Power transmission line state monitoring method and device and medium

Abstract

The invention discloses a power transmission line state monitoring method and device and a medium. The problem that manual inspection of a power transmission line is low in efficiency and reliability is solved. A wire contour image, an insulator contour image, wire temperature information and insulator temperature information on a power transmission line are collected through a sensing terminal installed on a power transmission tower. And converting the temperature information into a temperature image, and further performing image edge detection on all physical state images to obtain edge calculation information. And inputting the edge calculation information into a preset image processing deep learning model for processing to obtain a state type of the power transmission line, and outputting a corresponding monitoring result. The state of the power transmission line, including appearance damage or abnormity, is comprehensively reflected, potential fault points are detected through temperature change, data are unified into image types, follow-up processing is facilitated, and the data transmission amount and the calculation burden of a central server are reduced by performing image edge detection preprocessing on the edge side.

Description

Technical Field

[0001] The present application relates to the field of power grids, and in particular to a method, device, and medium for monitoring the status of a transmission line. Background Art

[0002] In recent years, the scale of power transmission system equipment has continued to expand, and society's demand for power supply reliability has also continued to increase. As the key to reliable power supply in the power grid, the normal operation of transmission lines is crucial. Therefore, it is urgent to accelerate the construction of intelligent transmission line inspection systems to achieve comprehensive perception of line operating status.

[0003] Traditional transmission line status monitoring often uses manual methods to identify faults through the human eye, but can only detect some obvious faults, such as insulator bursts, while faults such as line corrosion are difficult to detect.

[0004] It can be seen that how to solve the problem of low efficiency and low reliability of manual inspection of transmission lines is a technical problem that needs to be solved urgently by people in this field. Summary of the Invention

[0005] The purpose of this application is to provide a method, device and medium for monitoring the status of a transmission line to solve the problems of low efficiency and low reliability of manual inspection of transmission lines.

[0006] To solve the above technical problems, the present application provides a method for monitoring the state of a transmission line, which is applied to a sensing terminal installed on each high-voltage transmission tower, comprising:

[0007] Collect conductor contour images, insulator contour images, conductor temperature information, and insulator temperature information on transmission lines;

[0008] Converting the conductor temperature information and the insulator temperature information into a conductor temperature image and an insulator temperature image;

[0009] Performing image edge detection on each physical state image to obtain edge calculation information, wherein the physical state image includes a conductor contour image, an insulator contour image, a conductor temperature image, and an insulator temperature image;

[0010] The edge computing information is input into the preset image processing deep learning model for processing to obtain the status type of the transmission line;

[0011] Output the corresponding monitoring result according to the status type.

[0012] As an optional solution, in the above-mentioned transmission line status monitoring method, converting the conductor temperature information and the insulator temperature information into a conductor temperature image and an insulator temperature image includes:

[0013] performing data preprocessing on the conductor temperature information and the insulator temperature information;

[0014] The pre-processed conductor temperature information and the insulator temperature information are converted into corresponding pixel values ​​using a temperature-pixel mapping algorithm; wherein a mapping relationship is established between a temperature range and a pixel value range;

[0015] A conductor temperature image and an insulator temperature image are generated according to the mapped pixel values.

[0016] As an optional solution, in the above-mentioned transmission line state monitoring method, performing image edge detection on the physical state image to obtain edge computing information includes:

[0017] performing a median filter process on the physical state image to remove image noise;

[0018] Obtaining gradient information of the filtered physical state image through a Sobel operator;

[0019] Processing the gradient information by a non-maximum suppression method to determine edge points in the physical state image;

[0020] The edge points are classified by a dual threshold algorithm to determine first-level edge points and second-level edge points as edge calculation information, wherein the grayscale values ​​of the first-level edge points and the second-level edge points are different.

[0021] As an optional solution, in the above-mentioned transmission line status monitoring method, the step of inputting edge computing information into a preset image processing deep learning model for processing to obtain the status type of the transmission line includes:

[0022] Inputting the edge computing information into an embedding layer of a preset improved VisionTransformer model, wherein the embedding layer includes image block embedding, learnable embedding, and position embedding;

[0023] The output of the embedding layer is input into a Transformer encoder and feature extraction is performed through a sparse attention module. The Transformer encoder is provided with a sparse attention module and several layers of encoding modules, each layer of encoding modules includes a multi-head self-attention layer and a multi-layer perceptron layer;

[0024] The output of the Transformer encoder is classified and predicted through the multi-layer perceptron classification head to obtain the state type of the transmission line.

[0025] As an optional solution, in the above-mentioned transmission line status monitoring method, inputting the edge computing information into the embedding layer of the preset improved VisionTransformer model includes:

[0026] Segmenting the input edge computing information in the embedding layer to obtain image block tokens of fixed size;

[0027] Adding position embedding information to each image block token;

[0028] Obtaining a one-dimensional token sequence by concatenating the image block token and the learnable embedding category vector;

[0029] The one-dimensional token sequence is used as the output result of the embedding layer.

