Power transmission line sag detection method based on synthetic aperture radar image

Abstract

The invention relates to the technical field of sag detection, and discloses a power transmission line sag detection method based on a synthetic aperture radar image, and the method comprises the following steps: S101, judging that the current wind speed is smaller than a threshold value, carrying out the calculation to obtain an ice-free and wind-free specific load and an ice-coated wind-free specific load, or else, entering S104; s102, obtaining a lead stress value through a stress detection model; step S103, calculating to obtain an ice-free and wind-free sag and an icing wind-free sag, and entering step S105; step S104, obtaining an ice-free wind sag and an ice-coated wind sag through the sag detection model; and S105, judging whether a grounding risk exists or not. According to the method, the ice-free wind-free sag and the icing wind-free sag are obtained by combining the stress detection model and the target detection model with the empirical formula, and the change trend of the conductor sag in the synthetic aperture radar image along with time development is dynamically tracked through the sag detection model, so that all-weather and real-time power transmission line sag detection is realized.

Description

Technical Field

[0001] The invention relates to the technical field of sag detection, and more specifically, to a transmission line sag detection method based on synthetic aperture radar images. Background Art

[0002] As an important part of the power system, transmission lines undertake the critical task of transmitting electricity over long distances. Under the influence of their own weight and environmental loads, the conductors will form a suspension curve between the towers. The sag refers to the vertical distance between the lowest point of the conductor and the connection point of the tower. Too much sag means that there is a risk of contact between the conductor and the ground or vegetation, which may cause accidents such as conductor grounding and short circuit. Too little sag means that the conductor tension is too high, which may cause the conductor to break or damage related equipment.

[0003] Traditional transmission line sag detection mainly relies on manual inspection, that is, using simple tools such as hanging line distance meter or sag ruler, combined with empirical formula to calculate the sag size. However, manual inspection is inefficient, especially in complex terrain, where inspection is extremely difficult. In addition, the empirical formula is approximately based on the parabola assumption and is suitable for situations where the curvature of the suspension curve is small, so the accuracy of sag detection is low.

[0004] With the development of drone technology and high-performance cameras, drones are now used to fly along transmission lines to capture high-definition images, and then the outline of the conductor is extracted through image processing technology (such as edge detection algorithm, Hough line detection algorithm, etc.), and the sag size and the lowest point position are calculated through catenary or parabola fitting. Although the above scheme has improved the detection efficiency and accuracy to a certain extent, drone aerial photography is greatly affected by weather, making it difficult to achieve all-weather, real-time transmission line sag detection. Summary of the invention

[0005] The present invention provides a transmission line sag detection method based on synthetic aperture radar images, which solves the technical problems in the above-mentioned background technology.

[0006] The present invention provides a transmission line sag detection method based on synthetic aperture radar images, comprising the following steps: Step S101, if the current wind speed is less than a preset wind speed threshold, then the ice-free wind-free load ratio and the ice-covered wind-free load ratio are calculated according to the first conductor parameter, otherwise, the process goes to step S104; The first conductor parameters include: conductor mass per unit length, cross-sectional area and diameter; Step S102, obtaining meteorological data and second conductor parameters, performing normalization processing and inputting them into the trained stress detection model to obtain the conductor stress value; Meteorological data include: temperature, rainfall, ice thickness, wind speed and the angle between wind direction and the line connecting the towers; The second wire parameters include: the first wire parameter, the wire expansion coefficient, and the elastic coefficient; In step S103, output the synthetic aperture radar image to the target detection model to obtain the basic tower parameters, and calculate the ice-free and windless sag and the ice-covered and windless sag respectively according to the wire stress value, the specific load without ice and wind, and the specific load with ice and without wind, and then enter step S105; The basic tower parameters include: the span, the height difference angle, and the distance from the detection point to the smaller side; In step S104, collect synthetic aperture radar images within a preset time period, construct them into an image sequence and input it into the trained sag detection model to obtain the ice-free and windy sag and the ice-covered and windy sag; In step S105, calculate the height above the ground according to the ice-free and windless sag, the ice-covered and windless sag, the ice-free and windy sag, or the ice-covered and windy sag in combination with the ground elevation, and judge whether there is a risk of touching the ground according to the preset safety distance threshold.

