Power transmission line icing treatment method, device, equipment, medium and product

Abstract

The invention discloses a power transmission line icing treatment method, device and equipment, a medium and a product. The method comprises the following steps: if an icing detection request for a power transmission line is detected, determining a target icing image corresponding to the power transmission line and a pre-trained target detection model; according to the target icing image, adopting a multi-level segmentation module and a lightweight classification module in a target detection model to perform identification processing on the target icing image to obtain a corresponding icing type; and according to the icing type, determining icing response measures for the power transmission line, and according to the icing response measures, processing icing on the power transmission line. According to the technical scheme provided by the invention, the target detection model can be used for identifying and processing the icing image, so that the icing condition of the power transmission line is remotely analyzed, and the icing processing efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the field of big data, and in particular to a method, device, equipment, medium and product for processing ice coating on a power transmission line. Background Art

[0002] Icing is a complex physical phenomenon affected by a variety of natural factors. During the power transmission process, ice on the surface of the transmission line can easily induce various safety accidents. Under the existing technical environment of the power grid, manual inspection is still the main inspection method for transmission line inspection. However, when the transmission line encounters high-risk weather such as blizzards and freezing rain, the actual operating environment of the transmission line is even worse. It is not easy for manual or transportation to go directly to the site, and even if they are on site, there are still great risks, making emergency repairs more difficult.

[0003] Therefore, how to use the target detection model to identify and process ice-covered images, so as to remotely analyze the ice-covered conditions of transmission lines and improve the efficiency of ice-covered treatment, is a problem that needs to be solved urgently. Summary of the invention

[0004] The present invention provides a method, device, equipment, medium and product for icing treatment of power transmission lines, which utilize a target detection model to identify and process icing images, thereby remotely analyzing the icing conditions of the power transmission lines and improving the efficiency of icing treatment.

[0005] According to one aspect of the present invention, a method for treating ice coating on a power transmission line is provided, comprising:

[0006] If an ice detection request for a transmission line is detected, a target ice image corresponding to the transmission line and a pre-trained target detection model are determined;

[0007] According to the target ice-covered image, the multi-level segmentation module and lightweight classification module in the target detection model are used to identify and process the target ice-covered image to obtain the corresponding ice-covered type;

[0008] According to the type of icing, determine the icing countermeasures for the transmission lines, and deal with the icing on the transmission lines according to the icing countermeasures.

[0009] According to another aspect of the present invention, there is provided a transmission line ice treatment device, comprising:

[0010] A determination module, for determining a target ice-covered image corresponding to the transmission line and a pre-trained target detection model if an ice-covered detection request for the transmission line is detected;

[0011] The recognition module is used to identify and process the target ice-covered image based on the target ice-covered image by using the multi-level segmentation module and the lightweight classification module in the target detection model to obtain the corresponding ice-covered type;

[0012] The processing module is used to determine the icing countermeasures for the transmission line according to the icing type, so as to process the icing on the transmission line according to the icing countermeasures.

[0013] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0014] at least one processor; and

[0015] a memory communicatively connected to the at least one processor; wherein,

[0016] The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the method for processing ice coating on a power transmission line described in any embodiment of the present invention.

[0017] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the method for processing ice coating on a power transmission line as described in any embodiment of the present invention when executed.

[0018] According to another aspect of the present invention, a computer program product is provided. The computer program product includes a computer program. When the computer program is executed by a processor, the method for processing ice coating on a power transmission line according to any embodiment of the present invention is implemented.

[0019] The technical solution of the embodiment of the present invention, if an ice detection request for a transmission line is detected, then the target ice image corresponding to the transmission line and the pre-trained target detection model are determined; according to the target ice image, the multi-level segmentation module and the lightweight classification module in the target detection model are used to identify and process the target ice image to obtain the corresponding ice type; according to the ice type, the ice countermeasures for the transmission line are determined, and the ice on the transmission line is processed according to the ice countermeasures. By using the target detection model to identify and process the ice image, the ice condition of the transmission line can be remotely analyzed, which can effectively improve the efficiency of ice processing.

