Laser radar-assisted real-time monitoring method and system for power conductor data sag

Abstract

The invention discloses a laser radar-assisted power conductor data sag real-time monitoring method and system, and the method comprises the steps: collecting key working condition data and three-dimensional laser point cloud data, especially wind power data of a power line scene, enabling the wind power to cause the change of the power conductor data in a transverse direction, and affecting the accurate sag calculation; by acquiring the data, a necessary basis can be provided for subsequent sag calculation; further, the three-dimensional laser point cloud data are analyzed through clustering processing in combination with working condition data so as to identify a catenary equation y of the power conductor data, and the key of the step is to determine the shape of the conductor, so that a foundation is laid for accurate calculation of the sag; and finally, based on the identified catenary equation, the sag f of the power conductor data is further calculated, and the sag is accurately calculated through the maximum value and the minimum value of the catenary equation, so that the stability and the safety of the power conductor data under wind power and other environmental factors are ensured.

Description

Technical Field

[0001] The present invention relates to the field of power conductor data sag monitoring, and in particular to a laser radar-assisted power conductor data sag real-time monitoring method and system. Background Art

[0002] In the operation and maintenance of power systems, the shape and state of power conductors are crucial to the safety and stability of the system. Power conductors are affected by many factors during operation, such as wind. These factors can cause changes in the shape of the conductors, especially the bending (catenary shape) and sag (sag) of the conductors, which may pose potential risks to the operation of the power system. Therefore, monitoring and accurately calculating the actual state of power conductors, especially their catenary equations and sag characteristics, has become an important task in power engineering.

[0003] Traditional sag monitoring methods often fail to collect and integrate environmental data in real time, resulting in an inability to fully analyze the stress conditions of power conductors. For example, factors such as wind speed and wind direction have a significant impact on the shape of power conductors, and these factors have not been effectively integrated in many existing technologies. Summary of the invention

[0004] The purpose of the present invention is to provide a laser radar-assisted power conductor data sag real-time monitoring method and system, which solves the above-mentioned technical problems pointed out in the prior art.

[0005] The present invention provides a laser radar-assisted power conductor data sag real-time monitoring method, comprising the following operating steps:

[0006] Real-time collection of working condition data and three-dimensional laser point cloud data of power line scenes;

[0007] Based on the three-dimensional laser point cloud data, the catenary equation y of the power conductor data is identified by clustering the data and analyzing the data in combination with the working condition data;

[0008] Calculate the power conductor data sag f based on the catenary equation;

[0009] Preferably, the operating condition data includes wind power data of the current power line scene;

[0010] The wind force data includes wind speed data and wind direction angle data.

[0011] Preferably, the calculation method of the power conductor data sag f is:

[0012] f=y max -y min ;

[0013] In the formula, ymax is the maximum value of the catenary equation; y min is the minimum value of the catenary equation.

[0014] Preferably, the catenary equation y of the power conductor data is identified by clustering the three-dimensional laser point cloud data and analyzing it in combination with the working condition data, including the following steps:

[0015] Performing clustering processing based on the three-dimensional laser point cloud data to identify and obtain first power conductor data;

[0016] Calibrate the first power conductor data based on the operating condition data to obtain second power conductor data;

[0017] The catenary equation y is obtained by calculation based on the second power conductor data.

[0018] Preferably, calibrating the first power conductor data based on the operating condition data to obtain the second power conductor data comprises the following steps:

[0019] The operating condition data and the first power conductor data are input into a pre-trained conductor calibration model, and second power conductor data are outputted.

[0020] Preferably, the operating condition data and the first power conductor data are input into a pre-trained conductor calibration model, and the second power conductor data is output, which includes the following steps:

[0021] Inputting the working condition data and the first power conductor data into a pre-trained conductor calibration model, and outputting the coordinates of the lowest point of the second power conductor data;

[0022] The second power conductor data is obtained by calculation based on the coordinates of the lowest point of the second power conductor data.

