the power transmission channel. Compared to traditional manual inspections, this application significantly improves maintenance efficiency and accuracy, reduces reliance on human experience, and ensures the safety of the power transmission channel.

Description

Technical Field

[0001] This application relates to the field of power transmission channel safety maintenance technology, specifically to a power transmission channel safety maintenance method, device, equipment, and storage medium. Background Technology

[0002] With the continuous expansion of power grid construction and the widespread application of ultra-high voltage power transmission technology, the transmission channels, as the core lifeline of the power system, face severe challenges to their operational safety and stability. Transmission lines typically cross complex terrains such as mountains, hills, and forests, and are easily affected by external environmental factors such as tree growth, illegal construction, and geological disasters.

[0003] Traditional power transmission channel operation and maintenance mainly relies on manual periodic inspections. Inspection personnel use tools such as theodolites and telescopes to conduct on-site measurements. This method is not only labor-intensive and has blind spots, but the inspection results are also greatly affected by human experience, making it difficult to meet the high requirements of modern power grids for the safety maintenance of power transmission channels. Summary of the Invention

[0004] In view of this, this application provides a method, apparatus, equipment and storage medium for the safe maintenance of power transmission channels, which solves the problems that traditional methods are not only labor-intensive and have blind spots, but also have detection results that are greatly affected by human experience, making it difficult to meet the high requirements of modern power grids for the safe maintenance of power transmission channels.

[0005] To achieve the above objectives, the following solution is proposed:

[0006] Firstly, a method for the safe maintenance of power transmission channels includes:

[0007] Acquire 3D point cloud data of the power transmission channel;

[0008] The three-dimensional point cloud data is divided into point cloud types to determine ground points, power line points, tower points, and vegetation points;

[0009] Feature extraction and integration are performed on the ground points, power line points, tower points, and vegetation points respectively to obtain a point cloud statistical feature set;

[0010] The point cloud statistical feature set is matched with a preset procedure knowledge base to determine the operating status and potential risk level of each section of the power transmission channel.

[0011] The system invokes a set of preset handling strategies, generates operating procedures and strategies based on the operating status and potential risk level of each section, and applies them to the power transmission channel.

[0012] Preferably, the step of classifying the three-dimensional point cloud data into point cloud types to determine ground points, power line points, tower points, and vegetation points includes:

[0013] The three-dimensional point cloud data is flipped along the Z-axis in a preset spatial rectangular coordinate system to obtain the first point cloud data;

[0014] A virtual grid is generated and overlaid on the first point cloud data;

[0015] The virtual mesh is sunk in the opposite direction of the Z-axis. During the sunk process, when the virtual mesh collides with the first point cloud data, the virtual mesh is fixed at the position corresponding to the collision in the virtual mesh to determine each collision point.

[0016] The collision point is taken as the ground point, and the other points in the three-dimensional point cloud data other than the ground points are taken as non-ground points.

[0017] Determine the power line points, tower points, and vegetation points from each of the aforementioned non-ground points.

[0018] Preferably, determining the power line point, tower point, and vegetation point from the various non-ground points includes:

[0019] Calculate the average elevation of the ground points;

[0020] The target height is obtained by adding the average elevation to the preset vegetation starting height threshold.

[0021] Obtain the elevation values ​​of each of the non-ground points, and designate the non-ground points with elevation values ​​greater than the target height as vegetation points;

[0022] Calculate the linearity of each non-ground point other than the vegetation point;

[0023] Points with linearity greater than a preset linearity threshold are designated as power line points, and points with linearity less than the linearity threshold are designated as tower points.

[0024] Preferably, the process of feature extraction for power line points includes:

[0025] For each electric field line point, an initial subset is randomly selected, and a catenary equation is fitted to the initial subset.

[0026] Solve the catenary equation to obtain the current stress;

[0027] Obtain the current working condition, substitute the current working condition and current stress into the preset conductor state equation and solve it to obtain the future stress;

[0028] The sag characteristics are calculated based on the future stress.

[0029] Preferably, the process of feature extraction for vegetation points includes:

[0030] The three-dimensional point cloud data is spatially voxelized into a mesh according to a preset voxel size to obtain individual voxel meshes.

[0031] The voxel grid containing the vegetation points is determined as each vegetation grid, and the number of vegetation grids and the number of voxels in each vegetation grid are counted.

[0032] The voxel density of each vegetation grid is calculated based on the number of voxels and the voxel size.

[0033] The total vegetation volume is calculated based on the vegetation grid and voxel dimensions.

[0034] Preferably, the process of establishing the procedure knowledge base includes:

[0035] Determine the voltage levels included in the power system, set different minimum vertical safety distances and minimum horizontal safety distances according to different voltage levels, and establish a voltage level rule table based on each voltage level, minimum vertical safety distance, and minimum horizontal safety distance;

[0036] Determine the slope of each terrain where all existing transmission channels are located, set different operational constraints based on different terrain slopes, and establish an operational constraint rule table;

[0037] Multiple temperature ranges are established, and a safety distance correction factor is set for each of the temperature ranges to establish a meteorological correction rule table;

[0038] Multiple distance ranges are established for the distance between vegetation and power lines, and a risk level is set for each distance range to create a vegetation risk level rule table;

[0039] Multiple sag ranges are established for the sag status of power lines, and a warning level is set for each sag range to establish a sag warning level rule table;

[0040] The voltage level rule table, operation constraint rule table, meteorological correction rule table, vegetation risk level rule table, and sag warning level rule table are combined to obtain the procedure knowledge base.

[0041] Preferably, the method further includes global feature extraction of the three-dimensional point cloud data, including:

[0042] For each point in the three-dimensional point cloud data, determine the neighborhood point set of that point, and calculate the covariance matrix of the neighborhood point set;

[0043] Singular value decomposition is performed on the covariance matrix to obtain the first eigenvalue, second eigenvalue, and third eigenvalue;

[0044] The linearity feature is calculated using the first and second eigenvalues.

[0045] Flatness features are calculated using the first, second, and third eigenvalues;

[0046] The scattering characteristics are calculated using the first and third eigenvalues.

[0047] Secondly, a safety maintenance device for a power transmission channel includes:

[0048] The 3D point cloud data acquisition module is used to acquire 3D point cloud data of the power transmission channel;

[0049] The type classification module is used to classify the three-dimensional point cloud data into point cloud types to determine ground points, power line points, tower points, and vegetation points.

[0050] The feature extraction module is used to extract and integrate the features of the ground points, power line points, tower points and vegetation points respectively to obtain a point cloud statistical feature set;

[0051] The matching module is used to match the point cloud statistical feature set with a preset procedure knowledge base to determine the operating status and potential risk level of each section of the power transmission channel.