[0007] Further, the preset wind speed threshold has the following calculation formula: , where represents the preset reference wind speed, represents the height of the wire suspension point on the smaller side, represents the preset reference height, represents the terrain factor; the length of the image sequence is equal to the ratio of the preset time period to the preset acquisition time interval; among them, the preset reference wind speed, the preset reference height, the terrain factor, the preset time period, the preset acquisition time interval, and the preset safety distance threshold are all user-defined parameters.

[0008] Further, the specific load without ice and wind has the following calculation formula: , where represents the mass per unit length of the wire, represents the cross-sectional area of the wire, represents the acceleration due to gravity.

[0009] Further, the specific load with ice and without wind has the following calculation formula: , where represents the mass per unit length of the wire, represents the cross-sectional area of the wire, represents the acceleration due to gravity, represents the wire diameter, represents the ice thickness, represents the ice density.

[0010] Further, the stress detection model consists of two branches; The first branch is used to obtain the first stress value through matrix operation of the normalized meteorological data and the second wire parameter; The first stress value The calculation formula is as follows: where , , represents the combined vector formed by splicing the normalized meteorological data and the second wire parameter, represents the weight parameter between the i-th dimension value and the j-th dimension value of the combined vector, represents the first weight parameter, represents the second weight parameter, represents the exponential function with the natural constant e as the base, represents the transpose operation; The second branch is used to obtain the second stress value through weighted summation operation of the normalized meteorological data and the second wire parameter; The first stress value and the second stress value are subjected to weighted summation operation to obtain the wire stress value; where the weight coefficients corresponding to the first stress value and the second stress value are custom parameters with a total value of 1, and the sample labels of the training samples of the stress detection model are collected through a fiber Bragg grating stress sensor.

[0011] Furthermore, the object detection model is constructed based on the YOLOV8 model, and the sample labels of the training samples of the object detection model are obtained through manual annotation.

[0012] Furthermore, the calculation formulas for the ice-free and windless sag and the ice-covered and windless sag are the same, where the ice-free and windless sag The calculation formula is as follows: where represents the ice-free and windless specific load, represents the wire stress value, represents the span, represents the distance from the detection point to the smaller side, represents the height difference angle.

[0013] Furthermore, the sag detection model consists of N hidden layers, N is the same as the length of the image sequence, and each hidden layer consists of a first unit and a second unit; The first unit of the n-th hidden layer inputs the synthetic aperture radar image corresponding to the n-th sequence unit of the image sequence and outputs the first feature vector, 1 ≤ n ≤ N; The second unit of the n-th hidden layer inputs the first feature vector output by the first unit of the n-th hidden layer and outputs the second feature vector; The second feature vector output by the second unit of the Nth hidden layer is input into the first classifier and the second classifier, and the class spaces of the two respectively represent the ice-free and windy sag and the ice-covered and windy sag; The sample labels of the training samples of the sag detection model are obtained by a hand-held laser rangefinder.

[0014] Further, the first unit is constructed based on a sliding window type self-attention transformer, and the second unit is constructed based on a long short-term memory recurrent network unit.

[0015] Further, the height above the ground The calculation formula is as follows: , where and respectively represent the conductor suspension heights on the small-side and large-side, represents the span, the distance from the detection point to the small-side, represents the ground elevation below the detection point, represents one of the ice-free and windless sag, ice-covered and windless sag, ice-free and windy sag, or ice-covered and windy sag.