[0020] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0022] Figure 1A This is a flow chart of a method for treating ice coating on a power transmission line provided in Embodiment 1 of the present invention;

[0023] Figure 1B is a schematic diagram of the structure of a multi-level segmentation module provided in Embodiment 1 of the present invention;

[0024] Figure 1C is a schematic diagram of the structure of a CP module provided in Embodiment 1 of the present invention;

[0025] Figure 1D is a structural diagram of a lightweight classification module provided in Embodiment 1 of the present invention;

[0026] Figure 1E is a schematic diagram of the structure of the DS module provided in the first embodiment of the present invention;

[0027] Figure 2 This is a flow chart of a method for treating ice coating on a power transmission line provided in Embodiment 2 of the present invention;

[0028] Figure 3 This is a structural block diagram of a power transmission line ice treatment device provided in Embodiment 3 of the present invention;

[0029] Figure 4 It is a schematic diagram of the structure of an electronic device provided in Embodiment 4 of the present invention. DETAILED DESCRIPTION

[0030] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0031] It should be noted that the terms "first", "second", "target", "candidate", "alternative", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices. The acquisition, storage, use, processing, etc. of data in the technical solution of this application comply with the relevant provisions of national laws and regulations.

[0032] Embodiment 1

[0033] Figure 1A This is a flow chart of a method for treating ice coating on a power transmission line provided in Embodiment 1 of the present invention; Figure 1B is a schematic diagram of the structure of a multi-level segmentation module provided in Embodiment 1 of the present invention; Figure 1C is a schematic diagram of the structure of a CP module provided in Embodiment 1 of the present invention; Figure 1D is a structural diagram of a lightweight classification module provided in Embodiment 1 of the present invention; Figure 1E : is a structural diagram of the DS module provided in the first embodiment of the present invention. This embodiment can be applied to the situation where the ice-covered image of the transmission line is identified based on the target detection model to obtain the ice-covered type, so as to facilitate the subsequent adoption of corresponding strategies for processing. The method can be executed by a transmission line ice-covered processing device, and the transmission line ice-covered processing device can be implemented in the form of hardware and / or software. The transmission line ice-covered processing device can be configured in an electronic device, such as an electronic device with a transmission line ice-covered processing function, and is executed by a master control system in a distribution network, such as Figure 1A As shown, the transmission line icing treatment method includes:

[0034] S101. If an ice detection request for a power transmission line is detected, a target ice image corresponding to the power transmission line and a pre-trained target detection model are determined.

[0035] Among them, the ice detection request refers to a request for detecting the ice condition on the transmission line. The target ice image refers to an image obtained by image acquisition of the transmission line. The target ice image can characterize the ice condition of the transmission line. The target detection model refers to a pre-trained model for segmenting and classifying the target ice image. The target detection model may include a multi-level segmentation module and a lightweight classification module. The multi-level segmentation module includes a feature extraction layer, a convolution layer, and an upsampling layer. The lightweight classification module may include a first classification convolution layer, an information extraction layer, a second classification convolution layer, an average pooling layer, and a fully connected layer. The target detection model may be, for example, a neural network model.

[0036] Optionally, the master control system of the distribution network can detect the ambient temperature of the transmission lines in the power grid in real time. If the ambient temperature is lower than a preset freezing temperature threshold, it is determined that ice detection is required for the transmission line. At this time, it is considered that an ice detection request for the transmission line is detected.

[0037] Optionally, if an ice detection request for a transmission line is detected, the master control system of the distribution network may use an image acquisition device corresponding to the transmission line to acquire a target ice image corresponding to the transmission line.

[0038] Optionally, before determining the pre-trained target detection model, it also includes: constructing a training sample set based on historical ice cover images and their corresponding ice cover types; based on the training sample set, training the initial detection model based on the cross entropy loss function and the negative log-likelihood loss function to obtain the target detection model.