[0052] The operation procedure strategy generation module is used to call a preset set of disposal strategies, generate operation procedure strategies based on the operation status and potential risk level of each section, and apply them to the power transmission channel.

[0053] Thirdly, a safety maintenance device for a power transmission channel includes a memory and a processor;

[0054] The memory is used to store programs;

[0055] The processor is configured to execute the program to implement the various steps of the power transmission channel safety maintenance method as described in any of the first aspects.

[0056] Fourthly, a storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method for safe maintenance of a power transmission channel as described in any of the first aspects.

[0057] As can be seen from the above technical solution, this application obtains three-dimensional point cloud data of the power transmission channel; classifies the three-dimensional point cloud data into point cloud types to determine ground points, power line points, tower points, and vegetation points; extracts and integrates features from the ground points, power line points, tower points, and vegetation points respectively to obtain a point cloud statistical feature set; matches the point cloud statistical feature set with a preset procedure knowledge base to determine the operating status and potential risk level of each section of the power transmission channel; and calls a preset set of disposal strategies to generate operating procedure strategies based on the operating status and potential risk level of each section, and applies them to the power transmission channel. This application first acquires three-dimensional point cloud data of the power transmission channel. However, the original three-dimensional point cloud data only contains spatial coordinate information and lacks semantic meaning (i.e., what object each point represents). This application classifies the point cloud data by type, assigning category labels such as ground, power lines, towers, and vegetation, thus realizing the transformation from data to information. After separating the overall point cloud by category, subsequent feature extraction can use different algorithms for different types, avoiding the accuracy loss caused by a one-size-fits-all approach and improving the accuracy and efficiency of the analysis. Therefore, in subsequent feature extraction, key feature parameters that best reflect the safety status of different types of ground features can be extracted for subsequent risk assessment. During the analysis and matching process, a regulatory knowledge base is used for matching. This knowledge base integrates power industry standards, safety regulations, and expert experience, digitizing and structuring the rules scattered across various regulatory documents. This allows for accurate analysis of the operating status and potential risk levels of each section of the transmission channel, automatically determining whether there are safety hazards in each section and quantifying their risk levels. Finally, based on the operating status and potential risk levels, corresponding handling strategies are automatically matched and generated, achieving an automated closed loop from problem discovery to solution provision. Compared to traditional manual inspections, this significantly improves operation and maintenance efficiency and accuracy, reduces reliance on human experience, and ensures the safety of the transmission channel. Attached Figure Description

[0058] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0059] Figure 1 An optional flowchart of a safety maintenance method for a power transmission channel provided in an embodiment of this application;

[0060] Figure 2 A schematic diagram of the structure of a safety maintenance device for a power transmission channel provided in an embodiment of this application;

[0061] Figure 3 This is a schematic diagram of the structure of a safety maintenance device for a power transmission channel provided in an embodiment of this application. Detailed Implementation

[0062] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0063] This invention can be used in a wide variety of general-purpose or special-purpose computing environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor devices, distributed computing environments including any of the above devices, etc.

[0064] In recent years, remote sensing technologies, represented by airborne lidar (LiDAR), have gradually replaced traditional manual inspections as the mainstream method due to their advantages of penetrating vegetation canopies and obtaining high-precision three-dimensional spatial coordinates. However, existing point cloud processing technologies still have significant limitations. For example, determining the presence of potential hazards by calculating the minimum Euclidean distance (geometric distance) between the guide point cloud and vegetation points, while solving the accuracy problem of distance measurement, is essentially still an alarm mechanism based on a single geometric threshold. It lacks in-depth mining of the local statistical features of the point cloud, which means that the system has difficulty distinguishing noise from real obstacles and cannot quantify the biomass or growth status of vegetation through point cloud statistics, resulting in a high false alarm rate and an inability to assess the amount of cleanup work. Furthermore, while some existing implementations have achieved automation of inspection paths and 3D visualization of data, their backend applications are still in the "human-computer interaction interpretation" stage, only outputting a list of hidden danger points or alarm signals. The massive point cloud data forms data silos and fails to achieve logical integration with the standard operating procedures of the power industry, such as the "Operating Procedures for Overhead Transmission Lines". Therefore, after receiving an alarm, maintenance personnel still need to manually refer to the complex procedures and spend a lot of time writing work orders, formulating maintenance plans, and estimating the construction period. This disconnect between perception and decision-making leads to a lag in maintenance response speed, and the generated operating procedures often lack standardization due to differences in personnel quality.

[0065] To address the problems of the prior art, this invention provides a method for the safety maintenance of power transmission channels. This method can be applied to various computer terminals or smart terminals, and its execution entity can be the processor or server of the computer terminal or smart terminal. The flowchart of the method is shown below. Figure 1 As shown, it specifically includes:

[0066] S1: Acquire the three-dimensional point cloud data of the power transmission channel.

[0067] High-precision 3D point cloud data can be obtained by scanning the target power transmission channel using equipment such as drone LiDAR or ground-based 3D laser scanners. This data contains the 3D spatial coordinate information of all ground features (such as towers, power lines, vegetation, and ground) within the power transmission channel.

[0068] LiDAR (Light Detection and Ranging) calculates distance by emitting laser pulses and receiving echoes, based on the Time-of-Flight (ToF) method. It can output standard .las or .pcd files containing the spatial coordinates of the point and the reflection intensity.

[0069] S2: Divide the three-dimensional point cloud data into point cloud types to determine ground points, power line points, tower points, and vegetation points.

[0070] The purpose of point cloud type classification is to distinguish the chaotic point cloud data according to the land cover category. This can be achieved using various classification algorithms, such as machine learning-based segmentation algorithms and rule-based hierarchical classification methods.

[0071] By classifying, point cloud data is divided into four basic types: ground points, power line points, tower points, and vegetation points. This lays the foundation for subsequent feature extraction for different land features. Furthermore, the quality of the classification results directly affects the accuracy of subsequent risk assessments, so detailed and accurate extraction of point cloud features is necessary.

[0072] The point cloud was divided into the four key types mentioned above, covering the most critical elements in the safety maintenance of power transmission channels: conductors (power line points), supporting structures (tower points), potential intruders (vegetation points), and basic environment (ground points), laying the foundation for subsequent targeted analysis.

[0073] S3: Extract and integrate the features of the ground points, power line points, tower points and vegetation points respectively to obtain a point cloud statistical feature set.

[0074] This step involves extracting key feature parameters that reflect the safety status of different types of ground features. Compared to extracting directly from the entire 3D point cloud data, this method can extract more detailed features.

[0075] For example, for power line points, features such as sag, stress, and span can be extracted; for vegetation points, features such as vegetation height, volume, and clearance distance from the conductor can be extracted; for tower points, features such as tilt and foundation settlement can be extracted; and for ground points, features such as terrain slope and elevation change can be extracted.

