Three-dimensional point cloud monitoring method and system for breakage and galloping of overhead line of power transmission line iron tower

Abstract

The invention relates to the technical field of power transmission line monitoring, and provides a three-dimensional point cloud monitoring method and system for fracture and galloping of an overhead line of a power transmission line iron tower, and the method comprises the following steps: arranging a three-dimensional laser scanner along a power transmission line; acquiring original three-dimensional point cloud data of the power transmission line in real time; performing real-time noise reduction, filtering and compression on the original three-dimensional point cloud data to obtain three-dimensional point cloud data; based on the three-dimensional point cloud data, establishing a three-dimensional digital model of the power transmission line, extracting characteristic parameters of the three-dimensional digital model of the power transmission line, calculating a time sequence change rate of the characteristic parameters, and constructing a characteristic vector space; in the feature vector space, feature matching is carried out on the feature vectors, the operation state of the power transmission line is recognized in real time, and whether fracture galloping exists or not is judged; and generating a monitoring report according to the judgment result of the fracture galloping. According to the invention, the coverage range and acquisition precision of transmission line fracture galloping monitoring are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power transmission line monitoring, and in particular to a three-dimensional point cloud monitoring method and system for fracture and vibration of an overhead line of a power transmission line tower. Background Art

[0002] As an important part of the power system, the safe and stable operation of transmission lines is directly related to the reliability of power supply. Overhead transmission lines are prone to breakage and galloping during long-term operation. If these faults are not discovered and handled in time, they will lead to serious power supply accidents and economic losses.

[0003] Existing technologies generally use simple feature matching algorithms to monitor the status of power transmission lines. Simple feature matching algorithms only consider basic geometric feature parameters and identify abnormal conditions through threshold judgment. The monitoring coverage is limited, making it difficult to achieve comprehensive monitoring of long-distance power transmission lines. Monitoring blind spots are prone to occur, and due to the lack of multi-dimensional feature analysis and historical data support, misjudgments and missed judgments are prone to occur, and potential faults cannot be discovered in a timely manner. Summary of the invention

[0004] In view of this, the present invention proposes a three-dimensional point cloud monitoring method and system for the fracture and vibration of overhead lines of transmission line towers, which solves the problem that the existing technology is prone to misjudgment and missed judgment, and cannot detect potential faults in time due to the lack of multi-dimensional feature analysis and historical data support.

[0005] The technical solution of the present invention is implemented as follows: In the first aspect, the present invention provides a three-dimensional point cloud monitoring method for the fracture and vibration of an overhead line of a transmission line tower, comprising the following steps:

[0006] Arrange a three-dimensional laser scanner along the transmission line, including a main scanner and multiple slave scanners;

[0007] The main scanner and the slave scanner work together in a high-frequency scanning manner to obtain original three-dimensional point cloud data of the power transmission line in real time;

[0008] Performing real-time noise reduction, filtering and compression on the original three-dimensional point cloud data to obtain pre-processed three-dimensional point cloud data;

[0009] Based on the three-dimensional point cloud data, a three-dimensional digital model of the transmission line is established, characteristic parameters of the three-dimensional digital model of the transmission line are extracted, the time series change rate of the characteristic parameters is calculated, and a characteristic vector space is constructed;

[0010] In the feature vector space, based on the state recognition model, the feature vectors are matched in combination with the normal state feature library, the fracture and galloping feature library and the environmental interference feature library, so as to identify the operation state of the transmission line in real time and determine whether there is fracture and galloping;

[0011] Based on the judgment result of fracture and galloping, a monitoring report is generated, which includes the line status judgment result, abnormal position positioning information and characteristic parameter analysis chart. When the fracture and galloping state is identified, detailed point cloud data is recorded to mark the time and location of the abnormality.

[0012] On the basis of the above technical solution, preferably, the three-dimensional laser scanner arranged along the power transmission line includes a main scanner and multiple slave scanners, specifically including:

[0013] According to the topographical characteristics of the transmission line and the distribution of the iron towers, the layout positions of the main scanner and the slave scanner are determined, wherein the main scanner is preferentially arranged at the top of the key iron tower, and the slave scanner is arranged at the non-key position of the transmission line at a fixed interval;

[0014] A wireless communication network is established between monitoring devices, and a real-time data transmission network based on a wireless communication protocol is established. The data transmission between the master scanner and the slave scanner adopts a multi-channel parallel transmission method.

[0015] On the basis of the above technical solution, preferably, the main scanner and the slave scanner work together in a high-frequency scanning manner to obtain the original three-dimensional point cloud data of the power transmission line in real time, specifically including:

[0016] The main scanner and the slave scanner work together to synchronously scan the transmission line using a preset scanning frequency and scanning angle; the main scanner uses a 360° omnidirectional scanning mode to obtain a three-dimensional point cloud of the top of the tower and the surrounding transmission lines; the slave scanner uses a fixed-angle scanning mode, and the scanning range covers the conductor area of ​​the transmission line;

[0017] According to the scanning range of the main scanner and the slave scanner and the actual distribution of the transmission lines, the scanning parameters are dynamically adjusted to obtain the original three-dimensional point cloud data with complete coverage, and the scanning frequency and scanning angle are adjusted in real time according to the environmental condition parameters, which include wind speed, temperature and humidity; when it is detected that the vibration amplitude of the transmission line is large or there is a scanning blind area, the scanning frequency is dynamically increased or the scanning angle is adjusted.