[0039] Among them, the historical ice-covered image refers to an image obtained by collecting images of outdoor power transmission lines in a historical period of time, which can characterize the ice-covered conditions of the transmission lines. The ice-covered types can be, for example, normal without ice, rime type, snow type, rime type, and mixed type. Mixed type refers to an ice-covered type including at least two types of rime type, snow type, and rime type. The training sample set contains historical ice-covered images of different ice-covered types, specifically, normal without ice-covered power transmission line images, rime type ice-covered power transmission line images, snow type ice-covered power transmission line images, rime type ice-covered power transmission line images, and mixed type ice-covered power transmission line images. The initial detection model can be an initialized neural network model. The target detection model refers to a model obtained by iteratively updating the parameters of the initial detection model based on the training sample set and the loss function. The initial detection model can include a multi-level segmentation module and a lightweight classification module, and the multi-level segmentation module includes a feature extraction layer, a convolution layer, and an upsampling layer. The lightweight classification module may include a first classification convolution layer, an information extraction layer, a second classification convolution layer, an average pooling layer, and a fully connected layer.

[0040] Optionally, the padding method can be used to fill the specified value around each historical ice image to unify the image size, and the Labelme annotation software can be used to annotate the historical ice images according to the ice type corresponding to the historical ice image, so as to construct a training sample set. Among them, pad is a method for modifying the image size during the data set preprocessing of a neural network model. This method does not change the original data form and does not lose the original information of the image. Labelme annotation software is a common tool for annotating image information.

[0041] Optionally, the training sample set can be divided into a training set, a validation set, and a test set based on a ratio of 6:2:2 to train and validate the initial detection model. For example, a batch of training sample set data can be input into the initial detection model for training, and the training batch size (i.e., batch size) batch_size = 16, the training batch epoch = 200, the initial learning rate is 0.001, the multi-level segmentation module MUnet uses the cross entropy loss function to calculate the segmentation loss value seg_loss, and the lightweight classification module MEnet uses the negative log-likelihood loss function to calculate the classification loss value cls_loss. The overall loss calculation formula is as follows:

[0042] loss=0.5×seg_loss+0.5×cls_loss

[0043] After the training of all batches of data in the training sample set is completed, the training sample set can be input into the initial detection model in batches to obtain the corresponding batch loss value batch_loss. During training and verification, the initial detection model will automatically learn and adjust parameters according to each loss and batch_loss situation. When the loss and batch_loss values ​​tend to converge and the difference between them is very small, the initial detection model training ends and the final target detection model is obtained.

[0044] S102: According to the target ice-covered image, a multi-level segmentation module and a lightweight classification module in the target detection model are used to perform recognition processing on the target ice-covered image to obtain a corresponding ice-covered type.

[0045] Among them, the multilevel segmentation module (Multilevel segmentation net, MUnet) refers to the module used for image segmentation in the target detection model, and the lightweight classification module (Mini Efficientnet) refers to the module used for image classification in the target detection model. The multilevel segmentation module may include at least one feature extraction layer (abbreviated as CP, the feature extraction layer includes a convolution unit conv and a PSwish activation function processing unit). The lightweight classification module may include at least one information extraction layer (Depth and Spatial feature extraction module, DS).

[0046] Optionally, according to the target ice-covered image, a multi-level segmentation module and a lightweight classification module in the target detection model are used to identify and process the target ice-covered image to obtain a corresponding ice-covered type, including: self-stitching the target ice-covered image (denoted as M1) to obtain a first stitched ice-covered image (denoted as M2); using the multi-level segmentation module in the target detection model to segment the first stitched ice-covered image to obtain a first segmented ice-covered image (denoted as M3), and stitching the first segmented ice-covered image with the target ice-covered image to obtain a second stitched ice-covered image (denoted as M4); using the multi-level segmentation module in the target detection model to segment the second stitched ice-covered image to obtain a second segmented ice-covered image (denoted as M5), and using the lightweight classification module in the target detection model to classify the second segmented ice-covered image to obtain the corresponding ice-covered type.