[0023] Preferably, the training of the wire calibration model includes the following steps:

[0024] Collect and obtain a plurality of historical data; pre-process the historical data and then divide them into a training set and a test set;

[0025] constructing an initial wire calibration model;

[0026] Training the initial conductor calibration model: inputting the historical operating condition data corresponding to the training set into the initial conductor calibration model, and outputting the coordinates (x', y', z') of the lowest point of the training second power conductor data;

[0027] A difference loss function Loss is calculated based on the coordinates (x, y, z) of the lowest point of the historical second power conductor data and the coordinates (x', y', z') of the lowest point of the training second power conductor data;

[0028] Determine whether the difference loss function Loss is less than or equal to a preset maximum threshold of the difference loss function; if so, output the initial wire calibration model as the target wire calibration model to be determined; if not, optimize the weight of the initial wire calibration model using the gradient descent method to obtain a new initial wire calibration model, and return to the above-mentioned "training the initial wire calibration model" operation until the target wire calibration model to be determined is output;

[0029] Verifying the target conductor calibration model to be determined: inputting the historical operating condition data corresponding to the test set into the target conductor calibration model to be determined, and outputting the coordinates (x", y", z") of the lowest point of the verified second power conductor data;

[0030] Based on the lowest point coordinates (x", y", z") of the verified second power conductor data and the lowest point coordinates (x, y, z) of the historical second power conductor data, a verification difference loss function Loss' is calculated; whether the verification difference loss function Loss' is less than or equal to a preset maximum threshold of the final verification loss function is determined; if so, the target conductor calibration model to be determined is output as the target conductor calibration model; if not, the weight of the target conductor calibration model to be determined is optimized by using the gradient descent method to obtain a new target conductor calibration model to be determined, and the above-mentioned "verifying the target conductor calibration model to be determined" operation is returned until the target conductor calibration model is output.

[0031] Preferably, the historical data includes historical operating condition data, historical first power conductor data and the lowest point coordinates (x, y, z) of historical second power conductor data corresponding to the historical operating condition data; the historical operating condition data includes historical wind speed data and historical wind direction angle data.

[0032] Preferably, the initial wire calibration model includes an input layer, a hidden layer and an output layer;

[0033] The input layer includes node 1 and node 2;

[0034] The node 1 is used as an input node for wind speed data and wind direction angle data; the node 2 is used as an input node for the historical first power conductor data;

[0035] The hidden layer uses ReLU as the activation function;

[0036] The output layer includes a first node, a second node and a third node; the output layer uses a linear activation function;

[0037] The first node is used as the output node of the horizontal coordinate x of the lowest point coordinate of the second power conductor data; the second node is used as the output node of the vertical coordinate of the lowest point coordinate of the second power conductor data; and the third node is used as the output node of the vertical coordinate of the lowest point coordinate of the second power conductor data.

[0038] On the other hand, the present invention also provides a laser radar-assisted power conductor data sag real-time monitoring system, including a data acquisition module, an analysis and identification module, and a calculation module;

[0039] The data acquisition module is used to collect working condition data and three-dimensional laser point cloud data of power line scenes in real time;

[0040] The operating condition data includes wind power data of the current power line scene;

[0041] The wind data includes wind speed data and wind direction angle data;

[0042] The analysis and identification module is used to identify the catenary equation y of the power conductor data based on the three-dimensional laser point cloud data through clustering processing and combined with the working condition data for analysis;

[0043] The calculation module is used to calculate the power conductor data sag f based on the catenary equation.

[0044] Compared with the prior art, the embodiments of the present invention have at least the following technical advantages:

[0045] By analyzing the above-mentioned laser radar-assisted real-time monitoring method and system for sag of power conductor data provided by the present invention, it can be known that in specific applications, by collecting key operating condition data and three-dimensional laser point cloud data of the power line scene, especially wind data (wind speed and wind direction angle), because wind has an important influence on the sag calculation of power conductor data, wind will cause the power conductor data to change in the lateral direction, affecting its accurate sag calculation; by obtaining these data, it is possible to provide a necessary basis for subsequent sag calculation; further, by clustering processing and combining the operating condition data to analyze the three-dimensional laser point cloud data to identify the catenary equation y of the power conductor data. The key to this step is to determine the shape of the conductor, thereby laying a foundation for the accurate calculation of sag; finally, based on the identified catenary equation, the sag f of the power conductor data is further calculated, and the sag is accurately calculated by the maximum and minimum values ​​of the catenary equation, thereby ensuring the stability and safety of the power conductor data under wind and other environmental factors. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 A schematic diagram of the overall operation steps of a laser radar-assisted real-time monitoring method for power conductor data sag;