[0016] The beneficial effects of the present invention are as follows: The present invention divides the detection environment into 4 situations, namely ice-free and windless conditions, ice-covered and windless conditions, ice-free and windy conditions, and ice-covered and windy conditions. The present invention first obtains the conductor stress value through a stress detection model according to meteorological data and conductor parameters, then identifies the basic parameters of the tower pole in the synthetic aperture radar image through an object detection model, and finally obtains the ice-free and windless sag and the ice-covered and windless sag in combination with an empirical formula. And the present invention performs feature extraction and time series analysis on the image sequence of the synthetic aperture radar image through a sag detection model, dynamically tracks the change trend of the sag over time, so as to realize all-weather and real-time transmission line sag detection. Description of the Drawings

[0017] Figure 1 is a flowchart of the transmission line sag detection method based on synthetic aperture radar images of the present invention. Detailed Embodiments

[0018] Now, the subject matter described herein will be discussed with reference to example embodiments. It should be understood that discussing these embodiments is only to enable those skilled in the art to better understand and thus implement the subject matter described herein, and the functions and arrangements of the elements discussed can be changed without departing from the scope of protection of the content of this specification. Each example can omit, substitute, or add various processes or components as needed. Additionally, the features described relative to some examples can also be combined in other examples.

[0019] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in one or more embodiments of the present invention should have the ordinary meaning understood by those of ordinary skill in the field to which the present invention pertains. The terms "first", "second" and similar words used in one or more embodiments of the present invention do not denote any order, quantity or importance, but are only used to distinguish different components. Words such as "including" or "comprising" mean that the elements or objects appearing before this word cover the elements or objects listed after this word and their equivalents, without excluding other elements or objects. Words such as "connected" or "coupled" are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Upper", "lower", "left", "right", etc. are only used to represent relative position relationships. When the absolute position of the object being described changes, the relative position relationship may also change accordingly.

[0020] As Figure 1 shown, the method for detecting the sag of a transmission line based on a synthetic aperture radar image includes the following steps: It should be noted that the calculation formula of the empirical formula mentioned in the background technology is as follows: , where represents the sag (unit: m), represents the self-weight per unit length of the wire (unit: N / m), which can be obtained by referring to the engineering design archives according to the wire model, represents the span, that is, the horizontal projection distance between the tower connection points (unit: m), which can be obtained by referring to the engineering design archives according to the tower number, represents the wire tension (unit: N), which can be obtained by referring to the standard value of the wire tension under normal temperature and conventional loads specified in the engineering design archives or by empirical estimation; the empirical formula is not only approximately based on the parabola hypothesis, but also easily ignores the influence of environmental factors such as temperature, ice coating and wind load, resulting in low accuracy of calculating the sag by the empirical formula.

[0021] Step S101, determine whether the current wind speed is less than the preset wind speed threshold. If so, calculate the no-ice and no-wind specific load and the ice-coated and no-wind specific load respectively according to the first wire parameters, otherwise enter step S104; The first wire parameters include: the mass per unit length of the wire, the cross-sectional area and the diameter; Step S102, obtain meteorological data and the second wire parameters, and perform normalization processing and input them into the trained stress detection model to obtain the wire stress value; The meteorological data includes: temperature, rainfall, ice coating thickness, wind speed and the angle between the wind direction and the tower connection line; The second wire parameters include: the first wire parameters, the wire expansion coefficient and the elastic coefficient; Step S103: Output the synthetic aperture radar image to the target detection model to obtain the basic parameters of the tower pole, and calculate the ice-free and windless sag and the ice-covered and windless sag respectively based on the wire stress value, the ice-free and windless specific load, and the ice-covered and windless specific load, and then enter Step S105; The basic parameters of the tower pole include: span, height difference angle, and distance from the detection point to the smaller side; Step S104: Collect synthetic aperture radar images within a preset time period, construct them into an image sequence, and input the image sequence into the trained sag detection model to obtain the ice-free and windy sag and the ice-covered and windy sag; Step S105: Calculate the height from the ground based on the ice-free and windless sag, the ice-covered and windless sag, the ice-free and windy sag, or the ice-covered and windy sag in combination with the ground elevation, and determine whether there is a risk of touching the ground according to the preset safety distance threshold.

[0022] In an embodiment of the present invention, the preset wind speed threshold is calculated as follows: , where represents the preset reference wind speed, represents the height of the wire suspension point on the smaller side, represents the preset reference height, represents the terrain factor; the length of the image sequence is equal to the ratio of the preset time period to the preset acquisition time interval; among them, the preset reference wind speed, the preset reference height, the terrain factor, the preset time period, the preset acquisition time interval, and the preset safety distance threshold are all custom parameters; preferably, the preset reference wind speed is set to 5 m / s, the preset reference height is set to 10 m, the terrain factor is set according to the terrain, for example, it is set to 0.16 in the plain area, 0.22 in the hilly area, 0.30 in the urban area, the preset time period is set to 30 min, the preset acquisition time interval is set to 1 min, then the length of the image sequence is equal to 30, and the preset safety distance threshold is set to 6 m.