[0047] Exemplarily, the size of the target ice-covered image (denoted as M1) can be H×W×C, where H and W are the height and width of the image, and C is the number of channels of the image. The target ice-covered image can be concat-ed with itself, i.e., self-stitching, to obtain a feature map M2 of size H×W×2C, which is the first spliced ​​ice-covered image.

[0048] It should be noted that the concat operation combines the number of channels of two feature maps. The size of the resulting feature map remains unchanged, but the number of channels increases.

[0049] Exemplarily, after obtaining the first spliced ​​ice image M2, M2 can be input into MUnet for segmentation to obtain a feature map M3 that extracts the transmission line embodiment and the line ice contour and location information. M3 is spliced ​​with M1 to obtain a feature map M4, and M4 is input into MUnet for segmentation to obtain a second segmented ice image M5. The pixels in M5 are divided into three categories, namely, transmission lines, line ice and background, to achieve target segmentation in the image. Finally, M5 is input into MEnet for line ice classification, and the output is normal ice-free, rime type, snow type, rime type or mixed ice type.

[0050] Exemplarily, after concat splicing M3 and M1, a feature map M4 of size H×W×2C can be obtained. M4 is input into the segmentation module MUnet for processing to obtain a second segmented ice-covered image M5 of size H×W×C, and the image is segmented into three non-overlapping areas: transmission lines, line ice and background.

[0051] Optionally, a multi-level segmentation module in the target detection model is used to perform segmentation processing on the first spliced ​​ice-covered image to obtain a first segmented ice-covered image, including: determining the feature extraction layer, convolution layer and upsampling layer included in the multi-level segmentation module in the target detection model; using each feature extraction layer, convolution layer and upsampling layer to hierarchically perform extraction processing, convolution processing and upsampling processing on the first spliced ​​ice-covered image to segment and obtain position information of the transmission line area, ice-covered area and background area in the first spliced ​​ice-covered image, and marking them in the first spliced ​​ice-covered image according to the position to obtain a first segmented ice-covered image.

[0052] Among them, the number of feature extraction layers (CP) is at least two; the number of convolutional layers (conv) is at least two; and the number of upsampling layers (Upsample) is at least two.

[0053] Optionally, obtaining a first segmented ice-covered image includes: using a first feature extraction layer, a second feature extraction layer, a third feature extraction layer and a fourth feature extraction layer to extract and process the first spliced ​​ice-covered image M2 respectively to obtain a first feature, a second feature, a third feature and a fourth feature (expressed as U1, U2, U3 and U4); upsampling the fourth feature, and adding and convolving the fourth feature after the upsampling with the third feature to obtain a fifth feature (expressed as U5); upsampling the fifth feature, and performing an upsampling process on the upsampling. The processed fifth feature is added and convolved with the second feature to obtain the sixth feature (denoted as U6); the sixth feature is upsampled, and the upsampled sixth feature is added to the first feature to obtain the seventh feature (denoted as U7); the seventh feature is sequentially convolved, upsampled and reconvolved to obtain the position information of the power transmission line area, the ice-covered area and the background area in the first spliced ​​ice-covered image, and marked in the first spliced ​​ice-covered image according to the position to obtain the first segmented ice-covered image M3.

[0054] The addition process refers to the Add operation, specifically, adding the values ​​of the pixels in the feature map one by one. The first feature extraction layer, the second feature extraction layer, the third feature extraction layer, and the fourth feature extraction layer are essentially feature extraction layers that include a convolution unit conv and a PSwish activation function processing unit, which can be represented as a CP module.