[0018] On the basis of the above technical solution, preferably, the real-time noise reduction, filtering and compression of the original three-dimensional point cloud data to obtain the pre-processed three-dimensional point cloud data specifically includes:

[0019] A noise removal algorithm based on statistical analysis is used to calculate the noise weight value of each point in the original 3D point cloud data, and outliers are removed according to the noise weight value; combined with the region growing filter algorithm, based on the spatial distribution relationship of the neighborhood points, the point cloud data of non-target areas is removed to obtain the noise reduction filter point cloud data;

[0020] The calculation formula of the noise removal algorithm based on statistical analysis is:

[0021]

[0022] Among them, d avg (i) is point P i (x i ,y i ,z i ) is the average distance of the neighboring points, N1 is the number of neighboring points, P j is the jth neighborhood point, σ d (i) is point P i (x i ,y i ,z i )’s neighborhood points;

[0023] The judgment conditions of noise points are:

[0024]

[0025] in, is the noise point, k1 is the noise threshold coefficient;

[0026] The denoising filtered point cloud data is compressed by using a voxel-based gridding method, the denoising filtered point cloud data is divided into a number of voxel units, representative points are selected in each voxel unit, and a feature point extraction method based on curvature analysis is combined to retain key feature points and curvature significant points of the wire area of ​​the denoising filtered point cloud data to obtain three-dimensional point cloud data;

[0027] The voxel unit is V(x, y, z), the side length of the voxel unit is L1, and the representative point is the geometric center of all points in the voxel unit, and the calculation formula is:

[0028]

[0029] Where M1 is the number of points in the voxel unit, is the k2th point in the voxel unit, P c For representative points.

[0030] On the basis of the above technical solution, preferably, the three-dimensional digital model of the transmission line is established based on the three-dimensional point cloud data, the characteristic parameters of the three-dimensional digital model of the transmission line are extracted, the time series change rate of the characteristic parameters is calculated, and the characteristic vector space is constructed, which specifically includes:

[0031] The three-dimensional point cloud data is fitted using a B-spline curve fitting algorithm to obtain a spatial curve equation of the conductor, and a three-dimensional digital model of the transmission line is constructed based on a triangular mesh reconstruction algorithm;

[0032] Characteristic parameters are extracted from the three-dimensional digital model of the transmission line, wherein the characteristic parameters include conductor sag, space curve parameters, lateral displacement and vibration frequency; a time series data matrix of the characteristic parameters is established, the time series change rate of each characteristic parameter is calculated, and a characteristic vector space is constructed by principal component analysis.

[0033] On the basis of the above technical solution, preferably, in the feature vector space, based on the state recognition model, in combination with the normal state feature library, the fracture galloping feature library and the environmental interference feature library, feature matching is performed on the feature vector, the operation state of the transmission line is identified in real time, and whether there is fracture galloping is determined, specifically including:

[0034] A multi-level state recognition model is constructed based on a deep learning network, and the multi-level state recognition model includes a feature extraction layer, a feature fusion layer, and a state classification layer; the multi-level state recognition model is trained using sample data of a normal state feature library, a broken state feature library, and a dancing state feature library, and feature matching is performed between feature vectors and each state feature library through an adaptive feature matching algorithm;

[0035] Calculate the similarity score between the feature vector and each state feature library to generate the state recognition confidence; determine the operating state of the transmission line based on the multi-threshold judgment mechanism and combined with historical state information;

[0036] The calculation formula of the similarity score is:

[0037]

[0038] D(T) = α2·h(T) + α3·v(T);

[0039] Among them, V(T) is the feature vector at time T, S p is the pth sample feature vector in the feature library, ||V(T)-S p || is V(T) and S p The Euclidean distance between them, σ is the kernel function parameter, D(T) is the dynamic weight at time T, h(T) is the historical state factor, v(T) is the feature change rate factor, α2 and α3 are the weight coefficients of the historical state information and feature change rate respectively;

[0040] The calculation formula of the state recognition confidence is:

[0041]

[0042] Among them, Conf q (T) is the recognition confidence of the qth state at time T, N q is the total number of samples in the q-th state feature library, is the pth sample in the qth class state feature library, W q (T) is the state weight factor of the qth state, T q (T) is the duration of the qth state, μ q is the time gain coefficient, r is the state category index, corresponding to the normal state, the broken state and the dancing state respectively;

[0043] The decision formula for the operating state of the transmission line is:

[0044]

[0045] Where State(T) is the operating state of the transmission line at time T, L2 is the time window length, ψ(s) is the time decay function, s is the index in the time window, ΔT is the sampling time interval, Q q (T) is the state transition probability factor, To find the state category q that corresponds to the maximum value of the objective function.

[0046] On the basis of the above technical solution, preferably, a monitoring report is generated according to the judgment result of the fracture galloping, and the monitoring report includes the line state judgment result, abnormal position positioning information and characteristic parameter analysis chart. When the fracture galloping state is identified, detailed point cloud data is recorded to mark the time and location of the abnormality, specifically including:

[0047] Based on the judgment result of the fracture galloping, combined with the three-dimensional point cloud data and characteristic parameter analysis, a line status judgment report is generated, which includes the precise positioning information of the abnormal position and the time series change diagram of the fracture galloping characteristics;

[0048] When a fractured dancing state is identified, the storage of detailed point cloud data and the abnormality marking operation are triggered, and the time, location and related characteristic parameters of the abnormality are recorded.