[0023] For example, the preset reference wind speed is set to 5 m / s, the preset reference height is set to 10 m, the terrain factor is set to 0.30, and the height of the wire suspension point on the smaller side is 8 m, then the preset wind speed threshold is approximately equal to 4.68 m / s.

[0024] In an embodiment of the present invention, the ice-free and windless specific load is calculated as follows: , where represents the mass per unit length of the wire (unit: kg / km), represents the cross-sectional area of the wire (unit: mm²), represents the acceleration due to gravity (unit: m / s², assigned as 9.8 m / s²), and the unit of the ice-free and windless specific load is Pa / m.

[0025] In an embodiment of the present invention, the specific gravity of ice-covered conductors without wind is calculated as follows: , where represents the mass per unit length of the conductor (unit: kg / km), represents the cross-sectional area of the conductor (unit: mm²), represents the acceleration due to gravity (unit: m / s², assigned as 9.8 m / s²), represents the diameter of the conductor (unit: mm), represents the ice thickness (unit: mm), represents the ice density (unit: kg / m³, assigned as 900 kg / m³), and the unit of the specific gravity of ice-covered conductors without wind is Pa / m.

[0026] In an embodiment of the present invention, the stress detection model consists of two branches; The first branch is used to obtain the first stress value through matrix operations on the normalized meteorological data and the second conductor parameters; The first stress value is calculated as follows: , where , , represents the combined vector (with a dimension number of 8) formed by splicing the normalized meteorological data and the second conductor parameters, represents the weight parameter between the i-th dimension value and the j-th dimension value of the combined vector, represents the first weight parameter, represents the second weight parameter, represents the exponential function with the natural constant e as the base, represents the transpose operation; Then, according to the above calculation formula, the product of the transpose of the combined vector and the combined vector results in an 8×8 matrix. Therefore, the first weight parameter needs to be designed as a 1×8 matrix, and the second weight parameter needs to be designed as an 8×1 matrix; The second branch is used to obtain the second stress value through weighted summation operations on the normalized meteorological data and the second conductor parameters; The first stress value and the second stress value are subjected to weighted summation operations to obtain the conductor stress value; Among them, the weight coefficients corresponding to the first stress value and the second stress value are custom parameters with a total value of 1. Preferably, the weight coefficient corresponding to the first stress value is set to 0.8, and the weight coefficient corresponding to the second stress value is set to 0.2.

[0027] In an embodiment of the present invention, the sample label of the training sample of the stress detection model is obtained by collecting through a fiber Bragg grating stress sensor.

[0028] It should be noted that the parameters in the stress detection model (including the first branch weight parameter, bias parameter, and various weight coefficients of the second branch) are all learnable hyperparameters, which are given random initial values (such as normal distribution initialization) at the beginning of training. During the training process, the mean square error between the predicted value output by the stress detection model and the true value corresponding to the sample label of the training sample is used as the loss function, and is updated by backpropagation through a gradient optimizer (such as SGD, Adam, etc.), so as to minimize the loss value obtained by forward calculation through the loss function. The training of the neural network model belongs to conventional technical means and will not be elaborated here.

[0029] In one embodiment of the present invention, the object detection model is constructed based on the YOLOV8 model, and the sample labels of the training samples of the object detection model are obtained through manual annotation.