[0030] As an optional solution, in the above-mentioned transmission line status monitoring method, the output of the Transformer encoder is classified and predicted by the multi-layer perceptron classification head to obtain the status type of the transmission line, including:

[0031] Performing dimensionality reduction processing on the feature vector output by the Transformer encoder;

[0032] The eigenvector after dimension reduction is normalized by the Softmax function to obtain the probability distribution of the transmission line state type;

[0033] determining a state type of the transmission line according to the probability distribution;

[0034] The state type of the power transmission line is output.

[0035] As an optional solution, in the above-mentioned transmission line status monitoring method, outputting corresponding monitoring results according to the status type includes:

[0036] Displaying the monitoring results in a visual manner on a display device in a monitoring center;

[0037] Determine whether the state of the transmission line is normal according to the monitoring results,

[0038] When the monitoring result indicates an abnormal state, a fault report is generated, including the fault type, location, and severity.

[0039] To solve the above technical problems, the present application further provides a transmission line status monitoring device, which is applied to a sensing terminal installed on each high-voltage transmission tower, comprising:

[0040] An acquisition module is used to acquire conductor contour images, insulator contour images, conductor temperature information, and insulator temperature information on the transmission line;

[0041] A conversion module, configured to convert the conductor temperature information and the insulator temperature information into a conductor temperature image and an insulator temperature image;

[0042] An image edge detection module is used to perform image edge detection on each physical state image to obtain edge calculation information, wherein the physical state image includes a conductor contour image, an insulator contour image, a conductor temperature image, and an insulator temperature image;

[0043] An analysis and processing module is used to input edge computing information into a preset image processing deep learning model for processing to obtain the status type of the transmission line;

[0044] The output module is used to output corresponding monitoring results according to the status type.

[0045] To solve the above technical problems, the present application also provides a transmission line status monitoring device, comprising:

[0046] memory for storing computer programs;

[0047] A processor is configured to implement the steps of the above-mentioned method for monitoring the state of a power transmission line when executing the computer program.

[0048] In order to solve the above technical problems, the present application also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the above-mentioned transmission line status monitoring method are implemented.

[0049] The transmission line condition monitoring method provided in this application uses a sensing terminal installed on a transmission tower to collect conductor contour images, insulator contour images, conductor temperature information, and insulator temperature information on the transmission line. After preprocessing this information, the temperature information is converted into a temperature image. Furthermore, image edge detection is performed on all physical condition images to obtain edge computing information. Finally, the edge computing information is input into a preset image processing deep learning model for processing, determining the transmission line condition type and outputting the corresponding monitoring results. This application not only collects conductor and insulator contour images but also collects temperature information and converts it into a temperature image. Unifying the data into image types facilitates subsequent processing and can more comprehensively reflect the transmission line condition, including not only external damage or anomalies but also potential fault points detected through temperature changes. By performing image edge detection preprocessing on the edge, the amount of data transmitted and the computational burden on the central server are reduced. This method is applicable to transmission line monitoring in a variety of environmental conditions, improving the efficiency of condition monitoring and addressing the low efficiency and reliability of manual transmission line inspections.

[0050] In addition, the present application also provides a device and a medium corresponding to the above-mentioned transmission line status monitoring method, with the same effect as above. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the embodiments of the present application, the following is a brief introduction to the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0052] Figure 1 A flow chart of a method for monitoring the state of a power transmission line is provided for an embodiment of the present application;

[0053] Figure 2 A schematic diagram of the Vision Transformer model structure is provided for an embodiment of the present application;

[0054] Figure 3 A performance comparison chart of the Vision Transformer model before and after improvement is provided for the embodiment of this application;

[0055] Figure 4 A performance comparison chart of the Vision Transformer model before and after improvement is provided for the embodiment of this application;

[0056] Figure 5 A structural diagram of a power transmission line status monitoring device provided in an embodiment of the present application;

[0057] Figure 6 This is a structural diagram of another transmission line status monitoring device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0058] The following will be combined with the accompanying drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0059] The core of this application is to provide a method, device and medium for monitoring the status of a transmission line.

[0060] In order to enable those skilled in the art to better understand the present application, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0061] To solve the above problems, the present invention provides a method for monitoring the state of a power transmission line, which is applied to a sensing terminal installed on each high-voltage transmission tower, such as Figure 1 Shown, including:

[0062] S11: Collecting conductor contour images, insulator contour images, conductor temperature information, and insulator temperature information on the transmission line;

[0063] S12: Converting the conductor temperature information and the insulator temperature information into a conductor temperature image and an insulator temperature image;

[0064] S13: performing image edge detection on each physical state image to obtain edge calculation information, wherein the physical state image includes a conductor contour image, an insulator contour image, a conductor temperature image, and an insulator temperature image;

[0065] S14: Input the edge computing information into a preset image processing deep learning model for processing to obtain the state type of the transmission line;

[0066] S15: Output corresponding monitoring results according to the status type.

[0067] This embodiment provides a transmission line status monitoring method primarily applicable to real-time status monitoring of high-voltage transmission lines. This method is implemented using sensing terminals installed on high-voltage transmission towers and is applicable to a wide range of transmission line scenarios, including but not limited to urban power grids, remote mountainous areas, and transmission line monitoring in harsh environments. The sensing terminals, installed on high-voltage transmission towers, capture physical status images using image acquisition modules and temperature sensors.