[0047] Figure 2 A schematic diagram of power conductor simulation under windless conditions in a laser radar-assisted power conductor data sag real-time monitoring method;

[0048] Figure 3 A schematic diagram of a power conductor simulation under a crosswind state in a laser radar-assisted power conductor data sag real-time monitoring method;

[0049] Figure 4 A schematic diagram of the operation steps of identifying and obtaining the catenary equation y of power conductor data in a laser radar-assisted real-time monitoring method for sag of power conductor data;

[0050] Figure 5 A schematic diagram of the operation steps of outputting second power conductor data in a laser radar-assisted power conductor data sag real-time monitoring method;

[0051] Figure 6 A schematic diagram of the operation steps of training a conductor calibration model in a laser radar-assisted real-time monitoring method for power conductor data sag;

[0052] Figure 7 The figure is a schematic diagram of the overall architecture of a laser radar-assisted power conductor data sag real-time monitoring system.

[0053] Reference numerals: data acquisition module 10 , analysis and identification module 20 , calculation module 30 . DETAILED DESCRIPTION

[0054] The technical solution of the present invention will be described clearly and completely below in conjunction with the accompanying drawings. 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.

[0055] The present invention is further described in detail below through specific embodiments in conjunction with the accompanying drawings.

[0056] Embodiment 1

[0057] like Figure 1 As shown, the first embodiment of the present invention provides a laser radar-assisted power conductor data sag real-time monitoring method, including the following operating steps:

[0058] Step S10: real-time collection of working condition data and three-dimensional laser point cloud data of the power line scene;

[0059] The operating condition data includes wind power data of the current power line scene;

[0060] The wind data includes wind speed data and wind direction angle data;

[0061] It should be noted that the above-mentioned embodiment of the present application collects wind data (this is because wind data is very important for the sag calculation of power conductor data. Under the influence of wind, the power conductor data collected by the laser radar will cause the line to be erratic due to the influence of lateral wind, thereby causing the sag calculation to be incorrect, such as Figure 2 and Figure 3 The above working condition data and three-dimensional laser point cloud data provide a data basis for the subsequent sag calculation of power conductor data.

[0062] Step S20: performing clustering processing based on the three-dimensional laser point cloud data and analyzing in combination with the working condition data to identify and obtain the catenary equation y of the power conductor data;

[0063] Step S30: Calculating based on the catenary equation to obtain the power conductor data sag f;

[0064] The calculation method of the power conductor data sag f is:

[0065] f=y max -y min ;

[0066] In the formula, y max is the maximum value of the catenary equation; y min is the minimum value of the catenary equation;

[0067] It should be noted that the above-mentioned embodiment of the present application first collects key operating condition data and three-dimensional laser point cloud data of the power line scene, especially wind data (wind speed and wind direction angle), because wind has an important influence on the sag calculation of the power conductor data. Wind will cause the power conductor data to change in the lateral direction, affecting its accurate sag calculation; by obtaining these data, it can provide the necessary basis for the subsequent sag calculation; further, the three-dimensional laser point cloud data is analyzed by clustering processing and combining the operating condition data to identify the catenary equation y of the power conductor data. The key to this step is to determine the shape of the conductor, thereby laying the foundation for the accurate calculation of sag; finally, based on the identified catenary equation, the sag f of the power conductor data is further calculated, and the sag is accurately calculated by the maximum and minimum values ​​of the catenary equation, thereby ensuring the stability and safety of the power conductor data under wind and other environmental factors;

[0068] The above-mentioned embodiments of the present application accurately calculate the sag of power conductor data through real-time operating data and accurate three-dimensional laser point cloud data analysis, thereby improving the safety and reliability of the power system under different environmental conditions, especially in response to the impact of wind changes.