[0030] In one embodiment of the present invention, a preprocessing layer is added to the object detection model, that is, the input synthetic aperture radar image is filtered and denoised. The filtering and denoising algorithm can be any one of the Lee filtering algorithm, Frost filtering algorithm, and BM3D denoising algorithm; the Neck part of the YOLOV8 model is modified to add a P2 feature layer, where Neck represents the intermediate module connecting the backbone network and the detection head, mainly responsible for fusing feature maps of different scales, and the P2 feature layer represents the shallower and finer-grained feature maps extracted. The preset scale of the Anchor can be adjusted in the configuration file of the YOLOV8 model, such as 8×12, 12×16, 16×20, etc.; a high-resolution feature extractor is also added to the object detection model, and the original CSPDarknet (Cross Stage Partial Convolutional Network) is replaced with Swin Transformer (Sliding Window-based Self-Attention Transformer) as the backbone network to improve the ability of the backbone network to extract high-detail information; in addition, the loss function of the object detection model can be modified to EIoU Loss, and a small object penalty term is added, which will not be elaborated here.

[0031] It should be noted that the detection points refer to the feature sampling points selected at preset intervals along the wire direction for representative detection of the sag and ground clearance status of the wire, such as at intervals of 30m, 50m, etc. Because it is impossible to detect the sag of every point of the wire during actual inspection. In addition, the span can be obtained by referring to the engineering design archives according to the tower pole number, the elevation angle can be obtained using DEM (Digital Elevation Model) data, and the distance from the detection point to the small-side tower can be obtained using the GIS database or high-definition map, which will not be elaborated here.

[0032] In one embodiment of the present invention, the calculation formulas for the ice-free and windless sag and the ice-covered and windless sag are the same, where the ice-free and windless sag has the following calculation formula: , where represents the ice-free and windless specific load, represents the conductor stress value, represents the span, represents the distance from the detection point to the smaller-side tower, represents the elevation angle.

[0033] It should be noted that the unit of the conductor stress value is Pa. The elevation angle represents the angle between the connection line between the tower connection points and the horizontal direction, and the distance from the detection point to the smaller-side tower represents the horizontal projection distance between the detection point and the lower conductor suspension point.

[0034] In one embodiment of the present invention, the sag detection model consists of N hidden layers, where N is the same as the length of the image sequence, and each hidden layer consists of a first unit and a second unit; The first unit of the nth hidden layer inputs the synthetic aperture radar image corresponding to the nth sequence unit of the image sequence and outputs a first feature vector, where 1 ≤ n ≤ N; The second unit of the nth hidden layer inputs the first feature vector output by the first unit of the nth hidden layer and outputs a second feature vector; The second feature vector output by the second unit of the Nth hidden layer is input into the first classifier and the second classifier, and the class spaces of the two respectively represent the ice-free and windy sag and the ice-covered and windy sag.

[0035] In one embodiment of the present invention, the sample labels of the training samples of the sag detection model are obtained by collecting with a handheld laser rangefinder or a laser scanner.

[0036] In one embodiment of the present invention, the first unit is constructed based on Swin Transformer (sliding window self-attention transformer), and the second unit is constructed based on LSTM (long short-term memory recurrent network) unit, or can also be constructed based on GRU (gated recurrent unit) unit, which will not be elaborated here.

[0037] For example, the size of the synthetic aperture radar image input by the first unit is 512×512×C (C represents the number of channels. For example, when C is set to 1, it is a grayscale image). Then it is cut into non-overlapping small blocks (Patches). The size of each Patch is 4×4, so there are a total of 128×128 Patches. Each Patch corresponds to a vector of size 4×4×1 = 16. Then each Patch is encoded into a higher-dimensional embedding vector (such as 96) through an MLP (multi-layer perceptron). At this time, the size of the feature map is 128×128×96. Repeat the above operation to convert the size of the feature map to 16×16×512. Finally, it is converted into the first feature vector through global average pooling, that is, the dimension number of the first feature vector is 512.

[0038] In an embodiment of the present invention, the ground clearance is calculated as follows: , where and represent the wire suspension point heights on the small side and the large side respectively, represents the span, the distance from the detection point to the small side, represents the ground elevation below the detection point, represents one of the ice-free and windless sag, ice-covered and windless sag, ice-free and windy sag, or ice-covered and windy sag.