[0068] In step S11, the conductor and insulator contour images captured by the sensing terminal provide information about the transmission line's appearance, while the conductor and insulator temperature information reflects the line's operating status. The collection of this data is the foundation of the entire monitoring system, ensuring the accuracy and integrity of the data source for subsequent processing.

[0069] Under certain specific conditions, such as in extreme weather (such as heavy rain or snow) or at night, the image acquisition module may need to have waterproof, dustproof and night vision functions to ensure the accuracy and reliability of image acquisition.

[0070] Step S12 converts the conductor and insulator temperature information into a temperature image using a temperature-to-pixel mapping algorithm. This temperature image intuitively reflects the temperature distribution of the transmission line and provides a unified format for subsequent image processing, facilitating edge detection and input into deep learning models.

[0071] Control temperature sensors collect conductor and insulator temperature information. These sensors can be thermocouples, thermistors, or infrared thermal imagers. These sensors collect conductor and insulator temperature information in real time and convert it into electrical signals. The collected temperature data is also transmitted via wired or wireless communication modules. To improve data transmission reliability, verification and error correction can be performed during the data transmission process.

[0072] Step S13 performs edge detection on the physical state image to obtain edge calculation information. The Sobel operator is then used to calculate the gradient information of the filtered image to preliminarily extract the image's edge features. The Sobel operator is a commonly used edge detection operator that can calculate the horizontal and vertical gradients of an image. Non-maximum suppression can also be performed on the gradient information to identify edge points in the image. This method compares gradient amplitudes, retaining edge points with local maxima and removing other non-edge points. Edge points can also be classified using a dual-threshold algorithm to distinguish between strong and weak edge points. This dual-threshold algorithm effectively distinguishes the intensity of edge points, avoiding misjudgments.

[0073] Under certain specific conditions, such as when the image noise is large or the edge features are not obvious, a more advanced edge detection algorithm (Canny operator) can be used to improve the accuracy of edge detection.

[0074] Step S14 inputs the edge computing information into a preset image processing deep learning model for processing to obtain the status type of the transmission line. This step performs in-depth analysis on the image information after edge detection to identify the status type of the transmission line.

[0075] The edge computing information is input into the preset improved Vision Transformer model, which includes an embedding layer, a Transformer encoder, and a multi-layer perceptron classification head.

[0076] Figure 2 A schematic diagram of the Vision Transformer model structure is provided for the embodiment of this application, such as Figure 2 In this embodiment, the improved Vision Transformer model consists of an embedding layer, a Transformer encoder, and a multi-layer perceptron classification head (MLP Head).

[0077] Figure 3 A performance comparison chart of the Vision Transformer model before and after improvement is provided for the embodiment of this application; Figure 4 A performance comparison chart of the Vision Transformer model before and after improvement is provided for the embodiment of this application; Figure 3、 4 The improved Vision Transformer model achieves a lower loss function value, close to 0.1, and a higher state analysis accuracy of 97%. This is due to the sparse attention module designed into the improved Vision Transformer model, which retains global classification feature information, further improving the accuracy of transmission line state image analysis and accelerating model convergence.

[0078] Embedding layer: In the embedding layer, the input edge computing information is divided into fixed-size image block tokens, and position embedding information is added to each image block (token). By concatenating the image block tokens with the learnable embedding category vector, a one-dimensional token sequence is obtained.

[0079] Transformer Encoder: The output of the embedding layer is fed into the Transformer Encoder, which then extracts features from the token sequence using the sparse attention module. The sparse attention module can highlight important features and suppress irrelevant ones.

[0080] Multi-layer perceptron classification head: The multi-layer perceptron classification head classifies and predicts the output of the Transformer encoder to obtain the state type of the transmission line.

[0081] Step S15 outputs the applicable environment or conditions for the monitoring results based on the status type. The monitoring results are displayed visually on the monitoring center's display device for real-time viewing by maintenance personnel. For any detected abnormal conditions, a detailed fault report is generated, including information such as the fault type, location, and severity. The monitoring results and fault reports are stored in a cloud database for subsequent data analysis and historical tracing. When an abnormal condition is detected, maintenance personnel are promptly notified via SMS, email, or app push notifications.

[0082] Under certain specific conditions, if further analysis of the monitoring results is required, trend analysis can be performed in combination with historical data to identify potential failure trends.

[0083] The transmission line status monitoring method provided in the embodiments of the present application uses a sensing terminal installed on a transmission tower to collect conductor contour images, insulator contour images, conductor temperature information, and insulator temperature information on the transmission line. After preprocessing, the temperature information is converted into a temperature image. Furthermore, image edge detection is performed on all physical status images to obtain edge computing information. Finally, the edge computing information is input into a preset image processing deep learning model for processing, determining the transmission line status type and outputting the corresponding monitoring results. This application not only collects conductor and insulator contour images but also collects temperature information and converts it into a temperature image. Unifying the data into image types facilitates subsequent processing and can more comprehensively reflect the status of the transmission line, including not only external damage or anomalies but also potential fault points detected through temperature changes. By performing image edge detection preprocessing on the edge, the amount of data transmitted and the computational burden on the central server are reduced. This method is applicable to transmission line monitoring in a variety of environmental conditions, improves the efficiency of status monitoring, and addresses the low efficiency and reliability of manual transmission line inspections.