[0049] In a second aspect, the present invention further provides a three-dimensional point cloud monitoring system for the fracture and vibration of an overhead line of a transmission line tower, the system comprising:

[0050] A scanner arrangement module, used for arranging three-dimensional laser scanners along the transmission line, including a master scanner and a plurality of slave scanners;

[0051] A data acquisition module, used to acquire original three-dimensional point cloud data of the power transmission line in real time by the main scanner and the slave scanner working together in a high-frequency scanning manner;

[0052] A data processing module, used for performing real-time noise reduction, filtering and compression on the original three-dimensional point cloud data to obtain pre-processed three-dimensional point cloud data;

[0053] A feature extraction module, used to establish a three-dimensional digital model of the transmission line based on the three-dimensional point cloud data, extract characteristic parameters of the three-dimensional digital model of the transmission line, calculate the time series change rate of the characteristic parameters, and construct a characteristic vector space;

[0054] A state recognition module is used to perform feature matching on feature vectors in the feature vector space based on a state recognition model, in combination with a normal state feature library, a fracture and pulsation feature library, and an environmental interference feature library, so as to recognize the operation state of the transmission line in real time and determine whether there is fracture and pulsation;

[0055] The report generation module is used to generate a monitoring report based on the judgment results of fracture and galloping. The monitoring report includes the line status judgment results, abnormal position positioning information and characteristic parameter analysis charts. When the fracture and galloping state is identified, the detailed point cloud data is recorded to mark the time and location of the abnormality.

[0056] In a third aspect, the present invention further provides an electronic device, comprising: at least one processor, at least one memory, a communication interface and a bus;

[0057] Among them, the processor, memory, and communication interface communicate with each other through the bus, the memory stores program instructions that can be executed by the processor, and the processor calls the program instructions to implement steps such as a three-dimensional point cloud monitoring method for the fracture and vibration of overhead lines of a transmission line tower.

[0058] In a fourth aspect, the present invention further provides a computer-readable storage medium storing computer instructions, wherein the computer instructions enable a computer to implement steps of a three-dimensional point cloud monitoring method for fracture and vibration of overhead lines of a transmission line tower.

[0059] The three-dimensional point cloud monitoring method and system for the fracture and galloping of an overhead line of a transmission line tower of the present invention has the following beneficial effects compared with the prior art:

[0060] (1) By arranging a main scanner and multiple slave scanners along the transmission line to cover key areas and non-key areas, monitoring blind spots are avoided, a wireless communication network is established, and real-time transmission of point cloud data between the main scanner and the slave scanner is realized, thereby improving data sharing and processing efficiency. The main scanner and the slave scanner work together in a high-frequency scanning mode, which improves the coverage and acquisition accuracy of transmission line fracture and vibration monitoring; (2) By performing noise reduction, filtering and compression on the original three-dimensional point cloud data, the quality and processing efficiency of the point cloud data are improved. The noise removal algorithm based on statistical analysis and the regional growth filtering algorithm are used to remove abnormal points and non-target area point cloud data. Combined with the voxel grid compression method, the data storage volume is reduced while retaining key feature points and significant curvature points to ensure data integrity;

[0061] (3) By constructing a feature vector space and combining it with a state recognition model, feature matching is performed using the normal state feature library, fracture and vibration feature library, and environmental interference feature library to identify the operating state of the transmission line in real time, calculate the similarity score between the feature vector and the feature library, generate the state recognition confidence, and judge whether the transmission line is in a fracture and vibration state based on a multi-threshold judgment mechanism combined with historical state information. At the same time, it effectively distinguishes between environmental interference and abnormal states, thereby improving the accuracy of state recognition. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0063] Figure 1 A flow chart of a three-dimensional point cloud monitoring method for fracture and vibration of an overhead line of a power transmission line tower according to the present invention;

[0064] Figure 2 This is a structural diagram of a three-dimensional point cloud monitoring system for fracture and vibration of an overhead line of a transmission line tower according to the present invention. DETAILED DESCRIPTION

[0065] The following will be combined with the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0066] See also Figure 1 The present invention provides a three-dimensional point cloud monitoring method for the fracture and vibration of an overhead line of a transmission line tower, comprising the following steps:

[0067] Arrange 3D laser scanners along the transmission line, including a master scanner and multiple slave scanners; adopt a "one master and multiple slaves" arrangement, where the master scanner is set on the top of the tower, and the slave scanners are set at intervals along the line, and a wireless communication network is established between the monitoring devices to achieve real-time transmission of point cloud data;

[0068] The main scanner and the slave scanner work together in a high-frequency scanning manner to obtain original three-dimensional point cloud data of the power transmission line in real time;

[0069] Performing real-time noise reduction, filtering and compression on the original three-dimensional point cloud data to obtain pre-processed three-dimensional point cloud data;

[0070] Based on the three-dimensional point cloud data, a three-dimensional digital model of the transmission line is established, characteristic parameters of the three-dimensional digital model of the transmission line are extracted, the characteristic parameters include conductor sag, space curve, lateral displacement and vibration frequency, the time series change rate of the characteristic parameters is calculated, and a characteristic vector space is constructed;

[0071] In the feature vector space, based on the state recognition model, the feature vectors are matched in combination with the normal state feature library, the fracture and galloping feature library and the environmental interference feature library, so as to identify the operation state of the transmission line in real time and determine whether there is fracture and galloping;

[0072] Based on the judgment result of fracture and galloping, a monitoring report is generated, which includes the line status judgment result, abnormal position positioning information and characteristic parameter analysis chart. When the fracture and galloping state is identified, detailed point cloud data is recorded to mark the time and location of the abnormality.