[0069] Specifically, Figure 4 As shown, in step S20, the catenary equation y of the power conductor data is identified based on the clustering processing of the three-dimensional laser point cloud data and analyzed in combination with the working condition data, including the following operation steps:

[0070] Step S21: performing clustering processing based on the three-dimensional laser point cloud data to identify and obtain first power conductor data;

[0071] It should be noted that the above embodiment of the present application first pre-processes the three-dimensional laser point cloud data, removes the noise in the point cloud through pre-processing, and makes the point cloud smoother, which is convenient for subsequent processing and analysis; further, the pre-processed point cloud data is clustered to obtain multiple clusters, and the pre-processed three-dimensional laser point cloud data belonging to the power conductor data line (i.e., the above-mentioned first power conductor data) is screened according to the number of point clouds in the clusters and the size of the bounding box of the clusters;

[0072] The technical solution adopted in the above-mentioned embodiment of the present application is intended to screen out the three-dimensional laser point cloud data of the power conductor data, screen out the obstacle point clouds such as poles and trees in the scene, and provide a data basis for subsequent further processing and analysis;

[0073] Step S22: calibrating the first power conductor data based on the operating condition data to obtain second power conductor data;

[0074] Step S23: Calculating and obtaining a catenary equation y based on the second power conductor data;

[0075] It should be noted that, in the above-mentioned embodiment of the present application, the specific position and form of the power conductor data are firstly identified by using the three-dimensional laser point cloud data, and the accuracy of the subsequent data processing is ensured by removing the noise in the point cloud, and the discreteness of the point cloud is reduced by smoothing the point cloud, so that the data is more coherent and easier to process; this lays the foundation for the subsequent cluster analysis and identification of the power conductor data;

[0076] Furthermore, the preprocessed point cloud data is divided into multiple clusters (clustering clusters) by a clustering algorithm. The point clouds within the clusters have a high spatial correlation, so that the point clouds belonging to the power line data can be more easily identified; the power line data is screened according to the number of point clouds in each cluster and the size of the bounding box of the cluster. The power line data usually has a certain degree of coherence in space, so the number of point clouds in the cluster and the spatial size of the cluster can help distinguish the power line data from other non-target objects (such as poles, trees, etc.);

[0077] By screening out the three-dimensional laser point cloud data belonging to the power conductor data and excluding other obstacles in the scene (such as poles, trees, etc.), an accurate data basis is provided for subsequent analysis, which is a prerequisite for the subsequent steps to accurately calculate the form of the power conductor data (such as the catenary equation); further, the first power conductor data is further calibrated according to the working condition data (which may refer to environmental conditions, the operating status of the power conductor data, etc.) to obtain the second power conductor data, and the power conductor data (i.e., the first power conductor data) identified by the three-dimensional laser point cloud data is matched with the power conductor data model under the actual operating state, because in reality The shape of the power conductor data may be deformed (such as sagging or bending) due to factors such as wind force, so it is necessary to adjust the model through the operating condition data to more accurately reflect the actual situation, and obtain a more accurate power conductor data model that conforms to the actual operating status (i.e., the second power conductor data); finally, further calculations are performed based on the model of the second power conductor data to obtain the catenary equation y; by combining the three-dimensional laser point cloud data, the operating condition data and the catenary equation, an efficient and accurate power conductor data recognition and analysis method is provided, which can effectively identify the power conductor data through precise preprocessing, clustering and calibration.

[0078] Specifically, in step S22, the first power conductor data is calibrated based on the operating condition data to obtain second power conductor data, including the following operation steps:

[0079] Step S221: inputting the working condition data and the first power conductor data into a pre-trained conductor calibration model, and outputting second power conductor data;

[0080] It should be noted that the technical solution adopted in the above-mentioned embodiment of the present application provides a more accurate prediction of the conductor status by using a conductor calibration model, which helps to calculate and output the second power conductor data under strong winds (especially crosswinds), thereby facilitating the subsequent calculation of the catenary equation y.

[0081] Specifically, Figure 5 As shown, in step S221, the operating condition data and the first power conductor data are input into a pre-trained conductor calibration model, and the second power conductor data is output, which includes the following operation steps:

[0082] Step S2211: inputting the working condition data and the first power conductor data into a pre-trained conductor calibration model, and outputting the coordinates of the lowest point of the second power conductor data;

[0083] Step S2212: Calculating and obtaining second power conductor data based on the coordinates of the lowest point of the second power conductor data;

[0084] It should be noted that the above-mentioned embodiment of the present application combines the operating condition data and the first power conductor data, and uses the pre-trained conductor calibration model to output more accurate coordinates of the lowest point of the conductor, thereby calculating more accurate conductor data, which can effectively improve the accuracy, reliability and real-time performance of the calibration of the offset of the power conductor data caused by wind (especially lateral wind).