[0084] Furthermore, converting the conductor temperature information and the insulator temperature information into the conductor temperature image and the insulator temperature image includes:

[0085] Perform data preprocessing on conductor temperature information and insulator temperature information;

[0086] The pre-processed conductor temperature information and insulator temperature information are converted into corresponding pixel values ​​using a temperature-pixel mapping algorithm; wherein a mapping relationship is established between the temperature range and the pixel value range;

[0087] According to the mapped pixel values, the conductor temperature image and the insulator temperature image are generated.

[0088] Data preprocessing is the first step in converting conductor temperature information and insulator temperature information into a temperature image. The main purpose of this step is to eliminate noise and outliers in the raw data to ensure the data quality for subsequent processing. Specifically, data preprocessing may include the following aspects: Data calibration: Calibrate the collected temperature data to eliminate the impact of sensor errors and environmental factors on the data. For example, by comparing the actual measurement value of a known temperature point with the theoretical value, adjust the measurement parameters of the sensor to ensure the accuracy of the data. Data normalization: Normalize the temperature data to a specific range, such as 0 to 1 or 0 to 255. Normalization processing facilitates subsequent temperature-pixel mapping, allowing data from different temperature ranges to be processed uniformly.

[0089] Converting preprocessed temperature data into pixel values ​​is a key step in generating a temperature image. The core of the temperature-pixel mapping algorithm is to establish a mapping relationship between temperature ranges and pixel value ranges. This mapping relationship is determined based on the actual application requirements. For example, the lowest temperature within the range can be mapped to a pixel value of 0, the highest temperature to a pixel value of 255, and intermediate temperatures can be mapped to pixel values ​​between 0 and 255 using a linear or nonlinear relationship. Based on this mapping relationship, each temperature data point is converted to its corresponding pixel value. This process converts temperature information into a format that can be recognized by image processing, providing a foundation for subsequent image analysis.

[0090] Based on the mapped pixel values, conductor and insulator temperature images are generated. This step converts the one-dimensional temperature data into two-dimensional image data, allowing the temperature information to be visually represented spatially. The mapped pixel values ​​are then inserted into the image matrix to generate a complete temperature image. The temperature image can intuitively display the temperature distribution of the conductors and insulators, facilitating subsequent image analysis and processing.

[0091] Data preprocessing eliminates noise and outliers from the raw data, improving data quality and reliability and laying a solid foundation for subsequent image generation and analysis. Converting temperature data into temperature images allows for intuitive spatial representation of temperature information. This intuitive visualization helps quickly identify areas of temperature anomalies, improving monitoring accuracy and efficiency. Temperature images provide rich information for subsequent image analysis, making it possible to utilize image processing and deep learning technologies for transmission line condition monitoring. For example, image edge detection and deep learning models can more accurately identify faults and abnormal conditions in transmission lines.

[0092] Furthermore, in a specific embodiment, performing image edge detection on the physical state image to obtain edge computing information includes:

[0093] Perform median filtering on the physical state image to remove image noise;

[0094] Obtain the gradient information of the filtered physical state image through the Sobel operator;

[0095] The gradient information is processed by the non-maximum suppression method to determine the edge points in the physical state image;

[0096] The edge points are classified by a dual threshold algorithm to determine first-level edge points and second-level edge points as edge calculation information, wherein the grayscale values ​​of the first-level edge points and the second-level edge points are different.

[0097] Median filtering is performed on the physical state image to remove image noise. Median filtering is a nonlinear filtering technique that effectively removes salt and pepper noise and other random noise by replacing each pixel value in the image with the median value of its neighborhood. Median filtering preserves image edge information and avoids blurring.

[0098] It's important to choose an appropriate filter window size (e.g., 3×3, 5×5) to balance noise removal effectiveness and computational complexity. A smaller window size preserves more detail but may not completely remove noise; a larger window size can better remove noise but may blur image edges.

[0099] Perform median filtering to obtain the first image information The specific expression is:

[0100] ;

[0101] Where, represents various status information of the transmission line, x and y represent the coordinates of the first image information, k and l are two-dimensional templates, and med[ ] represents an algorithm for calculating the difference between two strings.

[0102] The gradient information of the filtered physical state image is obtained through the Sobel operator. The Sobel operator is a commonly used edge detection operator that highlights edge features by calculating the gradient of the image in the horizontal and vertical directions.

[0103] The Sobel operator is used to calculate the image gradient. The Sobel operator is applied to the filtered image in the horizontal and vertical directions to calculate the gradient magnitude and direction of each pixel.

[0104] The gradient information is processed by non-maximum suppression to determine the edge points in the physical state image. If the pixel gradient amplitude is the maximum value of the gradient amplitudes of adjacent pixels along the gradient direction, the pixel is considered an edge point.

[0105] The non-maximum suppression method compares the gradient amplitude, retains the edge points of the local maximum, and removes other non-edge points. The specific steps are as follows:

[0106] For each pixel, check its neighboring pixels in the gradient direction.