[0073] Specifically, this embodiment arranges a main scanner and multiple slave scanners along the transmission line to cover key areas and non-key areas, avoids monitoring blind spots, establishes a wireless communication network, and realizes real-time transmission of point cloud data between the main scanner and the slave scanners, thereby improving data sharing and processing efficiency. The main scanner and the slave scanner work together in a high-frequency scanning manner, which improves the coverage and collection accuracy of transmission line fracture and dancing monitoring.

[0074] The three-dimensional laser scanners are arranged along the transmission line, including a main scanner and a plurality of slave scanners, specifically including:

[0075] According to the topographical characteristics of the transmission line and the distribution of the iron towers, the layout positions of the main scanner and the slave scanner are determined, wherein the main scanner is preferentially arranged at the top of the key iron tower, and the slave scanner is arranged at the non-key position of the transmission line at a fixed interval;

[0076] Establish a wireless communication network between monitoring devices to realize real-time transmission of point cloud data between the master scanner and the slave scanner, and establish a real-time data transmission network based on wireless communication protocols, wherein the wireless communication protocols include Wi-Fi, 4G / 5G or LoRa; the data transmission between the master scanner and the slave scanner adopts a multi-channel parallel transmission method to improve the transmission speed and stability of the point cloud data.

[0077] Specifically, this embodiment realizes the rational allocation of monitoring resources and improves the pertinence and economy of the monitoring system by making differentiated arrangements according to the terrain characteristics of the transmission lines and the distribution of towers. The main scanner is preferentially arranged on the top of the key tower, and the slave scanners are arranged at fixed intervals in non-key positions. A wireless communication network based on Wi-Fi, 4G / 5G or LoRa is established, which provides a variety of optional communication methods for data transmission between the master and slave scanners, thereby enhancing adaptability.

[0078] The main scanner and the slave scanner work together in a high-frequency scanning manner to obtain the original three-dimensional point cloud data of the power transmission line in real time, specifically including:

[0079] The main scanner and the slave scanner work together to synchronously scan the transmission line using a preset scanning frequency and scanning angle; the main scanner uses a 360° omnidirectional scanning mode to obtain a three-dimensional point cloud of the top of the tower and the surrounding transmission lines; the slave scanner uses a fixed-angle scanning mode, and the scanning range covers the conductor area of ​​the transmission line; the main scanner and the slave scanner work in coordination through a time synchronization mechanism to ensure the time consistency of the point cloud data;

[0080] According to the scanning range of the main scanner and the slave scanner and the actual distribution of the transmission lines, the scanning parameters are dynamically adjusted to obtain the original three-dimensional point cloud data with complete coverage. The scanning frequency and scanning angle are adjusted in real time according to the environmental condition parameters, which include wind speed, temperature and humidity. When a large vibration amplitude of the transmission line is detected or a blind spot is scanned, the scanning frequency is dynamically increased or the scanning angle is adjusted to ensure the integrity and accuracy of the point cloud data.

[0081] Specifically, this embodiment can effectively cover all areas of the power transmission line by dynamically adjusting scanning parameters (such as scanning frequency and scanning angle), especially in complex terrain or in the presence of scanning blind spots, to ensure the comprehensiveness of point cloud data and avoid data loss. The scanning frequency and angle are adjusted in real time according to environmental conditions such as wind speed, temperature and humidity, which can adapt to the dynamic changes in the operating environment of the power transmission line and ensure the stability and accuracy of data collection.

[0082] When a large vibration amplitude of the transmission line is detected, dynamically increasing the scanning frequency or adjusting the scanning angle can more accurately capture the point cloud data under the vibration state. By dynamically adjusting the scanning parameters, the waste of resources or monitoring blind spots caused by fixed scanning parameters can be avoided, thereby improving monitoring efficiency while ensuring data integrity.

[0083] The real-time noise reduction, filtering and compression of the original three-dimensional point cloud data to obtain pre-processed three-dimensional point cloud data specifically includes:

[0084] A noise removal algorithm based on statistical analysis is used to calculate the noise weight value of each point in the original 3D point cloud data, and outliers are removed according to the noise weight value; combined with the region growing filter algorithm, based on the spatial distribution relationship of the neighborhood points, the point cloud data of non-target areas is removed to obtain the noise reduction filter point cloud data;

[0085] The calculation formula of the noise removal algorithm based on statistical analysis is:

[0086]

[0087] Among them, d avg (i) is point P i (x i ,y i ,z i ), N1 is the number of neighboring points, P j is the jth neighborhood point, σ d (i) is point P i (x i ,y i ,z i )’s neighborhood points;

[0088] The judgment conditions of noise points are:

[0089]

[0090] in, is the noise point, k1 is the noise threshold coefficient;

[0091] The denoising filtered point cloud data is compressed by using a voxel-based gridding method, the denoising filtered point cloud data is divided into a number of voxel units, representative points are selected in each voxel unit, and a feature point extraction method based on curvature analysis is combined to retain key feature points and curvature significant points of the wire area of ​​the denoising filtered point cloud data to obtain three-dimensional point cloud data;

[0092] The voxel unit is V(x, y, z), the side length of the voxel unit is L1, and the representative point is the geometric center of all points in the voxel unit, and the calculation formula is:

[0093]

[0094] Where M1 is the number of points in the voxel unit, is the k2th point in the voxel unit, P c For representative points.

[0095] Specifically, this embodiment uses a noise removal algorithm based on statistical analysis to calculate the noise weight value of each point and remove abnormal points, which can effectively eliminate noise interference in the original point cloud data and improve the purity of the data. By combining the regional growing filtering algorithm, the point cloud data of non-target areas is removed based on the spatial distribution relationship of neighborhood points, and the point cloud data of the transmission line conductor area can be accurately retained, reducing the interference of irrelevant data, and providing high-quality input data for feature extraction.