[0039] It should be noted that a training sample consists of sample data and a sample label. Then the sample data of the sag detection model is a synthetic aperture radar image sequence, and the sample label of the sag detection model is the ice-free and windy sag and the ice-covered and windy sag (which can be obtained by collecting with a hand-held laser rangefinder or a laser scanner). During the training process of the sag detection model, the mean square error is used as the loss function to calculate the deviation between the sag value predicted by the sag detection model at each iteration and the measured value (sample label), and the parameters of the model are updated by backpropagation through the gradient descent optimization algorithm (such as Adam, SGD optimizer, etc.), gradually reducing the deviation calculated by the loss function until the maximum number of iterations is reached or the deviation reaches the preset minimum loss threshold, then the training of the sag detection model is completed.

[0040] It should be noted that before inputting into the sag detection model, the synthetic aperture radar image can also be normalized, denoised, etc. And before training, it can also be pre-trained. The meteorological data is used as the sample label for pre-training of the first unit and the second unit. After pre-training, parameter fine-tuning is performed for the sag detection task, which can accelerate model convergence and improve the feature extraction ability, and to a certain extent, can reduce the number of training samples of the actual training sag detection model, thereby reducing the training difficulty of the model.

[0041] It should be noted that the detection environment of the present invention is divided into four situations, namely ice-free and windless conditions, ice-covered and windless conditions, ice-free and windy conditions, and ice-covered and windy conditions. The present invention first obtains the conductor stress value through the stress detection model according to the meteorological data and conductor parameters, then identifies the basic parameters of the tower pole in the synthetic aperture radar image through the target detection model, and finally obtains the ice-free and windless sag and the ice-covered and windless sag by combining the empirical formula. However, under windy conditions, affected by the wind force, the above empirical formula is no longer applicable. Therefore, the present invention extracts features and performs time-series analysis on the image sequence of the synthetic aperture radar image through the sag detection model, dynamically tracks the change trend of the sag over time, so as to realize all-weather and real-time detection of the transmission line sag.

[0042] It should be noted that the setting of the interval and the threshold size is for the convenience of comparison. Among them, the size of the threshold depends on the amount of sample data and the base quantity set by those skilled in the art for each group of sample data, as long as it does not affect the proportional relationship between the parameters and the quantified values. And the above formulas are all calculations of taking the numerical value after removing the dimension. The formulas are all obtained by software simulation of collecting a large amount of data to get a formula closest to the real situation. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0043] The above has described the embodiments of this embodiment, but this embodiment is not limited to the above specific implementation manners. The above specific implementation manners are only illustrative and not restrictive. Under the inspiration of this embodiment, those of ordinary skill in the art can also make many forms, all of which fall within the protection scope of this embodiment.

Claims

1. A method for detecting the sag of a transmission line based on synthetic aperture radar images, characterized in that, It includes the following steps: In step S101, if it is determined that the current wind speed is less than the preset wind speed threshold, the specific gravity without ice and without wind and the specific gravity with ice and without wind are respectively calculated according to the first wire parameters; otherwise, step S104 is entered; The first wire parameters include: the mass per unit length of the wire, the cross-sectional area, and the diameter; In step S102, meteorological data and second wire parameters are obtained, and after normalization processing, they are input into the stress detection model that has been trained to obtain the wire stress value; The meteorological data includes: temperature, rainfall, ice coating thickness, wind speed, and the angle between the wind direction and the connecting line of the tower pole; The second wire parameters include: the first wire parameters, the wire expansion coefficient, and the elastic coefficient; In step S103, the synthetic aperture radar image is output to the target detection model to obtain the basic tower pole parameters, and the sag without ice and without wind and the sag with ice and without wind are respectively calculated according to the wire stress value, the specific gravity without ice and without wind, and the specific gravity with ice and without wind, and step S105 is entered; The basic tower pole parameters include: span, elevation angle difference, and the distance from the detection point to the small-side; In step S104, synthetic aperture radar images are collected within a preset time period and constructed into an image sequence, which is input into the sag detection model that has been trained to obtain the sag with ice and with wind and the sag without ice and with wind; In step S105, the height from the ground is calculated according to the sag without ice and without wind, the sag with ice and without wind, the sag without ice and with wind, or the sag with ice and with wind in combination with the ground elevation, and it is judged whether there is a risk of touching the ground according to the preset safety distance threshold.