[0107] If the gradient magnitude of the current pixel is the local maximum in its gradient direction, the pixel is retained as an edge point; otherwise, it is set as a non-edge point.

[0108] The edge points are classified by a dual threshold algorithm to determine first-level edge points and second-level edge points as edge calculation information, wherein the grayscale values ​​of the first-level edge points and the second-level edge points are different.

[0109] The high threshold is used to determine strong edge points. Edge points with grayscale values ​​higher than the high threshold are considered strong edge points, and their grayscale values ​​are usually set to 255 (white).

[0110] The low threshold is used to determine weak edge points. Edge points with grayscale values ​​between the low threshold and the high threshold are considered weak edge points, and their grayscale values ​​are usually retained or set to a lower value (such as 128).

[0111] Points with grayscale values ​​below the low threshold are considered noise or non-edge points and are usually set to 0 (black).

[0112] The selection of high threshold and low threshold can be adjusted according to the specific application scenario and image characteristics. Generally, a high threshold is used to detect obvious edges, while a low threshold is used to detect possible weak edges.

[0113] Median filtering can effectively remove noise from images while preserving edge information, improving image quality and the efficiency of subsequent processing. Calculating gradient information using the Sobel operator accurately extracts image edge features, providing reliable data support for subsequent edge detection. Non-maximum suppression can precisely locate edge points in an image, avoiding misjudgments and redundant edge points. Classifying edge points using a dual-threshold algorithm distinguishes between strong and weak edges, providing richer information for subsequent condition monitoring and fault diagnosis. This method can adapt to varying image quality and environmental conditions. By adjusting parameters such as the filter window size and threshold, it can flexibly address a variety of practical application scenarios.

[0114] Furthermore, in a specific embodiment, the edge computing information is input into a preset image processing deep learning model for processing to obtain the state type of the transmission line, including:

[0115] Input edge computing information into the embedding layer of the preset improved VisionTransformer model, which includes image block embedding, learnable embedding, and position embedding;

[0116] The output of the embedding layer is input into the Transformer encoder and features are extracted through the sparse attention module. The Transformer encoder is equipped with a sparse attention module and several layers of encoding modules. Each encoding module includes a multi-head self-attention layer and a multi-layer perceptron layer.

[0117] The output of the Transformer encoder is classified and predicted through the multi-layer perceptron classification head to obtain the state type of the transmission line.

[0118] The edge computing information is input into the embedding layer of the preset improved Vision Transformer model, which includes image block embedding, learnable embedding, and position embedding.

[0119] Image patch embedding divides edge computing information into fixed-size image patches (e.g., 16×16 pixels). Each image patch is flattened into a one-dimensional vector and mapped to the model's dimensional space through a linear transformation.

[0120] Learnable embedding introduces a learnable category vector (classtoken) that is input into the model together with the image patch embedding. This category vector is continuously learned throughout the training process and is used to store the global feature information of the image.

[0121] Position embedding: Since the Transformer architecture itself does not retain spatial information, position embedding is needed to provide the model with the spatial position information of pixels. Position embedding can be learned or fixed, such as sinusoidal position encoding.

[0122] The output of the embedding layer is input into the Transformer encoder and feature extraction is performed through the sparse attention module. The Transformer encoder is equipped with a sparse attention module and several layers of encoding modules. Each layer of encoding module includes a multi-head self-attention layer and a multi-layer perceptron layer.

[0123] The Transformer encoder is composed of multiple identical encoding modules stacked together. Each encoding module consists of two main parts: a multi-head self-attention layer and a multi-layer perceptron layer.

[0124] The sparse attention module introduces sparsity constraints, allowing the model to process image features more efficiently. For example, the sparse attention mechanism can be used to reduce the amount of computation while retaining important feature information.

[0125] The learning weights are obtained by the sparse attention module according to the layer encoding module before the last layer encoding module , its specific expression is:

[0126] ;

[0127] In the formula, m is the first ordinal number, , N is the number of coding module layers, is the learning sub-weight, i is the second ordinal number, , is the total number of self-attention heads;

[0128] ;

[0129] In the formula, i is the second ordinal number, , and All are learning-based weights;

[0130] The sparse attention module is used to globally average pool the attention map into a descriptor. Two connected multi-head self-attention layers and multi-layer perceptron layers are used to build a model of the correlation between attention maps. The weight values ​​of the encoding modules SM of each layer are obtained, and the final attention weight value is calculated with the learning weight. , its specific expression is:

[0131] ;

[0132] Where, is the weight value of the m-th layer coding module SM;

[0133] Through the weight corresponding to the classification vector in the final value of the attention weight, the implicit feature corresponding to the maximum weight is screened out in the self-attention head as the implicit feature input to the last layer encoding module.

[0134] The multi-head self-attention mechanism allows the model to learn information in different representation subspaces, thereby capturing long-range dependencies in the image. Each multi-head self-attention layer is followed by a multi-layer perceptron (MLP) to further process the feature vector. The MLP classification head classifies the Transformer encoder output and predicts the transmission line status.