[0096] The voxel grid compression method is used to divide the point cloud data after noise reduction and filtering into several voxel units, and representative points are selected in each voxel unit, which significantly reduces the storage volume of point cloud data, while retaining key feature points and curvature significant points, ensuring that important information is not lost while data is compressed. Through the feature point extraction method based on curvature analysis, the key feature points and curvature significant points of the wire area can be retained on the basis of compressed data. By combining noise reduction, filtering and compression, the data volume can be reduced while ensuring data quality, and the processing speed and real-time performance of point cloud data can be improved.

[0097] The method of establishing a three-dimensional digital model of a power transmission line based on the three-dimensional point cloud data, extracting characteristic parameters of the three-dimensional digital model of the power transmission line, calculating the time series change rate of the characteristic parameters, and constructing a characteristic vector space specifically includes:

[0098] The three-dimensional point cloud data is fitted using a B-spline curve fitting algorithm to obtain a spatial curve equation of the conductor, and a three-dimensional digital model of the transmission line is constructed based on a triangular mesh reconstruction algorithm;

[0099] The calculation formula of the B-spline curve fitting algorithm is:

[0100]

[0101] W k =α1·ρ k +(1-α1)·κ k ;

[0102] Where C(·) is the B-spline curve equation, u is the curve parameter, m is the number of control points, N k,k(u) is the k-order B-spline basis function, W k is the weight coefficient of the kth point cloud data point, P k is the three-dimensional coordinate of the kth point cloud data point, α1 is the weight adjustment coefficient, ρ k is the local point cloud density of the kth point cloud data point, κ k is the local curvature of the kth point cloud data point;

[0103] Extract characteristic parameters from the three-dimensional digital model of the transmission line, wherein the characteristic parameters include conductor sag, space curve parameters, lateral displacement and vibration frequency; establish a time series data matrix of the characteristic parameters, calculate the time series change rate of each characteristic parameter, and construct a characteristic vector space by principal component analysis;

[0104] The calculation formula of the timing change rate is:

[0105]

[0106] Among them, R(t) is the time series change rate of the characteristic parameter, w n is the weight coefficient of the nth characteristic parameter, f n (t) is the value of the nth characteristic parameter at time t, f n (t-Δt) is the value of the nth characteristic parameter at time t-Δt, and Δt is the time interval between adjacent moments.

[0107] Specifically, this embodiment fits the three-dimensional point cloud data through the B-spline curve fitting algorithm, obtains the spatial curve equation of the conductor, and combines the triangular mesh reconstruction algorithm to construct a high-precision three-dimensional digital model of the transmission line. This embodiment extracts the key characteristic parameters of the conductor (such as sag, spatial curve parameters, lateral displacement and vibration frequency) based on the three-dimensional digital model to reflect the operating status of the transmission line and ensure the integrity of the extraction of characteristic parameters. By establishing a time series data matrix of characteristic parameters and calculating the time series change rate of each characteristic parameter, the operating status change of the transmission line is dynamically monitored.

[0108] By using principal component analysis to reduce the dimension of characteristic parameters and constructing a characteristic vector space, it is possible to effectively reduce redundant information and improve the efficiency and accuracy of feature matching and state recognition. By analyzing the time series change rate of characteristic parameters, it is possible to quickly capture abnormal changes in the operating status of the transmission line.

[0109] In the feature vector space, based on the state recognition model, in combination with the normal state feature library, the fracture galloping feature library and the environmental interference feature library, feature matching is performed on the feature vector, the operation state of the transmission line is recognized in real time, and whether there is fracture galloping is determined, specifically including:

[0110] A multi-level state recognition model is constructed based on a deep learning network, and the multi-level state recognition model includes a feature extraction layer, a feature fusion layer, and a state classification layer; the multi-level state recognition model is trained using sample data of a normal state feature library, a broken state feature library, and a dancing state feature library, and feature matching is performed between feature vectors and each state feature library through an adaptive feature matching algorithm;

[0111] Calculate the similarity score between the feature vector and each state feature library to generate the state recognition confidence; determine the operating state of the transmission line based on the multi-threshold judgment mechanism and combined with historical state information;

[0112] The calculation formula of the similarity score is:

[0113]

[0114] D(T) = α2·h(T) + α3·v(T);

[0115] Among them, V(T) is the feature vector at time T, S p is the pth sample feature vector in the feature library, ||V(T)-S p || is V(T) and S p The Euclidean distance between them, σ is the kernel function parameter, D(T) is the dynamic weight at time T, h(T) is the historical state factor, v(T) is the feature change rate factor, α2 and α3 are the weight coefficients of the historical state information and feature change rate respectively;

[0116] The calculation formula of the state recognition confidence is:

[0117]

[0118] Among them, Conf q (T) is the recognition confidence of the qth state at time T, N q is the total number of samples in the q-th state feature library, is the pth sample in the qth class state feature library, W q (T) is the state weight factor of the qth state, T q (T) is the duration of the qth state, μ q is the time gain coefficient, r is the state category index, corresponding to the normal state, the broken state and the dancing state respectively;

[0119] The decision formula for the operating state of the transmission line is:

[0120]

[0121] Where State(T) is the operating state of the transmission line at time T, L2 is the time window length, ψ(s) is the time decay function, s is the index in the time window, ΔT is the sampling time interval, Q q (T) is the state transition probability factor, To find the state category q that corresponds to the maximum value of the objective function.