[0055] For example, see Figure 1B , the feature map M2 with a size of H×W×2C is input into the first feature extraction layer (CP module), and a feature map U1 with a size of H / 2×W / 2×64 is obtained, that is, the first feature. The feature map U1 is further passed through three CP modules in sequence to obtain a feature map U2 with a size of H / 4×W / 4×128, a feature map U3 with a size of H / 8×W / 8×256, and a feature map U4 with a size of H / 16×W / 16×512, that is, the first feature, the second feature, the third feature, and the fourth feature.

[0056] Furthermore, U4 can be input to the upsampling layer Upsample for upsampling, and the result obtained is added to U3, and then passes through two conv layers with a convolution kernel size of 3×3, a step size of 1, and a padding of 1 to obtain a feature map U5 with a size of H / 8×W / 8×256. U5 is input to the upsampling layer Upsample for upsampling, and the result obtained is added to the feature map U2, and then passes through two conv layers with a convolution kernel size of 3×3, a step size of 1, and a padding of 1 to obtain a feature map U6 with a size of H / 4×W / 4×128. U6 is input to the upsampling layer Upsample for upsampling, and the result obtained is added to U1 to obtain a feature map U7 with a size of H / 2×W / 2×64.

[0057] Finally, U7 can be input into a conv layer with a convolution kernel size of 3×3, a step size of 1, and a padding of 1, and then upsampled by Upsample, and finally passed through a conv layer with a convolution kernel size of 1×1 and a step size of 1. Finally, the contour and position information of the transmission line and the line ice coverage are extracted, and the feature map M3 with a size of H×W×C is obtained. Only the pixel values ​​within the contour of the transmission line and the line ice coverage are retained in M3, and the values ​​of the remaining pixels outside the contour are all 0, thereby realizing the segmentation processing of the first spliced ​​ice-covered image.

[0058] Optionally, the location information of the retained pixel values ​​in M3 can be determined as the location information of the transmission line area and the ice-covered area. Furthermore, the area where the retained pixel values ​​are retained can be divided into the transmission line area and the ice-covered area according to the common pixel range of the transmission line area and the ice-covered area to respectively obtain the location information of the transmission line area and the location information of the ice-covered area, and the location information of the area with pixel 0 in M3 can be determined as the location information of the background area.

[0059] For example, see Figure 1C , the feature extraction layer included in the multi-level segmentation module in the target detection model, that is, the CP module can include three convolution units. Specifically, after the feature map M2 is input into the CP module, it can pass through two conv layers (i.e., convolution units) with a convolution kernel size of 3×3, a step size of 1, and a padding of 1 in sequence to obtain a feature map P1 with a size of H×W×2C. After adding the corresponding elements of P1 and M1, a feature map P2 with a size of H×W×2C is obtained. After P2 passes through a conv layer with a convolution kernel size of 3×3, a step size of 2, a padding of 1, and a number of convolution kernels of 64, as well as a PSwish activation function, a feature map U1 with a size of H / 2×W / 2×64 can be output.

[0060] Exemplarily, the calculation formula of the PSwish activation function can be as follows:

[0061]

[0062] In the formula, x is the value of the pixel in the input feature map, α is the parameter adjusted through training learning, and the calculation formula of sigmoid(x) is as follows:

[0063]

[0064] Among them, x is the value of the pixel in the input feature map, and e is an infinite non-repeating decimal.

[0065] For example, if the size of M1 is 1024×1024×3, after concatenating M1 with itself, a feature map M2 of size 1024×1024×6 can be obtained. M2 is input into MUnet, and after being processed by the CP module, a feature map U1 of size 512×512×64 can be obtained. After U1 passes through the CP module, a feature map U2 of size 256×256×128 is obtained. After U2 passes through the CP module, a feature map U3 of size 128×128×256 is obtained. After U3 passes through the CP module, a feature map U4 of size 64×64×512 is obtained. U4 is input into the upsampling layer Upsample for upsampling, and the result is added to U3, and then passed through two conv layers with a convolution kernel size of 3×3, a step size of 1, and a padding of 1 to obtain a feature map U5 of size 128×128×256. U5 is input to the upsampling layer Upsample for upsampling, and the result is added to the feature map U2, and then passes through two conv layers with a convolution kernel size of 3×3, a step size of 1, and a padding of 1 to obtain a feature map U6 of size 256×256×128. U6 is input to the upsampling layer Upsample for upsampling, and the result is added to U1 to obtain a feature map U7 of size 512×512×64.