[0135] The improved Vision Transformer model effectively extracts global and local features from images, providing rich information for power line status classification. The introduction of a sparse attention module makes the model more efficient when processing high-resolution images, reducing computational effort and memory usage. The multi-layer perceptron classification head accurately predicts the extracted features. Combined with the Softmax function, it outputs a probability distribution for each category, improving classification reliability.

[0136] Furthermore, in a specific embodiment, inputting edge computing information into the embedding layer of a preset improved VisionTransformer model includes:

[0137] In the embedding layer, the input edge computing information is segmented to obtain fixed-size image block tokens;

[0138] Add position embedding information to each image block token;

[0139] A one-dimensional token sequence is obtained by concatenating the image block token and the learnable embedded category vector;

[0140] The one-dimensional token sequence is used as the output of the embedding layer.

[0141] The input edge computing information is segmented in the embedding layer to obtain fixed-size image block tokens.

[0142] The edge computing input (typically a 2D image) is divided into fixed-size image blocks. For example, a 16×16 pixel block size can be chosen. Each image block is flattened into a 1D vector and mapped to the model's dimensional space using a linear transformation.

[0143] The size of the image patch can be adjusted based on the resolution of the input image and the computational power of the model. Larger image patches can reduce the amount of computation but may lose some detail information; smaller image patches can retain more details but increase computational complexity.

[0144] Since the Transformer architecture itself does not retain spatial information, it is necessary to add position embedding information to each image block token. Position embedding can be learned or fixed.

[0145] Position embeddings provide the model with information about the spatial location of pixels, enabling it to better understand spatial relationships within an image. A one-dimensional token sequence is obtained by concatenating the image patch tokens with the learnable embedding category vectors.

[0146] The category vector is concatenated element-wise with all image patch tokens to form a one-dimensional token sequence. This concatenated sequence serves as the output of the embedding layer and is fed into the subsequent Transformer encoder. Under certain conditions, such as multi-task learning, multiple category vectors can be introduced, each for different tasks, to enhance the model's versatility.

[0147] The one-dimensional token sequence is used as the output of the embedding layer, and the concatenated token sequence is used as the output of the embedding layer and directly input into the Transformer encoder.

[0148] Furthermore, in a specific embodiment, the output of the Transformer encoder is classified and predicted by a multi-layer perceptron classification head to obtain the state type of the transmission line, including:

[0149] Perform dimensionality reduction on the feature vector output by the Transformer encoder;

[0150] The eigenvector after dimension reduction is normalized by the Softmax function to obtain the probability distribution of the transmission line state type;

[0151] Determine the state type of the transmission line according to the probability distribution;

[0152] Output the status type of the transmission line.

[0153] Perform dimensionality reduction on the feature vector output by the Transformer encoder to reduce the dimension of the feature vector to reduce computational complexity and avoid overfitting, especially when processing high-dimensional feature data.

[0154] A multi-layer perceptron (MLP) is used to reduce the dimensionality of the feature vector output by the Transformer encoder. An MLP typically consists of one or more fully connected layers that transform features using nonlinear activation functions such as ReLU. This maps the high-dimensional feature vector to a lower-dimensional space while retaining the information most useful for the classification task. For example, a feature vector can be reduced from 512 dimensions to 128 dimensions.

[0155] The dimensionality-reduced feature vector is normalized by the softmax function to obtain the probability distribution of the transmission line state type, and the feature vector is converted into a probability distribution, especially when performing multi-category classification tasks.

[0156] The state type of the transmission line is determined based on the probability distribution, and the category with the highest probability is selected as the final classification result. For example, if the probability distribution is [0.1, 0.3, 0.6], the third category is selected as the state type of the transmission line.

[0157] A confidence threshold can be set, such as 0.5. If the highest probability is below the threshold, the classification result is considered unreliable and can be marked as "uncertain" or requested for further inspection.

[0158] Output the status type of the transmission line, output the classification results and perform subsequent processing. The output content may include information such as status type, probability value, and timestamp.

[0159] Classification results are displayed in real time on the monitoring center's display devices, allowing operators to keep abreast of the transmission line's status. If the classification result indicates an abnormality, operators can be notified promptly via SMS, email, or app push notifications. Under certain conditions, if further analysis of the classification results is required, the results can be stored in a database for subsequent trend analysis and historical tracing.

[0160] Furthermore, in a specific embodiment, outputting corresponding monitoring results according to the state type includes:

[0161] Display the monitoring results in a visual manner on the display device of the monitoring center;

[0162] According to the monitoring results, it is judged whether the status of the transmission line is normal.

[0163] When the monitoring results indicate an abnormal state, a fault report is generated, including the fault type, location, and severity.

[0164] Monitoring results are visualized on a display device in the monitoring center. This device can be a large-screen monitor, a monitoring terminal, or a mobile device. These devices require high resolution and excellent color rendering to clearly display monitoring results. Monitoring results can be displayed in a variety of visualization methods, such as directly displaying images of the physical status of the transmission line, including conductor outlines, insulator outlines, and conductor and insulator temperature images. Colors or icons can be used to indicate the transmission line status type, such as green for normal status and red for abnormal status. Real-time monitoring data, such as temperature values ​​and edge detection results, can be displayed.