[0122] Specifically, this embodiment constructs a feature vector space and combines it with a state recognition model to perform real-time analysis and judgment on the operating status of the transmission line. It combines the normal state feature library, the fracture and dancing feature library, and the environmental interference feature library, and compares and analyzes the feature vectors through a feature matching algorithm, which can effectively distinguish between the normal state, the fracture and dancing state, and the environmental interference state.

[0123] By introducing the environmental interference feature library, it is possible to identify and filter non-abnormal conditions caused by environmental factors such as wind speed, temperature, and humidity to avoid misjudgment. By using a multidimensional feature matching method based on feature vector space, a variety of characteristic parameters such as conductor sag, spatial curve, lateral displacement, and vibration frequency are comprehensively analyzed to fully reflect the operating status of the transmission line and ensure the comprehensiveness of status identification.

[0124] According to the judgment result of the fracture galloping, a monitoring report is generated, which includes the line state judgment result, abnormal position location information and characteristic parameter analysis chart. When the fracture galloping state is identified, detailed point cloud data is recorded to mark the time and location of the abnormality, specifically including:

[0125] Based on the judgment result of the fracture galloping, combined with the three-dimensional point cloud data and characteristic parameter analysis, a line status judgment report is generated, which includes the precise positioning information of the abnormal position and the time series change diagram of the fracture galloping characteristics;

[0126] The point cloud of the abnormal area is segmented through the spatial clustering algorithm of three-dimensional point cloud data to extract the geometric features of the abnormal area; the occurrence mechanism of the fracture galloping is analyzed by combining the geometric features of the abnormal area and the time series change rate of the characteristic parameters, and the time series change curve of the fracture galloping characteristics is generated; according to the positioning information and characteristic analysis results of the abnormal area, a detailed line status judgment report is generated and transmitted to the monitoring center in real time through the wireless communication network;

[0127] When the fracture dancing state is identified, the storage of detailed point cloud data and the abnormality marking operation are triggered, and the time, location and related characteristic parameters of the abnormality are recorded;

[0128] When the fracture dancing state is identified, the high-precision point cloud acquisition mode is automatically triggered to record the detailed point cloud data of the abnormal area; based on the detailed point cloud data of the abnormal area, the time, location and corresponding characteristic parameters of the fracture dancing are marked, and the marked information is stored in the monitoring system database for analysis and fault handling.

[0129] Specifically, this embodiment uses a spatial clustering algorithm of three-dimensional point cloud data to segment the point cloud of the abnormal area, extract the geometric features of the abnormal area, accurately locate the position where the fracture dance occurs, and generate a time series change curve of the fracture dance characteristics by combining the geometric features of the abnormal area and the time series change rate of the characteristic parameters.

[0130] According to the positioning information and characteristic analysis results of the abnormal area, a detailed monitoring report containing the line status determination results, abnormal position positioning information and characteristic parameter analysis charts is generated to ensure that the report content is comprehensive and intuitive, providing a clear basis for decision-making for the monitoring center. When the fracture galloping state is identified, the high-precision point cloud acquisition mode is automatically triggered to record the detailed point cloud data of the abnormal area, and mark the time, location and corresponding characteristic parameters of the fracture galloping to ensure the integrity and traceability of the abnormal data. The monitoring report is transmitted to the monitoring center in real time through the wireless communication network, realizing the rapid reporting and real-time monitoring of the fracture galloping state, and improving the response speed of the monitoring system.

[0131] See also Figure 2 The present invention also provides a three-dimensional point cloud monitoring system for the fracture and vibration of an overhead line of a transmission line tower, the system comprising:

[0132] A scanner arrangement module, used for arranging three-dimensional laser scanners along the transmission line, including a master scanner and a plurality of slave scanners;

[0133] A data acquisition module, used to acquire original three-dimensional point cloud data of the power transmission line in real time by the main scanner and the slave scanner working together in a high-frequency scanning manner;

[0134] A data processing module, used for performing real-time noise reduction, filtering and compression on the original three-dimensional point cloud data to obtain pre-processed three-dimensional point cloud data;

[0135] A feature extraction module, used to establish a three-dimensional digital model of the transmission line based on the three-dimensional point cloud data, extract characteristic parameters of the three-dimensional digital model of the transmission line, calculate the time series change rate of the characteristic parameters, and construct a characteristic vector space;

[0136] A state recognition module is used to perform feature matching on feature vectors in the feature vector space based on a state recognition model, in combination with a normal state feature library, a fracture and pulsation feature library, and an environmental interference feature library, so as to recognize the operation state of the transmission line in real time and determine whether there is fracture and pulsation;

[0137] The report generation module is used to generate a monitoring report based on the judgment results of fracture and galloping. The monitoring report includes the line status judgment results, abnormal position positioning information and characteristic parameter analysis charts. When the fracture and galloping state is identified, the detailed point cloud data is recorded to mark the time and location of the abnormality.

[0138] Specifically, a three-dimensional point cloud monitoring system for the fracture and vibration of overhead lines on power transmission line towers in this embodiment realizes real-time monitoring, accurate identification and rapid warning of fracture and vibration of power transmission lines through the collaborative work of scanner arrangement module, point cloud data processing module, three-dimensional digital model construction module, state recognition module and report generation module. The system adopts a "one master and multiple slaves" scanner arrangement method, combined with wireless communication network and multi-channel parallel transmission to ensure the real-time and integrity of point cloud data acquisition; improves data quality through noise reduction, filtering and compression, and extracts feature parameters based on three-dimensional digital models to construct feature vector space for state identification and abnormality judgment; and finally generates a monitoring report containing abnormality location and feature analysis. The system significantly improves the comprehensiveness, real-time and intelligent level of monitoring.