[0066] For example, U7 is input into a conv layer with a convolution kernel size of 3×3, a step size of 1, and a padding of 1, and then upsampled by Upsample, and finally passed through a conv layer with a convolution kernel size of 1×1 and a step size of 1, to obtain a feature map M3 of size 1024×1024×3, which extracts the contour and position information of the transmission line and the line ice. In M3, only the pixel values ​​within the contour of the transmission line and the line ice are retained, and the values ​​of the remaining pixels outside the contour are all 0.

[0067] For example, after concat splicing M3 and M1, a feature map M4 with a size of 1024×1024×6 can be obtained. M4 is input into the segmentation module MUnet for processing to obtain a second segmented ice-covered image M5 with a size of 1024×1024×3, and the image is segmented into three non-overlapping areas: transmission lines, line ice and background.

[0068] Optionally, a lightweight classification module in the target detection model is used to classify the second segmented ice-covered image to obtain a corresponding ice-covered type, including: determining the first classification convolution layer, information extraction layer, second classification convolution layer, average pooling layer and fully connected layer included in the lightweight classification module in the target detection model; using the first classification convolution layer to convolve the second segmented ice-covered image M5 to obtain a convolution image E1, and using each information extraction layer to extract the convolution image in turn to obtain an extracted image; using the second classification convolution layer, average pooling layer and fully connected layer to classify the extracted image to obtain the corresponding ice-covered type.

[0069] Among them, the number of information extraction layers (DS) is at least two, and the first classification convolution layer can be specifically: a conv layer with a convolution kernel size of 3×3, a step size of 2, a padding of 1, and a number of convolution kernels of 32. The second classification convolution layer can be specifically a conv layer with a convolution kernel size of 1×1 and a step size of 1.

[0070] For example, see Figure 1D , if the size of the second segmented ice-covered image M5 is H×W×C, then after the second segmented ice-covered image M5 is input into the lightweight classification module, it will first pass through the first classification convolution layer to obtain a feature map E1 of size H / 2×W / 2×32, that is, a convolution image, and further pass through the same information extraction layer in sequence, such as three identical information extraction layers, using the first information extraction layer to process the convolution image E1 to obtain a feature map E2, using the second information extraction layer to process the feature map E2 to obtain a feature map E3, and using the third feature extraction layer to process the feature map E3 to obtain a feature map E4, that is, an extracted image. Among them, the size of the feature map E3 is H / 8×W / 8×128, and the size of the feature map E4 is H / 16×W / 16×256.

[0071] Exemplarily, the extracted image E4 can be input into the second classification convolution layer, the average pooling layer and the fully connected layer to obtain the corresponding ice cover type, wherein the second classification convolution layer and the Pooling layer (i.e., the average pooling layer) perform global average pooling on the feature map output by the last DS layer, compressing the feature map of each channel into a single numerical value, thereby obtaining a global feature vector. This step helps to reduce the number of parameters of the model and prevent overfitting, and outputs a global feature vector; the fully connected layer maps the global feature vector obtained from the second classification convolution layer and the Pooling layer to the final classification result. This is a typical classification layer, which is responsible for mapping high-level features to class probabilities, outputting probability values ​​for each category, and finally, based on the output probability values ​​for each category, determining the final ice cover type with the highest probability value.