[0085] Specifically, Figure 6 As shown, in step S2211, the training of the wire calibration model includes the following steps:

[0086] Step S22111: collecting and acquiring a plurality of historical data; preprocessing the historical data and dividing the data into a training set and a test set;

[0087] The historical data includes historical operating condition data, historical first power conductor data and the lowest point coordinates (x, y, z) of historical second power conductor data corresponding to the historical operating condition data; the historical operating condition data includes historical wind speed data and historical wind direction angle data;

[0088] It should be noted that the above-mentioned preprocessing refers to the preprocessing operation of standardizing and normalizing historical data, which aims to eliminate the dimensional differences between different features and improve the model training effect.

[0089] Step S22112: constructing an initial wire calibration model;

[0090] The initial wire calibration model includes an input layer, a hidden layer and an output layer;

[0091] The input layer includes node 1 and node 2;

[0092] The node 1 is used as an input node for wind speed data and wind direction angle data; the node 2 is used as an input node for the historical first power conductor data;

[0093] The hidden layer uses ReLU as the activation function;

[0094] The output layer includes a first node, a second node and a third node; the output layer uses a linear activation function;

[0095] The first node is used as an output node of the abscissa x of the coordinate of the lowest point of the second power conductor data; the second node is used as an output node of the ordinate of the coordinate of the lowest point of the second power conductor data; the third node is used as an output node of the vertical coordinate of the coordinate of the lowest point of the second power conductor data;

[0096] It should be noted that the above-mentioned initial conductor calibration model inputs the above-mentioned historical wind speed data, historical wind direction angle data and historical first power conductor data into the initial conductor calibration model as the input layer, and then calculates the activation function of the hidden layer ReLU, and then calculates the linear activation function of the output layer to output the lowest point coordinates (x, y, z) of the historical power conductor data.

[0097] Step S22113: training the initial conductor calibration model: inputting the historical operating condition data corresponding to the training set into the initial conductor calibration model, and outputting the coordinates (x', y', z') of the lowest point of the training second power conductor data;

[0098] Step S22114: Calculate the difference loss function Loss based on the coordinates (x, y, z) of the lowest point of the historical second power conductor data and the coordinates (x', y', z') of the lowest point of the training second power conductor data;

[0099] The difference loss function Loss is calculated using a mean square error function;

[0100] Step S22115: determine whether the difference loss function Loss is less than or equal to a preset maximum threshold of the difference loss function; if so, output the initial wire calibration model as the target wire calibration model to be determined; if not, optimize the weight of the initial wire calibration model using the gradient descent method to obtain a new initial wire calibration model, and return to the "training the initial wire calibration model" operation of the above step S22113 until the target wire calibration model to be determined is output;

[0101] Step S22116: verifying the calibration model of the target conductor to be determined: inputting the historical operating condition data corresponding to the test set into the calibration model of the target conductor to be determined, and outputting the coordinates (x", y", z") of the lowest point of the verified second power conductor data;

[0102] Step S22117: Calculate and verify the difference loss function Loss' based on the lowest point coordinates (x", y", z") of the verified second power conductor data and the lowest point coordinates (x, y, z) of the historical second power conductor data; determine whether the verification difference loss function Loss' is less than or equal to the preset maximum threshold of the final verification loss function; if so, output the target conductor calibration model to be determined as the target conductor calibration model; if not, optimize the weight of the target conductor calibration model to be determined by using the gradient descent method to obtain a new target conductor calibration model to be determined, and return to the "verify the target conductor calibration model to be determined" operation of the above step S22116 until the target conductor calibration model is output;

[0103] The verification difference loss function Loss' is calculated using a mean square error function.

[0104] It should be noted that the above-mentioned embodiment of the present application is optimized through multiple training, loss calculation, and gradient descent method, and the final model can effectively capture the complex relationship between wind speed, wind direction angle, the position of the first power conductor data and the coordinates of the lowest point of the second power conductor data. Each optimization improves the prediction accuracy of the model and finally reaches the target accuracy level; by introducing a multi-layer neural network (multi-layer perceptron) and a nonlinear activation function (ReLU), the nonlinear relationship can be effectively processed, which greatly improves the accuracy of the model predicting the coordinates (x, y, z) of the lowest point of the second power conductor data; the above-mentioned embodiment of the present application adopts standardization and normalization processing to ensure the consistency of the input data in training and avoid the influence of features of different scales on the model training effect; the use of the gradient descent method can make the model gradually converge in different training cycles, prevent overfitting, and improve the stability of the model; by separating the training set and the test set, and by monitoring the loss function during the model verification process, the generalization ability of the model in practical applications is ensured, and accurate predictions can be made under different working conditions;