[0076] The extracted features are then normalized and integrated to form a multi-dimensional point cloud statistical feature set, which serves as the benchmark data for subsequent risk assessment.

[0077] S4: Match the point cloud statistical feature set with a preset procedure knowledge base to determine the operating status and potential risk level of each section of the power transmission channel.

[0078] The regulations knowledge base is a pre-established set of rules that integrates power industry standards, safety regulations, and expert experience. This knowledge base includes various safety thresholds, risk assessment rules, and corresponding risk level classification standards.

[0079] The point cloud statistical feature set obtained in step S3 is compared with the rules in the procedure knowledge base to determine whether there are safety hazards in each section of the power transmission channel. This determines the current objective physical state of each section as the operating state. For example, whether the conductor sag is too large, whether the vegetation has encroached on the safety distance, and whether the tower has abnormal tilt are all operating states. Based on the operating state, the potential risk level can be determined again by combining the risk level classification standard in the procedure knowledge base. Therefore, the determination of the potential risk level requires the assistance of the operating state. The potential risk level can include low risk, medium risk, high risk, etc.

[0080] S5: Invoke the preset set of handling strategies, generate operating procedures and strategies based on the operating status and potential risk level of each section, and apply them to the power transmission channel.

[0081] The set of response strategies includes predefined response plans for different operating states and different potential risk levels. In other words, different operating states correspond to different response plans, and different potential risk levels correspond to different response plans. Therefore, merging the response plans corresponding to the operating states and potential risk levels constitutes the operating procedure strategy.

[0082] For example, for sections with a low potential risk level and an operating status where the clearance distance is close to the threshold, a strategy of regular monitoring and clearance distance retesting can be generated to prompt maintenance personnel to pay attention to the changing trend of the section; for sections with a medium potential risk level and an operating status where vegetation is growing rapidly, a strategy of planned pruning and growth cycle tracking can be generated to arrange on-site verification or vegetation clearing; for sections with a high potential risk level and an operating status where conductor sag is severely excessive, an emergency maintenance and temporary load limiting strategy can be generated immediately and distributed to maintenance personnel or relevant automated systems.

[0083] By applying the generated operating procedures and strategies to the transmission channels, such as notifying maintenance personnel, triggering early warning systems, or directly linking with the inspection system, closed-loop safety maintenance of the transmission channels can be achieved. At the same time, this application ensures that the generated operating procedures and strategies not only meet the rigor of industry standards but also have practical operability for on-site operations, which can significantly improve the safety of power grid operation and maintenance.

[0084] A template engine can be used to create output files corresponding to the operating procedures and strategies. For example, a "Tree Obstacle Handling Work" template in .docx format can be created in advance, with placeholders such as {{Tower_ID}} and {{Action_Type}} reserved in it. The operating status and potential risk level of each section output in step S4 can be automatically filled into the placeholders and saved as the final procedure file for power transmission channels.

[0085] As can be seen from the above technical solution, this application obtains three-dimensional point cloud data of the power transmission channel; classifies the three-dimensional point cloud data into point cloud types to determine ground points, power line points, tower points, and vegetation points; extracts and integrates features from the ground points, power line points, tower points, and vegetation points respectively to obtain a point cloud statistical feature set; matches the point cloud statistical feature set with a preset procedure knowledge base to determine the operating status and potential risk level of each section of the power transmission channel; and calls a preset set of disposal strategies to generate operating procedure strategies based on the operating status and potential risk level of each section, and applies them to the power transmission channel. This application first acquires three-dimensional point cloud data of the power transmission channel. However, the original three-dimensional point cloud data only contains spatial coordinate information and lacks semantic meaning (i.e., what object each point represents). This application classifies the point cloud data by type, assigning category labels such as ground, power lines, towers, and vegetation, thus realizing the transformation from data to information. After separating the overall point cloud by category, subsequent feature extraction can use different algorithms for different types, avoiding the accuracy loss caused by a one-size-fits-all approach and improving the accuracy and efficiency of the analysis. Therefore, in subsequent feature extraction, key feature parameters that best reflect the safety status of different types of ground features can be extracted for subsequent risk assessment. During the analysis and matching process, a regulatory knowledge base is used for matching. This knowledge base integrates power industry standards, safety regulations, and expert experience, digitizing and structuring the rules scattered across various regulatory documents. This allows for accurate analysis of the operating status and potential risk levels of each section of the transmission channel, automatically determining whether there are safety hazards in each section and quantifying their risk levels. Finally, based on the operating status and potential risk levels, corresponding handling strategies are automatically matched and generated, achieving an automated closed loop from problem discovery to solution provision. Compared to traditional manual inspections, this significantly improves operation and maintenance efficiency and accuracy, reduces reliance on human experience, and ensures the safety of the transmission channel.

[0086] The method provided in this embodiment of the invention involves classifying the three-dimensional point cloud data into point cloud types to determine ground points, power line points, tower points, and vegetation points. The specific details are as follows:

[0087] The three-dimensional point cloud data is flipped along the Z-axis in a preset spatial rectangular coordinate system to obtain the first point cloud data;

[0088] A virtual grid is generated and overlaid on the first point cloud data;

[0089] The virtual mesh is sunk in the opposite direction of the Z-axis. During the sunk process, when the virtual mesh collides with the first point cloud data, the virtual mesh is fixed at the position corresponding to the collision in the virtual mesh to determine each collision point.

[0090] The collision point is taken as the ground point, and the other points in the three-dimensional point cloud data other than the ground points are taken as non-ground points.

[0091] Determine the power line points, tower points, and vegetation points from each of the aforementioned non-ground points.

[0092] Specifically, to facilitate the extraction of ground points, the original point cloud data is flipped along the Z-axis (elevation axis), transforming the ground point cloud data, originally at the bottom, into a ceiling-like shape at the top of the point cloud data. This facilitates subsequent mesh settling processing. The virtual mesh can be understood as a flexible, deformable cloth simulation mesh. The size of this mesh should be able to completely cover the target point cloud area. The initial position of the mesh is at the top of the flipped point cloud. Then, the virtual mesh is controlled to gradually settle downwards (in the opposite direction of the Z-axis) from the initial position. When a node of the mesh encounters (collides) a point in the first point cloud data, the node stops settling and is fixed (fixed within the virtual mesh). By simulating this physical process of cloth settling, the virtual mesh will eventually conform to the surface of the flipped point cloud, and these fixed points are the collision points. Since the original ground points are at the top in the flipped point cloud, the collision points actually correspond to the ground points in the original point cloud.

[0093] This embodiment uses this method to accurately separate ground points from non-ground points such as buildings, vegetation, and power lines, achieving efficient and accurate ground point cloud extraction.

[0094] The steps for determining power line points, tower points, and vegetation points from the various non-ground points in the above process are explained in detail below.