[0139] The present invention also discloses an electronic device, comprising: at least one processor, at least one memory communication interface and a bus: wherein the processor, memory and communication interface communicate with each other through the bus; the memory stores program instructions that can be executed by the processor, and the processor calls the program instructions to implement a three-dimensional point cloud monitoring method for fracture and vibration of overhead lines of transmission line towers.

[0140] The present invention also discloses a computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions, wherein the computer instructions enable the computer to implement all or part of the steps of a three-dimensional point cloud monitoring method for fracture and galloping of an overhead line of a transmission line tower described in an embodiment of the present invention. The storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory ROM, a random access memory RAM, a magnetic disk or an optical disk.

[0141] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A three-dimensional point cloud monitoring method for the fracture and vibration of an overhead line of a transmission line tower, characterized in that: The following steps are involved: Arrange a three-dimensional laser scanner along the transmission line, including a main scanner and multiple slave scanners; The main scanner and the slave scanner work together in a high-frequency scanning manner to obtain original three-dimensional point cloud data of the power transmission line in real time; Performing real-time noise reduction, filtering and compression on the original three-dimensional point cloud data to obtain pre-processed three-dimensional point cloud data; Based on the three-dimensional point cloud data, a three-dimensional digital model of the transmission line is established, characteristic parameters of the three-dimensional digital model of the transmission line are extracted, the time series change rate of the characteristic parameters is calculated, and a characteristic vector space is constructed; In the feature vector space, based on the state recognition model, the feature vectors are matched in combination with the normal state feature library, the fracture and galloping feature library and the environmental interference feature library, so as to identify the operation state of the transmission line in real time and determine whether there is fracture and galloping; Based on the judgment result of fracture and galloping, a monitoring report is generated, which includes the line status judgment result, abnormal position positioning information and characteristic parameter analysis chart. When the fracture and galloping state is identified, detailed point cloud data is recorded to mark the time and location of the abnormality.

2. A three-dimensional point cloud monitoring method for fracture and vibration of an overhead line of a transmission line tower as claimed in claim 1, characterized in that: The three-dimensional laser scanners are arranged along the transmission line, including a main scanner and a plurality of slave scanners, specifically including: According to the topographical characteristics of the transmission line and the distribution of the iron towers, the layout positions of the main scanner and the slave scanner are determined, wherein the main scanner is preferentially arranged at the top of the key iron tower, and the slave scanner is arranged at the non-key position of the transmission line at a fixed interval; A wireless communication network is established between monitoring devices, and a real-time data transmission network based on a wireless communication protocol is established. The data transmission between the master scanner and the slave scanner adopts a multi-channel parallel transmission method.

3. A three-dimensional point cloud monitoring method for fracture and vibration of an overhead line of a transmission line tower as claimed in claim 1, characterized in that: The main scanner and the slave scanner work together in a high-frequency scanning manner to obtain the original three-dimensional point cloud data of the power transmission line in real time, specifically including: The main scanner and the slave scanner work together to synchronously scan the transmission line using a preset scanning frequency and scanning angle; the main scanner uses a 360° omnidirectional scanning mode to obtain a three-dimensional point cloud of the top of the tower and the surrounding transmission lines; the slave scanner uses a fixed-angle scanning mode, and the scanning range covers the conductor area of ​​the transmission line; According to the scanning range of the main scanner and the slave scanner and the actual distribution of the transmission lines, the scanning parameters are dynamically adjusted to obtain the original three-dimensional point cloud data with complete coverage, and the scanning frequency and scanning angle are adjusted in real time according to the environmental condition parameters, which include wind speed, temperature and humidity; when it is detected that the vibration amplitude of the transmission line is large or there is a scanning blind area, the scanning frequency is dynamically increased or the scanning angle is adjusted.

4. A three-dimensional point cloud monitoring method for fracture and vibration of an overhead line of a transmission line tower as claimed in claim 1, characterized in that: The real-time noise reduction, filtering and compression of the original three-dimensional point cloud data to obtain pre-processed three-dimensional point cloud data specifically includes: A noise removal algorithm based on statistical analysis is used to calculate the noise weight value of each point in the original 3D point cloud data, and outliers are removed according to the noise weight value; combined with the region growing filter algorithm, based on the spatial distribution relationship of the neighborhood points, the point cloud data of non-target areas is removed to obtain the noise reduction filter point cloud data; The calculation formula of the noise removal algorithm based on statistical analysis is: Among them, d avg (i) is point P i (x i ,y i ,z i ) is the average distance of the neighboring points, N1 is the number of neighboring points, P j is the jth neighborhood point, σ d (i) is point P i (x i ,y i ,z i )’s neighborhood points; The judgment conditions of noise points are: in, is the noise point, k1 is the noise threshold coefficient; The denoising filtered point cloud data is compressed by using a voxel-based gridding method, the denoising filtered point cloud data is divided into a number of voxel units, representative points are selected in each voxel unit, and a feature point extraction method based on curvature analysis is combined to retain key feature points and curvature significant points of the wire area of ​​the denoising filtered point cloud data to obtain three-dimensional point cloud data; The voxel unit is V(x, y, z), the side length of the voxel unit is L1, and the representative point is the geometric center of all points in the voxel unit, and the calculation formula is: Where M1 is the number of points in the voxel unit, is the k2th point in the voxel unit, P c For representative points.