[0105] The above-mentioned embodiment of the present application realizes efficient and accurate prediction of the coordinates of the lowest point of the power conductor through data collection and preprocessing, neural network model construction, training, verification and optimization. Through multiple training and optimization, the model can accurately reflect the influence of factors such as wind speed, wind direction angle, and the position of the first power conductor on the coordinates of the lowest point of the second power conductor, and has strong accuracy, robustness and generalization ability.

[0106] Based on the same concept of the above method embodiments, the embodiments of the present invention also provide a laser radar-assisted power conductor data sag real-time monitoring system for implementing the above method of the present invention. Since the principles and methods of solving the problems in the system embodiments are similar, it has at least all the beneficial effects brought by the technical solutions of the above embodiments, which will not be described one by one here.

[0107] Embodiment 2

[0108] like Figure 7 As shown, the second embodiment of the present invention provides a laser radar-assisted power conductor data sag real-time monitoring system, including a data acquisition module 10, an analysis and identification module 20, and a calculation module 30;

[0109] The data acquisition module 10 is used to collect working condition data and three-dimensional laser point cloud data of power line scenes in real time;

[0110] The operating condition data includes wind power data of the current power line scene;

[0111] The wind data includes wind speed data and wind direction angle data;

[0112] The analysis and identification module 20 is used to identify the catenary equation y of the power conductor data based on the three-dimensional laser point cloud data through clustering processing and combined with the working condition data for analysis;

[0113] The calculation module 30 is used to calculate the power conductor data sag f based on the catenary equation;

[0114] The calculation method of the power conductor data sag f is:

[0115] f=y max -y min ;

[0116] In the formula, y max is the maximum value of the catenary equation; y min is the minimum value of the catenary equation;

[0117] In summary, a laser radar-assisted real-time monitoring method and system for sag of power conductor data proposed in an example of the present invention first collects key operating condition data and three-dimensional laser point cloud data of the power line scene, especially wind data (wind speed and wind direction angle), because wind has an important influence on the sag calculation of power conductor data. Wind will cause the power conductor data to change in the lateral direction, affecting its accurate sag calculation; by obtaining these data, it can provide the necessary basis for subsequent sag calculation; further, the three-dimensional laser point cloud data is analyzed by clustering processing and combining the operating condition data to identify the catenary equation y of the power conductor data. The key to this step is to determine the shape of the conductor, thereby laying the foundation for the accurate calculation of sag; finally, based on the identified catenary equation, the sag f of the power conductor data is further calculated, and the sag is accurately calculated by the maximum and minimum values ​​of the catenary equation, thereby ensuring the stability and safety of the power conductor data under wind and other environmental factors.

[0118] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. A person skilled in the art may modify the technical solutions described in the above embodiments, or replace part or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A laser radar-assisted real-time monitoring method for power conductor data sag, characterized in that: The steps are as follows: Real-time collection of working condition data and three-dimensional laser point cloud data of power line scenes; Based on the three-dimensional laser point cloud data, the catenary equation y of the power conductor data is identified by clustering the data and analyzing the data in combination with the working condition data; The power conductor data sag f is calculated based on the catenary equation.

2. A laser radar-assisted real-time monitoring method for power conductor data sag according to claim 1, characterized in that: The operating condition data includes wind power data of the current power line scene; The wind force data includes wind speed data and wind direction angle data.

3. The laser radar-assisted real-time monitoring method for power conductor data sag according to claim 2 is characterized in that: The calculation method of the power conductor data sag f is: f=y max -y min ; In the formula, y max is the maximum value of the catenary equation; y min is the minimum value of the catenary equation.

4. The method for real-time monitoring of power conductor data sag assisted by laser radar according to claim 3 is characterized in that: Based on the three-dimensional laser point cloud data, the catenary equation y of the power conductor data is identified by clustering the data and analyzing the data in combination with the working condition data, including the following steps: Performing clustering processing based on the three-dimensional laser point cloud data to identify and obtain first power conductor data; Calibrate the first power conductor data based on the operating condition data to obtain second power conductor data; The catenary equation y is obtained by calculation based on the second power conductor data.