[0095] Calculate the average elevation of the ground points;

[0096] The target height is obtained by adding the average elevation to the preset vegetation starting height threshold.

[0097] Obtain the elevation values ​​of each of the non-ground points, and designate the non-ground points with elevation values ​​greater than the target height as vegetation points;

[0098] Calculate the linearity of each non-ground point other than the vegetation point;

[0099] Points with linearity greater than a preset linearity threshold are designated as power line points, and points with linearity less than the linearity threshold are designated as tower points.

[0100] Specifically, the average elevation of the ground points represents the ground reference height of the area where the power transmission channel is located. The average elevation can be obtained by statistically analyzing the elevation values ​​of all the determined ground points and calculating their average value.

[0101] The preset vegetation starting height threshold can be set based on actual experience, such as 2 meters, to distinguish low vegetation or ground debris from higher vegetation that poses a potential risk. The average elevation is added to the vegetation starting height threshold to obtain the target height, which serves as a reference benchmark for distinguishing vegetation from other ground features. Then, all non-ground points are traversed, and the elevation value of each point is extracted. If the elevation value of a non-ground point is greater than the target height, the point is classified as a vegetation point. This is because vegetation that may pose a threat to the power transmission line in the transmission channel usually has a relatively obvious height.

[0102] For the remaining point cloud that is not classified as vegetation point among non-ground points, it is necessary to distinguish whether it is a power line point or a tower point. This involves calculating the linearity feature of each remaining point cloud that is not classified as vegetation point among non-ground points. Linearity is a feature quantity that describes the local geometry of the point cloud, and it can be obtained by calculating the eigenvalues ​​of the covariance matrix of the neighboring point set of the point.

[0103] Since power lines are linearly distributed in space and have high linearity, points with linearity greater than a preset linearity threshold are designated as power line points. While towers are also columnar, their tops and crossarms have planar or scattered features, resulting in relatively low linearity. Therefore, points with linearity less than the linearity threshold can be designated as tower points. This application achieves accurate classification of vegetation points, power line points, and tower points by combining elevation thresholds and linearity features.

[0104] The following embodiments provide a detailed explanation of the process of feature extraction of power line points in this application.

[0105] For each electric field line point, an initial subset is randomly selected, and a catenary equation is fitted to the initial subset.

[0106] Solve the catenary equation to obtain the current stress;

[0107] Obtain the current working condition, substitute the current working condition and current stress into the preset conductor state equation and solve it to obtain the future stress;

[0108] The sag characteristics are calculated based on the future stress.

[0109] Specifically, the shape of the power line between suspension points conforms to the catenary model, and the model formula is:

[0110] ;

[0111] in, Represents the horizontal coordinate. Indicates correspondence The vertical coordinate (i.e., height) of the power line at that location. For the shape parameters of the catenary, , This represents the horizontal position offset of the lowest point, when hour, Take the minimum value, that is, the lowest point of the power line. It represents the vertical translation amount, which determines the overall height by which the catenary is raised or lowered in the longitudinal direction.

[0112] From the power line point cloud obtained through classification, an initial subset can be randomly selected using methods such as the Random Sampling Consensus (RANSAC) algorithm to fit the catenary equation, thereby eliminating the interference of noise points and outliers and obtaining accurate catenary model parameters. Based on the fitted catenary equation and combined with the specific load (the gravity acting on a unit length of conductor), the current stress of the conductor can be calculated. This stress reflects the stress state of the conductor under the current operating conditions. Then, the current operating conditions are obtained, and the conductor state equation describes the relationship between the stress, temperature, and specific load of the conductor under different operating conditions.

[0113] By substituting the current operating conditions, current stress, and predicted future operating conditions into the preset conductor state equation, the conductor stress under the future operating conditions, i.e., the future stress, can be calculated. Here, the operating conditions can refer to current temperature, wind speed, icing conditions, etc. The conductor state equation is:

[0114] ;

[0115] in, Indicates future stress under future operating conditions. This indicates the current stress under the current operating conditions. , These represent the conductor specific load under future operating conditions and under suppressed operating conditions, respectively. For gear distance, This represents the elastic modulus of the conductor. This represents the coefficient of linear expansion of the conductor. , These represent the temperatures under future operating conditions and the current operating conditions, respectively.

[0116] Sag characteristic can be a specific numerical value or a deviation rate from the design sag. Sag refers to the vertical distance between the line connecting the conductor suspension points and the lowest point of the conductor. It is a key indicator for the safe operation of transmission lines. Excessive sag can lead to insufficient safe distances to the ground or objects being crossed, while insufficient sag can result in excessive conductor tension, posing a risk of conductor breakage. Based on the calculated future stress, the conductor sag value under future operating conditions can be calculated, for example, using the parabolic sag formula.

[0117] ;

[0118] in, This indicates the sag value of the conductor.

[0119] The following embodiments provide a detailed explanation of the process of feature extraction for vegetation points in this application.

[0120] The three-dimensional point cloud data is spatially voxelized into a mesh according to a preset voxel size to obtain individual voxel meshes.

[0121] The voxel grid containing the vegetation points is determined as each vegetation grid, and the number of vegetation grids and the number of voxels in each vegetation grid are counted.

[0122] The voxel density of each vegetation grid is calculated based on the number of voxels and the voxel size.

[0123] The total vegetation volume is calculated based on the vegetation grid and voxel dimensions.

[0124] Specifically, a voxel size is set (e.g., 0.5m × 0.5m × 0.5m), and the three-dimensional space of the entire power transmission channel is divided into several regular three-dimensional voxel grids. Each voxel grid is a small cube used to count the point cloud information contained within it. All voxel grids are traversed to determine whether each voxel grid contains vegetation points.

[0125] The voxel grid containing vegetation points is labeled as a vegetation grid. The total number of all vegetation grids is counted, and the number of voxels (i.e., 3D points) contained in each vegetation grid is recorded. For each vegetation grid, its voxel density can be calculated by dividing the number of points in the grid by the volume of the voxel grid. Voxel density reflects the density of vegetation in the area; higher density indicates denser vegetation. The total vegetation volume can be obtained by summing the volumes of all vegetation grids, and the volume of each vegetation grid is the cube of the voxel size.

[0126] By combining the total volume of vegetation with its growth trend, it is possible to assess whether it has encroached on the safe distance of the line. This application, through voxelization processing, can quantify the voxel density and total volume of vegetation, providing a data basis for subsequent risk detection and safety maintenance processes. It can also estimate the biomass that subsequent workers need to clear based on voxel density and total vegetation volume, generate a list of clearing workload, and set the type of clearing tools and estimated working hours in the operation procedure strategy.