5. A three-dimensional point cloud monitoring method for fracture and vibration of an overhead line of a transmission line tower as claimed in claim 1, characterized in that: The method of establishing a three-dimensional digital model of a power transmission line based on the three-dimensional point cloud data, extracting characteristic parameters of the three-dimensional digital model of the power transmission line, calculating the time series change rate of the characteristic parameters, and constructing a characteristic vector space specifically includes: The three-dimensional point cloud data is fitted using a B-spline curve fitting algorithm to obtain a spatial curve equation of the conductor, and a three-dimensional digital model of the transmission line is constructed based on a triangular mesh reconstruction algorithm; Characteristic parameters are extracted from the three-dimensional digital model of the transmission line, wherein the characteristic parameters include conductor sag, space curve parameters, lateral displacement and vibration frequency; a time series data matrix of the characteristic parameters is established, the time series change rate of each characteristic parameter is calculated, and a characteristic vector space is constructed by principal component analysis.

6. A three-dimensional point cloud monitoring method for fracture and vibration of an overhead line of a transmission line tower as claimed in claim 5, characterized in that: In the feature vector space, based on the state recognition model, in combination with the normal state feature library, the fracture galloping feature library and the environmental interference feature library, feature matching is performed on the feature vector, the operation state of the transmission line is recognized in real time, and whether there is fracture galloping is determined, specifically including: A multi-level state recognition model is constructed based on a deep learning network, and the multi-level state recognition model includes a feature extraction layer, a feature fusion layer, and a state classification layer; the multi-level state recognition model is trained using sample data of a normal state feature library, a broken state feature library, and a dancing state feature library, and feature matching is performed between feature vectors and each state feature library through an adaptive feature matching algorithm; Calculate the similarity score between the feature vector and each state feature library to generate the state recognition confidence; determine the operating state of the transmission line based on the multi-threshold judgment mechanism and combined with historical state information; The calculation formula of the similarity score is: D(T) = α2·h(T) + α3·v(T); Among them, V(T) is the feature vector at time T, S p is the pth sample feature vector in the feature library, ||V(T)-S p || is V(T) and S p The Euclidean distance between them, σ is the kernel function parameter, D(T) is the dynamic weight at time T, h(T) is the historical state factor, v(T) is the feature change rate factor, α2 and α3 are the weight coefficients of the historical state information and feature change rate respectively; The calculation formula of the state recognition confidence is: Among them, Conf q (T) is the recognition confidence of the qth state at time T, N q is the total number of samples in the q-th state feature library, is the pth sample in the qth class state feature library, W q (T) is the state weight factor of the qth state, T q (T) is the duration of the qth state, μ q is the time gain coefficient, r is the state category index, corresponding to the normal state, the broken state and the dancing state respectively; The decision formula for the operating state of the transmission line is: Where State(T) is the operating state of the transmission line at time T, L2 is the time window length, ψ(s) is the time decay function, s is the index in the time window, ΔT is the sampling time interval, Q q (T) is the state transition probability factor, To find the state category q that corresponds to the maximum value of the objective function.

7. A three-dimensional point cloud monitoring method for fracture and vibration of an overhead line of a transmission line tower as claimed in claim 1, characterized in that: According to the judgment result of the fracture galloping, a monitoring report is generated, which includes the line state judgment result, abnormal position location information and characteristic parameter analysis chart. When the fracture galloping state is identified, detailed point cloud data is recorded to mark the time and location of the abnormality, specifically including: Based on the judgment result of the fracture galloping, combined with the three-dimensional point cloud data and characteristic parameter analysis, a line status judgment report is generated, which includes the precise positioning information of the abnormal position and the time series change diagram of the fracture galloping characteristics; When a fractured dancing state is identified, the storage of detailed point cloud data and the abnormality marking operation are triggered, and the time, location and related characteristic parameters of the abnormality are recorded.

8. A three-dimensional point cloud monitoring system for the fracture and vibration of transmission line tower overhead lines, characterized in that: The system comprises: A scanner arrangement module, used for arranging three-dimensional laser scanners along the transmission line, including a master scanner and a plurality of slave scanners; A data acquisition module, used to acquire original three-dimensional point cloud data of the power transmission line in real time by the main scanner and the slave scanner working together in a high-frequency scanning manner; A data processing module, used for performing real-time noise reduction, filtering and compression on the original three-dimensional point cloud data to obtain pre-processed three-dimensional point cloud data; A feature extraction module, used to establish a three-dimensional digital model of the transmission line based on the three-dimensional point cloud data, extract characteristic parameters of the three-dimensional digital model of the transmission line, calculate the time series change rate of the characteristic parameters, and construct a characteristic vector space; A state recognition module is used to perform feature matching on feature vectors in the feature vector space based on a state recognition model, in combination with a normal state feature library, a fracture and pulsation feature library, and an environmental interference feature library, so as to identify the operating state of the transmission line in real time and determine whether there is fracture and pulsation; The report generation module is used to generate a monitoring report based on the judgment results of fracture and galloping. The monitoring report includes the line status judgment results, abnormal position positioning information and characteristic parameter analysis charts. When the fracture and galloping state is identified, the detailed point cloud data is recorded to mark the time and location of the abnormality.

9. An electronic device, characterized in that: include: at least one processor, at least one memory, a communication interface, and a bus; The processor, memory, and communication interface communicate with each other via the bus, the memory stores program instructions executable by the processor, and the processor calls the program instructions to implement the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions enable a computer to implement the method according to any one of claims 1 to 7.

Images