5. The laser radar-assisted real-time monitoring method for power conductor data sag according to claim 4 is characterized in that: The first power conductor data is calibrated based on the operating condition data to obtain second power conductor data, including the following steps: The operating condition data and the first power conductor data are input into a pre-trained conductor calibration model, and second power conductor data are outputted.

6. A laser radar-assisted real-time monitoring method for power conductor data sag according to claim 5, characterized in that: Inputting the operating condition data and the first power conductor data into a pre-trained conductor calibration model, and outputting second power conductor data, comprises the following steps: Inputting the working condition data and the first power conductor data into a pre-trained conductor calibration model, and outputting the coordinates of the lowest point of the second power conductor data; The second power conductor data is obtained by calculation based on the coordinates of the lowest point of the second power conductor data.

7. A laser radar-assisted real-time monitoring method for power conductor data sag according to claim 6, characterized in that: The training of the wire calibration model includes the following steps: Collect and obtain a plurality of historical data; pre-process the historical data and then divide them into a training set and a test set; constructing an initial wire calibration model; Training the initial conductor calibration model: inputting the historical operating condition data corresponding to the training set into the initial conductor calibration model, and outputting the coordinates (x', y', z') of the lowest point of the training second power conductor data; A difference loss function Loss is calculated based on the coordinates (x, y, z) of the lowest point of the historical second power conductor data and the coordinates (x', y', z') of the lowest point of the training second power conductor data; Determine whether the difference loss function Loss is less than or equal to a preset difference loss function maximum threshold; If yes, the initial wire calibration model is output as the target wire calibration model to be determined; if no, the weight of the initial wire calibration model is optimized by using the gradient descent method to obtain a new initial wire calibration model, and the above-mentioned "training the initial wire calibration model" operation is returned until the target wire calibration model to be determined is output; Verifying the target conductor calibration model to be determined: inputting the historical operating condition data corresponding to the test set into the target conductor calibration model to be determined, and outputting the coordinates (x", y", z") of the lowest point of the verified second power conductor data; Calculate and verify the difference loss function Loss' based on the lowest point coordinate (x", y", z") of the verified second power conductor data and the lowest point coordinate (x, y, z) of the historical second power conductor data; Determine whether the verification difference loss function Loss' is less than or equal to a preset maximum threshold of the final verification loss function; If so, the target wire calibration model to be determined is output as the target wire calibration model; if not, the weight of the target wire calibration model to be determined is optimized by using the gradient descent method to obtain a new target wire calibration model to be determined, and return to the above-mentioned "verifying the target wire calibration model to be determined" operation until the target wire calibration model is output.

8. A laser radar-assisted real-time monitoring method for power conductor data sag according to claim 7, characterized in that: The historical data includes historical operating condition data, historical first power conductor data and the lowest point coordinates (x, y, z) of historical second power conductor data corresponding to the historical operating condition data; the historical operating condition data includes historical wind speed data and historical wind direction angle data.

9. A laser radar-assisted real-time monitoring method for power conductor data sag according to claim 8, characterized in that: The initial wire calibration model includes an input layer, a hidden layer and an output layer; The input layer includes node 1 and node 2; The node 1 is used as an input node for wind speed data and wind direction angle data; the node 2 is used as an input node for the historical first power conductor data; The hidden layer uses ReLU as the activation function; The output layer includes a first node, a second node and a third node; the output layer uses a linear activation function; The first node is used as an output node of the horizontal coordinate x of the coordinate of the lowest point of the second power conductor data; The second node is used as an output node of the ordinate of the lowest point coordinate of the second power conductor data; The third node serves as an output node of the vertical coordinate of the lowest point coordinate of the second power conductor data.

10. A laser radar-assisted power conductor data sag real-time monitoring system, characterized in that: It includes data acquisition module, analysis and identification module, and calculation module; The data acquisition module is used to collect working condition data and three-dimensional laser point cloud data of power line scenes in real time; The operating condition data includes wind power data of the current power line scene; The wind data includes wind speed data and wind direction angle data; The analysis and identification module is used to identify the catenary equation y of the power conductor data based on the three-dimensional laser point cloud data through clustering processing and combined with the working condition data for analysis; The calculation module is used to calculate the power conductor data sag f based on the catenary equation.

Images