[0127] Furthermore, in addition to feature extraction for different point cloud types, features can also be extracted from a global perspective of 3D point cloud data, which may include the following processes:

[0128] For each point in the three-dimensional point cloud data, determine the neighborhood point set of that point, and calculate the covariance matrix of the neighborhood point set;

[0129] Singular value decomposition is performed on the covariance matrix to obtain the first eigenvalue, second eigenvalue, and third eigenvalue;

[0130] The linearity feature is calculated using the first and second eigenvalues.

[0131] Flatness features are calculated using the first, second, and third eigenvalues;

[0132] The scattering characteristics are calculated using the first and third eigenvalues.

[0133] Specifically, for each point p in the point cloud, its neighborhood point set can be determined based on methods such as K-nearest neighbors (KNN) or radius search, and the three-dimensional covariance matrix of the neighborhood point set can be calculated. The three-dimensional covariance matrix is ​​a 3×3 symmetric matrix that can reflect the distribution of the neighborhood point set in space.

[0134] Singular value decomposition (SVD) of the covariance matrix yields three eigenvalues ​​λ1, λ2, and λ3, satisfying λ1 ≥ λ2 ≥ λ3. These three eigenvalues ​​describe the dispersion of the neighborhood point set along the three principal directions.

[0135] Linearity features are used to describe the degree to which a point cloud exhibits a local linear distribution. The calculation formula is as follows: The larger the linearity value, the closer the point cloud distribution is to a straight line.

[0136] Flatness is a feature used to describe the degree to which a point cloud locally exhibits a planar distribution. Its calculation formula is: Flatness The larger the flatness value, the closer the point cloud distribution is to a plane.

[0137] Scattering intensity is a feature used to describe the degree to which a point cloud exhibits a scattered distribution in a localized area. Its calculation formula is: Scattering Intensity The larger the scattering value, the more scattered the point cloud distribution is, and the less obvious the geometric structure.

[0138] These features can effectively describe the local geometry of point clouds, which is crucial for further, more detailed structural discrimination of classified point clouds in power transmission channels.

[0139] Furthermore, this application can also acquire historical point cloud data of the transmission channel at a preset historical moment, analyze the height change of voxels at the same location between the historical point cloud data and the current three-dimensional point cloud data acquired in step S1 of this application (the M3C2 algorithm or the Cloud-to-Cloud distance calculation method can be used), and use a linear extrapolation formula to predict growth, thereby further predicting the subsequent vegetation growth trend. The linear extrapolation formula is as follows:

[0140] ;

[0141] in, This indicates the predicted vegetation height at a future point in time. Indicates the current vegetation height. This represents the height difference between the future vegetation height and the current vegetation height. This represents the time difference between a future point in time and the present point in time. It indicates a future point in time.

[0142] The probability of vegetation touching the power transmission line at a future time can be predicted using the linear extrapolation formula. If the probability of contact is greater than the preset probability threshold and the time difference is short, a corresponding handling strategy needs to be added to the operation procedure strategy. If the probability of contact is less than the probability threshold or the time difference is long, an operation procedure strategy can still be generated, but it needs to be marked as an "annual pruning order" to indicate that no action is needed at present, but it needs to be completed within the year.

[0143] Furthermore, to mitigate risks more meticulously, this application can also verify whether a sufficient safe distance is maintained between the conductor and vegetation after swaying under extreme wind conditions. Specifically, this involves calculating the conductor's wind deflection angle, rotating the conductor's point cloud coordinates around the suspension point according to the deflection angle, and then calculating the distance between the conductor and the vegetation point. The calculation formula is:

[0144] ;

[0145] in, This represents the horizontal thrust exerted by wind on the conductor, expressed in N / m or kN. This represents the weight of the conductor itself.

[0146] In the embodiments provided in this application, the process of establishing the procedure knowledge base includes:

[0147] Determine the voltage levels included in the power system, set different minimum vertical safety distances and minimum horizontal safety distances according to different voltage levels, and establish a voltage level rule table based on each voltage level, minimum vertical safety distance, and minimum horizontal safety distance;

[0148] Determine the slope of each terrain where all existing transmission channels are located, set different operational constraints based on different terrain slopes, and establish an operational constraint rule table;

[0149] Multiple temperature ranges are established, and a safety distance correction factor is set for each of the temperature ranges to establish a meteorological correction rule table;

[0150] Multiple distance ranges are established for the distance between vegetation and power lines, and a risk level is set for each distance range to create a vegetation risk level rule table;

[0151] Multiple sag ranges are established for the sag status of power lines, and a warning level is set for each sag range to establish a sag warning level rule table;

[0152] The voltage level rule table, operation constraint rule table, meteorological correction rule table, vegetation risk level rule table, and sag warning level rule table are combined to obtain the procedure knowledge base.

[0153] Specifically, according to power industry standards, the minimum safe distance requirements for transmission lines of different voltage levels differ regarding conductors to the ground, buildings, trees, etc. Therefore, a voltage level rule table can be established, which meticulously defines the minimum vertical and horizontal safe distances for each voltage level. Terrain slope is a crucial factor affecting the safety of operation and maintenance. For example, in mountainous areas with steep slopes, the difficulty and risk of manual or drone inspections increase, and the access conditions for operations become more stringent. Therefore, an operation constraint rule table can be established, setting operation constraints for different slope ranges, such as whether special protection is required or whether specific types of equipment are allowed to enter. Weather conditions affect conductor sag and wind deflection. For instance, conductor sag increases in hot weather, requiring a corresponding increase in safe distance. Therefore, a weather correction rule table can be established, setting safety distance correction coefficients for different temperature ranges (or different weather combinations) to dynamically adjust the safety distance threshold. Based on the clearance distance between vegetation and the conductor, multiple risk levels can be defined. For example, a distance greater than 1.2 times the safe distance is considered low risk; a distance between 1.0 and 1.2 times the safe distance is considered medium risk; and a distance less than the safe distance is considered high risk. This allows for the establishment of a vegetation risk level rule table, which can be used to directly determine the risk level based on measured distance. Based on the conductor sag value or its deviation rate from the design sag, multiple warning levels can be defined. For example, sag within the normal range is considered normal; sag exceeding the design value by 15% is considered a warning; exceeding 30% is a warning; and exceeding 50% is a severe warning. This allows for the establishment of a sag warning level rule table, which can be used to determine the warning level based on sag characteristics.

[0154] The rules can be stored in a relational database (such as MySQL or SQLite) to form rule tables, and SQL query statements or Python logical judgment statements can be written for matching.

[0155] By integrating the above rule tables, a multi-dimensional and multi-level procedural knowledge base can be formed. During risk assessment, rules from multiple dimensions can be combined for comprehensive judgment, thereby improving the comprehensiveness and accuracy of risk assessment.