[0072] Exemplarily, if the size of the second segmented ice-covered image M5 is 1024×1024×3, the process of inputting it into the lightweight classification module MEnet for classification can be: after a first classification convolution layer with a convolution kernel size of 3×3, a step size of 2, and a padding of 1, a convolution image E1 with a size of 512×512×32 is obtained. After E1 passes through the DS module, a feature map E2 with a size of 256×256×64 is obtained. After E2 passes through the DS module, a feature map E3 with a size of 128×128×128 is obtained. After E3 passes through the DS module, a feature map E4 with a size of 64×64×256 is obtained, that is, the extracted image. After the extracted image E4 passes through a conv layer with a convolution kernel size of 1×1 and a step size of 1, an average pooling layer Pooling and a fully connected layer FC, the ice-covered type of the image can be finally output.

[0073] For example, see Figure 1E , the information extraction layer contained in the lightweight classification module in the target detection model, that is, the DS module can include a normal convolution conv layer, a depthwise separable convolution layer (DWconv), a maximum pooling layer (MaxPool) and a channel reconstruction convolution layer (Spatial and Channel Reconstruction Convolution, SCconv). Specifically, after inputting the DS module, the convolution image E1 can first pass through a normal convolution conv layer with a convolution kernel size of 1×1 and a step size of 1, and the number of channels of E1 is increased by n times (after experimental testing, setting n=6 is the best effect, which enriches the channel feature information without increasing too much calculation) to obtain a feature map D1 of size H / 2×W / 2×192.

[0074] Furthermore, D1 can be evenly divided into two groups according to the number of channels, and feature maps D2 and D3 of size H / 2×W / 2×96 are obtained. D2 is passed through a common convolution conv layer with a convolution kernel size of 1×1 and a stride of 1 to increase the number of channels, and then passes through a depthwise convolution DWconv layer with a convolution kernel size of 3×3 and a stride of 2 for feature extraction, and finally passes through a common convolution conv layer with a convolution kernel size of 1×1 and a stride of 1 to reduce the number of channels, obtaining a feature map D4 of size H / 4×W / 4×96.

[0075] Furthermore, D3 first passes through a 2×2 maximum pooling layer MaxPool, and then passes through a channel reconstruction convolution SCconv to obtain a feature map D5 of size H / 4×W / 4×96. Finally, D4 and D5 are added, and then pass through a normal convolution conv layer with a convolution kernel size of 1×1, a step size of 1, and a number of convolution kernels of 64, and the output size is a feature map E2 of size H / 4×W / 4×64.

[0076] S103: Determine icing countermeasures for the transmission line according to the icing type, and handle the icing on the transmission line according to the icing countermeasures.

[0077] Among them, the icing countermeasures refer to the measures to remove ice on the transmission line. For example, the icing countermeasures can be robot deicing, manual deicing, or increasing the load current or using short-circuit current to increase the conductor temperature.

[0078] Optionally, if the icing type is normal without icing, the corresponding icing countermeasure may be not performing any operation.

[0079] Optionally, according to the icing type, the icing countermeasures for the transmission line can be determined in combination with the preset correspondence between the icing type and the icing countermeasures, so that the icing on the transmission line can be handled according to the icing countermeasures.

[0080] The technical solution of the embodiment of the present invention, if an ice detection request for a transmission line is detected, then the target ice image corresponding to the transmission line and the pre-trained target detection model are determined; according to the target ice image, the multi-level segmentation module and the lightweight classification module in the target detection model are used to identify and process the target ice image to obtain the corresponding ice type; according to the ice type, the ice countermeasures for the transmission line are determined, and the ice on the transmission line is processed according to the ice countermeasures. By using the target detection model to identify and process the ice image, the ice condition of the transmission line can be remotely analyzed, which can effectively improve the efficiency of ice processing.