[0165] Based on the status type predicted by classification, determine whether the transmission line is in a normal state. For example, the status type can be categorized as "normal," "minor abnormal," or "serious abnormal." A threshold is set to distinguish between normal and abnormal states. For example, if the probability of the "normal" category in the probability distribution of the status type falls below a certain threshold (such as 0.7), the transmission line is considered to be in an abnormal state. Other anomaly detection algorithms (such as threshold-based detection, statistical detection, or machine learning-based detection) can be combined to improve the accuracy of anomaly detection.

[0166] This embodiment ensures the efficiency and reliability of the transmission line status monitoring method through the detailed design of monitoring result output and fault report generation steps, and can effectively improve the operational safety and maintenance efficiency of the transmission line.

[0167] In the above embodiments, a method for monitoring the state of a power transmission line is described in detail. This application also provides corresponding embodiments of a device for monitoring the state of a power transmission line. It should be noted that this application describes the embodiments of the device from two perspectives: one is based on the functional module perspective, and the other is based on the hardware perspective.

[0168] Based on the perspective of functional modules, Figure 5 A structural diagram of a power transmission line status monitoring device provided in an embodiment of the present application is shown in FIG. Figure 5 As shown, a transmission line status monitoring device is applied to a sensing terminal installed on each high-voltage transmission tower, comprising:

[0169] The acquisition module 21 is used to acquire the conductor contour image, insulator contour image, conductor temperature information, and insulator temperature information on the transmission line;

[0170] A conversion module 22, configured to convert the conductor temperature information and the insulator temperature information into a conductor temperature image and an insulator temperature image;

[0171] An image edge detection module 23 is configured to perform image edge detection on each physical state image to obtain edge calculation information, wherein the physical state image includes a conductor contour image, an insulator contour image, a conductor temperature image, and an insulator temperature image;

[0172] An analysis and processing module 24 is used to input edge computing information into a preset image processing deep learning model for processing to obtain the status type of the transmission line;

[0173] The output module 25 is used to output corresponding monitoring results according to the status type.

[0174] Since the embodiments of the apparatus part correspond to the embodiments of the method part, please refer to the description of the embodiments of the method part for the embodiments of the apparatus part, and they will not be repeated here.

[0175] Figure 6 A structural diagram of another power transmission line status monitoring device provided in an embodiment of the present application is shown in FIG. Figure 6 As shown, the power transmission line status monitoring device includes: a memory 30 for storing a computer program;

[0176] The processor 31 is configured to implement the steps of the method for obtaining user operation habit information in the above embodiment (power transmission line status monitoring method) when executing the computer program.

[0177] The power transmission line status monitoring device provided in this embodiment may include but is not limited to a mobile terminal, a personal computer, a workstation, and the like.

[0178] The processor 31 may include one or more processing cores, such as a quad-core processor or an octa-core processor. The processor 31 may be implemented in at least one of the following hardware forms: a digital signal processor (DSP), a field-programmable gate array (FPGA), or a programmable logic array (PLA). The processor 31 may also include a main processor and a coprocessor. The main processor is used to process data in the awake state, also known as a central processing unit (CPU); the coprocessor is a low-power processor used to process data in the standby state. In some embodiments, the processor 31 may be integrated with a graphics processing unit (GPU), which is responsible for rendering and drawing content required to be displayed on the display. In some embodiments, the processor 31 may also include an artificial intelligence (AI) processor, which is used to handle computational operations related to machine learning.

[0179] The memory 30 may include one or more computer-readable storage media, which may be non-transitory. The memory 30 may also include a high-speed random access memory, and a non-volatile memory, such as one or more disk storage devices, flash memory storage devices. In this embodiment, the memory 30 is at least used to store the following computer program 301, wherein, after the computer program is loaded and executed by the processor 31, it can implement the relevant steps of the transmission line status monitoring method disclosed in any of the aforementioned embodiments. In addition, the resources stored in the memory 30 may also include an operating system 302 and data 303, etc., and the storage method may be temporary storage or permanent storage. Among them, the operating system 302 may include Windows, Unix, Linux, etc. The data 303 may include but is not limited to data involved in implementing the transmission line status monitoring method, etc.

[0180] In some embodiments, the power transmission line status monitoring device may further include a display screen 32 , an input / output interface 33 , a communication interface 34 , a power supply 35 , and a communication bus 36 .

[0181] Those skilled in the art will understand that Figure 6 The structure shown in the figure does not constitute a limitation to the power transmission line condition monitoring device, and may include more or fewer components than shown in the figure.

[0182] The power transmission line status monitoring device provided in an embodiment of the present application includes a memory and a processor. When the processor executes a program stored in the memory, it can implement the following method: a power transmission line status monitoring method.

[0183] Finally, the present application also provides an embodiment corresponding to a computer-readable storage medium. The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps described in the above-mentioned embodiment of the power transmission line status monitoring method.

[0184] It is understandable that if the method in the above embodiment is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and executes all or part of the steps of the methods of each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, and other media that can store program code.

[0185] The computer-readable storage medium provided in this embodiment stores a computer program. When a processor executes the program, the following method can be implemented: a method for monitoring the state of a power transmission line.