[0156] Further, in determining the potential risk level, it can be based on the vegetation risk level rule table and / or the sag warning level rule table. Additionally, based on the point cloud measurement error distribution and environmental uncertainties, Monte Carlo simulation can be used to calculate the probability density function of the fault occurrence. The calculated fault probability is mapped to a visual cycle suggestion. For high-risk areas, instructions to shorten the inspection cycle are automatically generated and added to the operation procedure strategy. For Monte Carlo simulation, the measurement error of the lidar can be set to follow a normal distribution N(0, 0.05m). 10,000 sets of point cloud coordinate samples with random errors can be constructed. For each set of samples, the distance between the guide wire and the trees is calculated, and the proportion of samples with distances less than the safety threshold is the fault probability.

[0157] Furthermore, this application can also generate a three-dimensional visualization work instruction book based on the operation procedure strategy. This instruction book can highlight the target to be processed in the three-dimensional point cloud scene of the power transmission channel, and automatically plan and display the optimal access path and work station based on the statistical characteristics of the terrain point cloud.

[0158] and Figure 1 Corresponding to the method described above, embodiments of the present invention also provide a safety maintenance device for power transmission channels, used for... Figure 1 In a specific implementation of the method, the power transmission channel safety maintenance device provided in this embodiment of the invention can be integrated into a computer terminal or various mobile devices. Figure 2 The safety maintenance devices for power transmission channels are introduced, such as... Figure 2 As shown, the device may include:

[0159] The 3D point cloud data acquisition module 10 is used to acquire the 3D point cloud data of the power transmission channel;

[0160] The type classification module 20 is used to classify the three-dimensional point cloud data into point cloud types to determine ground points, power line points, tower points and vegetation points;

[0161] Feature extraction module 30 is used to extract and integrate features from the ground points, power line points, tower points and vegetation points respectively to obtain a point cloud statistical feature set;

[0162] The matching module 40 is used to match the point cloud statistical feature set with a preset procedure knowledge base to determine the operating status and potential risk level of each section of the power transmission channel.

[0163] The operation procedure strategy generation module 50 is used to call a preset set of disposal strategies, generate operation procedure strategies based on the operation status and potential risk level of each section, and apply them to the power transmission channel.

[0164] As can be seen from the above technical solution, this application obtains three-dimensional point cloud data of the power transmission channel; classifies the three-dimensional point cloud data into point cloud types to determine ground points, power line points, tower points, and vegetation points; extracts and integrates features from the ground points, power line points, tower points, and vegetation points respectively to obtain a point cloud statistical feature set; matches the point cloud statistical feature set with a preset procedure knowledge base to determine the operating status and potential risk level of each section of the power transmission channel; and calls a preset set of disposal strategies to generate operating procedure strategies based on the operating status and potential risk level of each section, and applies them to the power transmission channel. This application first acquires three-dimensional point cloud data of the power transmission channel. However, the original three-dimensional point cloud data only contains spatial coordinate information and lacks semantic meaning (i.e., what object each point represents). This application classifies the point cloud data by type, assigning category labels such as ground, power lines, towers, and vegetation, thus realizing the transformation from data to information. After separating the overall point cloud by category, subsequent feature extraction can use different algorithms for different types, avoiding the accuracy loss caused by a one-size-fits-all approach and improving the accuracy and efficiency of the analysis. Therefore, in subsequent feature extraction, key feature parameters that best reflect the safety status of different types of ground features can be extracted for subsequent risk assessment. During the analysis and matching process, a regulatory knowledge base is used for matching. This knowledge base integrates power industry standards, safety regulations, and expert experience, digitizing and structuring the rules scattered across various regulatory documents. This allows for accurate analysis of the operating status and potential risk levels of each section of the transmission channel, automatically determining whether there are safety hazards in each section and quantifying their risk levels. Finally, based on the operating status and potential risk levels, corresponding handling strategies are automatically matched and generated, achieving an automated closed loop from problem discovery to solution provision. Compared to traditional manual inspections, this significantly improves operation and maintenance efficiency and accuracy, reduces reliance on human experience, and ensures the safety of the transmission channel.

[0165] In one example, type partitioning module 20 includes:

[0166] The flipping unit is used to flip the three-dimensional point cloud data along the Z-axis in a preset spatial rectangular coordinate system to obtain the first point cloud data;

[0167] An overlay unit is used to generate a virtual grid and overlay the virtual grid on top of the first point cloud data;

[0168] The collision point determination unit is used to sink the virtual mesh in the opposite direction of the Z-axis. During the sinking process, when the virtual mesh collides with the first point cloud data, the virtual mesh is fixed at the position corresponding to the collision in the virtual mesh to determine each collision point.

[0169] The ground point and non-ground point determination unit is used to treat the collision point as a ground point and other points in the three-dimensional point cloud data other than the ground points as non-ground points.

[0170] The point determination unit is used to determine power line points, tower points, and vegetation points from the various non-ground points.

[0171] In one example, the point determination unit includes:

[0172] The average elevation calculation block is used to calculate the average elevation of the ground points.

[0173] The summation block is used to add the average elevation to a preset vegetation starting height threshold to obtain the target height;

[0174] A vegetation point determination block is used to obtain the elevation values ​​of each of the non-ground points, and to designate non-ground points with elevation values ​​greater than the target height as vegetation points.

[0175] A linearity calculation block is used to calculate the linearity of each non-ground point other than the vegetation point.

[0176] The tower point determination block is used to designate points with linearity greater than a preset linear threshold as power line points and points with linearity less than the linear threshold as tower points.

[0177] In one example, feature extraction module 30 includes:

[0178] An equation fitting unit is used to randomly select an initial subset from the electric field points and fit a catenary equation to the initial subset.

[0179] The current stress calculation unit is used to solve the catenary equation to obtain the current stress;

[0180] The future stress calculation unit is used to obtain the current working condition, substitute the current working condition and the current stress into the preset conductor state equation and solve it to obtain the future stress.

[0181] A sag feature calculation unit is used to calculate the sag feature based on the future stress.

[0182] In one example, feature extraction module 30 includes:

[0183] A voxel mesh determination unit is used to perform spatial voxelization meshing on the three-dimensional point cloud data according to a preset voxel size to obtain each voxel mesh.

[0184] A statistical unit is used to determine the voxel grid containing the vegetation points as each vegetation grid, and to count the number of vegetation grids and the number of voxels in each vegetation grid.

[0185] A voxel density calculation unit is used to calculate the voxel density of each vegetation grid based on the number of voxels and the voxel size.

[0186] The vegetation total volume calculation unit is used to calculate the total vegetation volume based on the vegetation grid and voxel size.

[0187] In one example, the device also includes a procedure knowledge base building module, which includes:

[0188] The voltage level rule table establishment unit is used to determine the voltage levels included in the power system, set different minimum vertical safety distances and minimum horizontal safety distances according to different voltage levels, and establish a voltage level rule table according to each voltage level, minimum vertical safety distance, and minimum horizontal safety distance.