2. The method for detecting the sag of a transmission line based on a synthetic aperture radar image according to claim 1, wherein, Preset wind speed threshold The calculation formula is as follows: , where represents the preset reference wind speed, represents the conductor suspension point height on the small side, represents the preset reference height, represents the terrain factor; the length of the image sequence is equal to the ratio of the preset time period to the preset acquisition time interval; Among them, the preset reference wind speed, the preset reference height, the terrain factor, the preset time period, the preset acquisition time interval, and the preset safety distance threshold are all user-defined parameters.

3. The method for detecting the sag of a transmission line based on a synthetic aperture radar image according to claim 1, wherein, Specific loading without ice and wind The calculation formula is as follows: , where represents the mass per unit length of the conductor, represents the cross-sectional area of the conductor, represents the acceleration of gravity.

4. The method for detecting the sag of a transmission line based on a synthetic aperture radar image according to claim 1, wherein Specific ice weight without wind The calculation formula is as follows: , where represents the mass per unit length of the conductor, represents the cross-sectional area of the conductor, represents the acceleration of gravity, represents the diameter of the conductor, represents the ice thickness, represents the ice density.

5. The method for detecting the sag of a transmission line based on a synthetic aperture radar image according to claim 1, characterized in that The stress detection model consists of two branches; The first branch is used to obtain the first stress value through matrix operation on the normalized meteorological data and second wire parameters; The first stress value The calculation formula is as follows: , where , , represents a combined vector formed by splicing the normalized meteorological data and the second wire parameters, represents the weight parameter between the i-th dimensional value and the j-th dimensional value of the combined vector, represents the first weight parameter, represents the second weight parameter, represents the exponential function with the natural constant e as the base, represents the transpose operation; The second branch is used to obtain the second stress value through weighted summation operation on the normalized meteorological data and second wire parameters; The first stress value and the second stress value are subjected to weighted summation operation to obtain the wire stress value; Among them, the weight coefficients corresponding to the first stress value and the second stress value are user-defined parameters with a total value of 1, and the sample labels of the training samples of the stress detection model are collected through fiber Bragg grating stress sensors.

6. The method for detecting the sag of a transmission line based on a synthetic aperture radar image according to claim 1, wherein The target detection model is constructed based on the YOLOV8 model, and the sample labels of the training samples of the target detection model are obtained through manual annotation.

7. The method for detecting the sag of a transmission line based on a synthetic aperture radar image according to claim 1, characterized in that The calculation formulas for the sag without ice and without wind and the sag with ice and without wind are the same. The calculation formula for the sag without ice and without wind is as follows: , where represents the specific load without ice and wind, represents the conductor stress value, represents the span, represents the distance from the detection point to the smaller side, represents the height difference angle.

8. The method for detecting the sag of a transmission line based on a synthetic aperture radar image according to claim 1, wherein, The sag detection model consists of N hidden layers, where N is the same as the length of the image sequence, and each hidden layer consists of a first unit and a second unit; The first unit of the nth hidden layer inputs the synthetic aperture radar image corresponding to the nth sequence unit of the image sequence and outputs the first feature vector, where 1 ≤ n ≤ N; The second unit of the nth hidden layer inputs the first feature vector output by the first unit of the nth hidden layer and outputs the second feature vector; The second feature vector output by the second unit of the Nth hidden layer is input into the first classifier and the second classifier, and the category spaces of the two respectively represent the sag with ice and with wind and the sag without ice and with wind; The sample labels of the training samples of the sag detection model are obtained through a hand-held laser rangefinder.

9. The method for detecting the sag of a transmission line based on a synthetic aperture radar image according to claim 8, wherein The first unit is constructed based on a sliding window self-attention transformer, and the second unit is constructed based on a long short-term memory recurrent network unit.

10. The method for detecting the sag of a transmission line based on a synthetic aperture radar image according to claim 1, characterized in that, Height above the ground The calculation formula is as follows: , where and represent the wire suspension point heights on the small side and the large side respectively, represents the span, the distance from the detection point to the small side, represents the ground elevation below the detection point, represents one of the ice-free and windless sag, ice-covered and windless sag, ice-free and windy sag, or ice-covered and windy sag.

Images