[0186] The above is a detailed introduction to the transmission line status monitoring method, device and medium provided by the present application. The various embodiments in the specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same and similar parts between the various embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description. It should be pointed out that for ordinary technicians in this technical field, without departing from the principles of this application, several improvements and modifications can be made to the present application, and these improvements and modifications also fall within the scope of protection of the claims of this application.

[0187] It should also be noted that, in this specification, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus comprising the element.

Claims

1. A method for monitoring the state of a transmission line, characterized in that: Applicable to sensing terminals installed on high-voltage transmission towers, including: Collect conductor contour images, insulator contour images, conductor temperature information, and insulator temperature information on transmission lines; Converting the conductor temperature information and the insulator temperature information into a conductor temperature image and an insulator temperature image; Performing image edge detection on each physical state image to obtain edge calculation information, wherein the physical state image includes a conductor contour image, an insulator contour image, a conductor temperature image, and an insulator temperature image; The edge computing information is input into the preset image processing deep learning model for processing to obtain the status type of the transmission line; Output the corresponding monitoring result according to the status type.

2. The method for monitoring the state of a transmission line according to claim 1, wherein: The converting the conductor temperature information and the insulator temperature information into a conductor temperature image and an insulator temperature image comprises: performing data preprocessing on the conductor temperature information and the insulator temperature information; The pre-processed conductor temperature information and the insulator temperature information are converted into corresponding pixel values ​​using a temperature-pixel mapping algorithm; wherein a mapping relationship is established between a temperature range and a pixel value range; A conductor temperature image and an insulator temperature image are generated according to the mapped pixel values.

3. The method for monitoring the state of a transmission line according to claim 2, wherein: The performing image edge detection on the physical state image to obtain edge computing information includes: performing a median filter process on the physical state image to remove image noise; Obtaining gradient information of the filtered physical state image through a Sobel operator; Processing the gradient information by a non-maximum suppression method to determine edge points in the physical state image; The edge points are classified by a dual threshold algorithm to determine first-level edge points and second-level edge points as edge calculation information, wherein the grayscale values ​​of the first-level edge points and the second-level edge points are different.

4. The method for monitoring the state of a power transmission line according to claim 3, wherein: The edge computing information is input into a preset image processing deep learning model for processing to obtain the state type of the transmission line, including: Inputting the edge computing information into an embedding layer of a preset improved VisionTransformer model, wherein the embedding layer includes image block embedding, learnable embedding, and position embedding; The output of the embedding layer is input into a Transformer encoder and feature extraction is performed through a sparse attention module. The Transformer encoder is provided with a sparse attention module and several layers of encoding modules, each layer of encoding modules includes a multi-head self-attention layer and a multi-layer perceptron layer; The output of the Transformer encoder is classified and predicted through the multi-layer perceptron classification head to obtain the state type of the transmission line.

5. The method for monitoring the state of a power transmission line according to claim 4, characterized in that: Inputting the edge computing information into an embedding layer of a preset improved VisionTransformer model includes: Segmenting the input edge computing information in the embedding layer to obtain image block tokens of fixed size; Adding position embedding information to each image block token; Obtaining a one-dimensional token sequence by concatenating the image block token and the learnable embedding category vector; The one-dimensional token sequence is used as the output result of the embedding layer.

6. The method for monitoring the state of a power transmission line according to claim 4, wherein: The output of the Transformer encoder is classified and predicted by the multi-layer perceptron classification head to obtain the status type of the transmission line, including: Performing dimensionality reduction processing on the feature vector output by the Transformer encoder; The eigenvector after dimension reduction is normalized by the Softmax function to obtain the probability distribution of the transmission line state type; determining a state type of the transmission line according to the probability distribution; The state type of the power transmission line is output.

7. The method for monitoring the state of a power transmission line according to any one of claims 1 to 6, characterized in that: Output corresponding monitoring results according to the status type, including: Displaying the monitoring results in a visual manner on a display device in a monitoring center; Determining whether the state of the transmission line is normal according to the monitoring result; When the monitoring result indicates an abnormal state, a fault report is generated, including the fault type, location, and severity.

8. A transmission line status monitoring device, characterized in that: Applicable to sensing terminals installed on high-voltage transmission towers, including: An acquisition module is used to acquire conductor contour images, insulator contour images, conductor temperature information, and insulator temperature information on the transmission line; A conversion module, configured to convert the conductor temperature information and the insulator temperature information into a conductor temperature image and an insulator temperature image; An image edge detection module is used to perform image edge detection on each physical state image to obtain edge calculation information, wherein the physical state image includes a conductor contour image, an insulator contour image, a conductor temperature image, and an insulator temperature image; An analysis and processing module is used to input edge computing information into a preset image processing deep learning model for processing to obtain the status type of the transmission line; The output module is used to output corresponding monitoring results according to the status type.

9. A transmission line status monitoring device, characterized in that: include: memory for storing computer programs; A processor, configured to implement the steps of the transmission line status monitoring method according to any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the power transmission line state monitoring method according to any one of claims 1 to 7 are implemented.

Images