[0189] The task constraint rule table establishment unit is used to determine the slope of each terrain where all existing transmission channels are located, set different task constraints according to different terrain slopes, and establish a task constraint rule table.

[0190] The meteorological correction rule table establishment unit is used to establish multiple temperature ranges and set a safety distance correction coefficient for each of the temperature ranges to establish a meteorological correction rule table.

[0191] The vegetation risk level rule table establishment unit is used to establish multiple distance ranges for the distance between vegetation and power lines, and to set a risk level for each distance range in order to establish a vegetation risk level rule table.

[0192] The sag warning level rule table establishment unit is used to establish multiple sag ranges for the sag status of power lines and set a warning level for each sag range to establish a sag warning level rule table.

[0193] The procedure knowledge base determination unit is used to combine the voltage level rule table, operation constraint rule table, meteorological correction rule table, vegetation risk level rule table and sag warning level rule table to obtain the procedure knowledge base.

[0194] In one example, the device also includes a global feature extraction module, which includes:

[0195] The covariance matrix calculation unit is used to determine the neighborhood point set of each point in the three-dimensional point cloud data and calculate the covariance matrix of the neighborhood point set.

[0196] The singular value decomposition unit is used to perform singular value decomposition on the covariance matrix to obtain each first eigenvalue, second eigenvalue, and third eigenvalue.

[0197] A linearity feature calculation unit is used to calculate the linearity feature using the first feature value and the second feature value;

[0198] A flatness feature calculation unit is used to calculate flatness features using the first feature value, the second feature value, and the third feature value;

[0199] The scattering characteristic calculation unit is used to calculate the scattering characteristic using the first characteristic value and the third characteristic value.

[0200] Each module in the aforementioned safety maintenance device for power transmission channels 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, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.

[0201] Furthermore, embodiments of this application provide a safety maintenance device for power transmission channels. Optionally, Figure 3 The hardware structure block diagram of the safety maintenance equipment for power transmission channels is shown. (Refer to...) Figure 3 The hardware structure of the safety maintenance equipment for power transmission channels may include: at least one processor 01, at least one communication interface 02, at least one memory 03 and at least one communication bus 04.

[0202] In this embodiment, the number of processor 01, communication interface 02, memory 03 and communication bus 04 is at least one, and processor 01, communication interface 02 and memory 03 communicate with each other through communication bus 04.

[0203] Processor 01 may be a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement embodiments of the present invention.

[0204] Memory 03 may include high-speed RAM, and may also include non-volatile memory, such as at least one disk storage device.

[0205] The memory stores a program that the processor can call. The program is used to execute the following safety maintenance methods for the power transmission channel, including:

[0206] Acquire 3D point cloud data of the power transmission channel;

[0207] The three-dimensional point cloud data is divided into point cloud types to determine ground points, power line points, tower points, and vegetation points;

[0208] Feature extraction and integration are performed on the ground points, power line points, tower points, and vegetation points respectively to obtain a point cloud statistical feature set;

[0209] The point cloud statistical feature set is matched with a preset procedure knowledge base to determine the operating status and potential risk level of each section of the power transmission channel.

[0210] The system invokes a set of preset handling strategies, generates operating procedures and strategies based on the operating status and potential risk level of each section, and applies them to the power transmission channel.

[0211] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0212] The three-dimensional point cloud data is flipped along the Z-axis in a preset spatial rectangular coordinate system to obtain the first point cloud data;

[0213] A virtual grid is generated and overlaid on the first point cloud data;

[0214] The virtual mesh is sunk in the opposite direction of the Z-axis. During the sunk process, when the virtual mesh collides with the first point cloud data, the virtual mesh is fixed at the position corresponding to the collision in the virtual mesh to determine each collision point.

[0215] The collision point is taken as the ground point, and the other points in the three-dimensional point cloud data other than the ground points are taken as non-ground points.

[0216] Determine the power line points, tower points, and vegetation points from each of the aforementioned non-ground points.

[0217] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0218] Calculate the average elevation of the ground points;

[0219] The target height is obtained by adding the average elevation to the preset vegetation starting height threshold.

[0220] Obtain the elevation values ​​of each of the non-ground points, and designate the non-ground points with elevation values ​​greater than the target height as vegetation points;

[0221] Calculate the linearity of each non-ground point other than the vegetation point;

[0222] Points with linearity greater than a preset linearity threshold are designated as power line points, and points with linearity less than the linearity threshold are designated as tower points.

[0223] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0224] For each electric field line point, an initial subset is randomly selected, and a catenary equation is fitted to the initial subset.

[0225] Solve the catenary equation to obtain the current stress;

[0226] Obtain the current working condition, substitute the current working condition and current stress into the preset conductor state equation and solve it to obtain the future stress;

[0227] The sag characteristics are calculated based on the future stress.

[0228] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0229] The three-dimensional point cloud data is spatially voxelized into a mesh according to a preset voxel size to obtain individual voxel meshes.

[0230] The voxel grid containing the vegetation points is determined as each vegetation grid, and the number of vegetation grids and the number of voxels in each vegetation grid are counted.

[0231] The voxel density of each vegetation grid is calculated based on the number of voxels and the voxel size.

[0232] The total vegetation volume is calculated based on the vegetation grid and voxel dimensions.

[0233] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0234] Determine the voltage levels included in the power system, set different minimum vertical safety distances and minimum horizontal safety distances according to different voltage levels, and establish a voltage level rule table based on each voltage level, minimum vertical safety distance, and minimum horizontal safety distance;

[0235] Determine the slope of each terrain where all existing transmission channels are located, set different operational constraints based on different terrain slopes, and establish an operational constraint rule table;

[0236] Multiple temperature ranges are established, and a safety distance correction factor is set for each of the temperature ranges to establish a meteorological correction rule table;

[0237] Multiple distance ranges are established for the distance between vegetation and power lines, and a risk level is set for each distance range to create a vegetation risk level rule table;

[0238] Multiple sag ranges are established for the sag status of power lines, and a warning level is set for each sag range to establish a sag warning level rule table;

[0239] The voltage level rule table, operation constraint rule table, meteorological correction rule table, vegetation risk level rule table, and sag warning level rule table are combined to obtain the procedure knowledge base.

[0240] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0241] For each point in the three-dimensional point cloud data, determine the neighborhood point set of that point, and calculate the covariance matrix of the neighborhood point set;

[0242] Singular value decomposition is performed on the covariance matrix to obtain the first eigenvalue, second eigenvalue, and third eigenvalue;

[0243] The linearity feature is calculated using the first and second eigenvalues.

[0244] Flatness features are calculated using the first, second, and third eigenvalues;

[0245] The scattering characteristics are calculated using the first and third eigenvalues.

[0246] This application embodiment also provides a storage medium that can store a program suitable for execution by a processor. When the program runs, it controls the device containing the storage medium to perform the following power transmission channel safety maintenance method, including:

[0247] Acquire 3D point cloud data of the power transmission channel;

[0248] The three-dimensional point cloud data is divided into point cloud types to determine ground points, power line points, tower points, and vegetation points;

[0249] Feature extraction and integration are performed on the ground points, power line points, tower points, and vegetation points respectively to obtain a point cloud statistical feature set;

[0250] The point cloud statistical feature set is matched with a preset procedure knowledge base to determine the operating status and potential risk level of each section of the power transmission channel.

[0251] The system invokes a set of preset handling strategies, generates operating procedures and strategies based on the operating status and potential risk level of each section, and applies them to the power transmission channel.

[0252] Specifically, the storage medium can be a computer-readable storage medium, which can be an electronic storage device such as flash memory, EEPROM (Electrically Erasable Programmable Read-Only Memory), EPROM, hard disk, or ROM.

[0253] Optionally, the refined and extended functions of the program can be found in the description of the power transmission channel safety maintenance method in the method embodiments.

[0254] Furthermore, the functional modules in the various embodiments of this disclosure can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part. If the function is implemented as a software functional module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this disclosure, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, a live streaming device, or a network device, etc.) to execute all or part of the steps of the methods in the various embodiments of this disclosure.

[0255] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0256] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0257] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for security maintenance of a power transmission corridor, characterized in that, include: Acquire 3D point cloud data of the power transmission channel; The three-dimensional point cloud data is divided into point cloud types to determine ground points, power line points, tower points, and vegetation points; Feature extraction and integration are performed on the ground points, power line points, tower points, and vegetation points respectively to obtain a point cloud statistical feature set; The point cloud statistical feature set is matched with a preset procedure knowledge base to determine the operating status and potential risk level of each section of the power transmission channel. The system invokes a set of preset handling strategies, generates operating procedures and strategies based on the operating status and potential risk level of each section, and applies them to the power transmission channel.

2. The method of claim 1, wherein, The step of classifying the three-dimensional point cloud data into point cloud types to determine ground points, power line points, tower points, and vegetation points includes: The three-dimensional point cloud data is flipped along the Z-axis in a preset spatial rectangular coordinate system to obtain the first point cloud data; A virtual grid is generated and overlaid on the first point cloud data; The virtual mesh is sunk in the opposite direction of the Z-axis. During the sunk process, when the virtual mesh collides with the first point cloud data, the virtual mesh is fixed at the position corresponding to the collision in the virtual mesh to determine each collision point. The collision point is taken as the ground point, and the other points in the three-dimensional point cloud data other than the ground points are taken as non-ground points. Determine the power line points, tower points, and vegetation points from each of the aforementioned non-ground points.

3. The method of claim 2, wherein, The process of determining power line points, tower points, and vegetation points from each of the aforementioned non-ground points includes: Calculate the average elevation of the ground points; The target height is obtained by adding the average elevation to the preset vegetation starting height threshold. Obtain the elevation values ​​of each of the non-ground points, and designate the non-ground points with elevation values ​​greater than the target height as vegetation points; Calculate the linearity of each non-ground point other than the vegetation point; Points with linearity greater than a preset linearity threshold are designated as power line points, and points with linearity less than the linearity threshold are designated as tower points.

4. The method of claim 1, wherein, The process of feature extraction from electric field line points includes: For each electric field line point, an initial subset is randomly selected, and a catenary equation is fitted to the initial subset. Solve the catenary equation to obtain the current stress; Obtain the current working condition, substitute the current working condition and current stress into the preset conductor state equation and solve it to obtain the future stress; The sag characteristics are calculated based on the future stress.

5. The method of claim 1, wherein, The process of feature extraction from vegetation points includes: The three-dimensional point cloud data is spatially voxelized into a mesh according to a preset voxel size to obtain individual voxel meshes. The voxel grid containing the vegetation points is identified as each vegetation grid, and the number of vegetation grids and the number of voxels in each vegetation grid are counted. The voxel density of each vegetation grid is calculated based on the number of voxels and the voxel size. The total vegetation volume is calculated based on the vegetation grid and voxel dimensions.

6. The method of claim 1, wherein, The process of establishing the procedure knowledge base includes: Determine the voltage levels included in the power system, set different minimum vertical safety distances and minimum horizontal safety distances according to different voltage levels, and establish a voltage level rule table based on each voltage level, minimum vertical safety distance, and minimum horizontal safety distance; Determine the slope of each terrain where all existing transmission channels are located, set different operational constraints based on different terrain slopes, and establish an operational constraint rule table; Multiple temperature ranges are established, and a safety distance correction factor is set for each of the temperature ranges to establish a meteorological correction rule table; Multiple distance ranges are established for the distance between vegetation and power lines, and a risk level is set for each distance range to create a vegetation risk level rule table; Multiple sag ranges are established for the sag status of power lines, and a warning level is set for each sag range to establish a sag warning level rule table; The voltage level rule table, operation constraint rule table, meteorological correction rule table, vegetation risk level rule table, and sag warning level rule table are combined to obtain the procedure knowledge base.

7. The method of claim 1, wherein, The method further includes global feature extraction of the 3D point cloud data, including: For each point in the three-dimensional point cloud data, determine the neighborhood point set of that point, and calculate the covariance matrix of the neighborhood point set; Singular value decomposition is performed on the covariance matrix to obtain the first eigenvalue, second eigenvalue, and third eigenvalue; The linearity feature is calculated using the first and second eigenvalues. Flatness features are calculated using the first, second, and third eigenvalues; The scattering characteristics are calculated using the first and third eigenvalues.

8. A security maintenance device for a power transmission corridor, characterized by include: The 3D point cloud data acquisition module is used to acquire 3D point cloud data of the power transmission channel; The type classification module is used to classify the three-dimensional point cloud data into point cloud types to determine ground points, power line points, tower points, and vegetation points. The feature extraction module is used to extract and integrate the features of the ground points, power line points, tower points and vegetation points respectively to obtain a point cloud statistical feature set; The matching module is used to match the point cloud statistical feature set with a preset procedure knowledge base to determine the operating status and potential risk level of each section of the power transmission channel. The operation procedure strategy generation module is used to call a preset set of disposal strategies, generate operation procedure strategies based on the operation status and potential risk level of each section, and apply them to the power transmission channel.

9. A security maintenance device for a power transmission corridor, characterized by Including memory and processor; The memory is used to store programs; The processor is used to execute the program to implement the various steps of the power transmission channel safety maintenance method as described in any one of claims 1-7.

10. A storage medium having stored thereon a computer program, characterized in that When the computer program is executed by the processor, it implements the various steps of the power transmission channel safety maintenance method as described in any one of claims 1-7.

Images