[0081] Embodiment 2

[0082] Figure 2 1 is a flow chart of a method for handling ice coating on a power transmission line provided in the second embodiment of the present invention; based on the above embodiment, this embodiment provides a preferred solution for handling ice coating. Figure 2 As shown, the method includes the following process:

[0083] Optionally, if a request for ice detection on a transmission line is detected, image acquisition can be performed on the outdoor transmission line to obtain an outdoor transmission line image, that is, a target ice image M1. M1 is concatenated with itself to obtain a first concatenated ice image M2. M2 is input into a multi-level segmentation module MUnet for segmentation to obtain a first segmented ice image M3 that extracts the transmission line embodiment and the line ice contour and position information.

[0084] Optionally, M3 and M1 are concatenated to obtain a second concatenated ice-covered image M4, and M4 is input into a multi-level segmentation module MUnet for segmentation to obtain a second segmented ice-covered image M5, where the pixels in M5 are divided into three categories, namely, transmission lines, line ice and background, to achieve segmentation of targets within the image. Finally, the second segmented ice-covered image M5 is input into a lightweight classification module MEnet for line ice classification, and the ice classification and segmented image corresponding to the target ice-covered image are output, so that the target ice-covered image is classified into a normal non-ice-covered transmission line image, a rime-type ice-covered transmission line image, a snow-type ice-covered transmission line image, a rime-type ice-covered transmission line image, or a mixed ice-covered transmission line image.

[0085] The technical solution of the present invention can effectively and accurately segment the image of an outdoor power transmission line into a power transmission line, line ice and background, and classify the line ice into a normal non-icing power transmission line image, a rime type iced power transmission line image, a snow type iced power transmission line image, a rime type iced power transmission line image or a mixed type iced power transmission line image, which is helpful for determining the countermeasures for icing. At the same time, the target detection model designed by the present invention has fewer computing parameters, higher computing efficiency and more accurate results.

[0086] Embodiment 3

[0087] Figure 3 It is a structural block diagram of a transmission line icing treatment device provided in the third embodiment of the present invention; this embodiment can be applied to the situation where the icing image of the transmission line is identified based on the target detection model to obtain the icing type, so as to facilitate the subsequent adoption of corresponding strategies for processing. The transmission line icing treatment device provided in the embodiment of the present invention can execute the transmission line icing treatment method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method; the transmission line icing treatment device can be implemented in the form of hardware and / or software, and configured in an electronic device with a transmission line icing treatment function, such as an electronic device with a transmission line icing treatment function, and is executed by the master control system in the distribution network, such as Figure 3 As shown, the transmission line ice treatment device specifically includes:

[0088] The determination module 301 is used to determine the target ice-covered image corresponding to the transmission line and the pre-trained target detection model if an ice-covered detection request for the transmission line is detected;

[0089] The recognition module 302 is used to recognize and process the target ice-covered image according to the target ice-covered image by using the multi-level segmentation module and the lightweight classification module in the target detection model to obtain the corresponding ice-covered type;

[0090] The processing module 303 is used to determine the icing countermeasures for the transmission line according to the icing type, so as to process the icing on the transmission line according to the icing countermeasures.

[0091] The technical solution of the embodiment of the present invention, if an ice detection request for a transmission line is detected, then the target ice image corresponding to the transmission line and the pre-trained target detection model are determined; according to the target ice image, the multi-level segmentation module and the lightweight classification module in the target detection model are used to identify and process the target ice image to obtain the corresponding ice type; according to the ice type, the ice countermeasures for the transmission line are determined, and the ice on the transmission line is processed according to the ice countermeasures. By using the target detection model to identify and process the ice image, the ice condition of the transmission line can be remotely analyzed, which can effectively improve the efficiency of ice processing.

[0092] Further, the identification module 302 may include:

[0093] A first stitching unit, used for performing self-stitching processing on the target ice-covered image to obtain a first stitched ice-covered image;

[0094] A second splicing unit is used to segment the first spliced ​​ice-covered image using a multi-level segmentation module in the target detection model to obtain a first segmented ice-covered image, and to splice the first segmented ice-covered image with the target ice-covered image to obtain a second spliced ​​ice-covered image;

[0095] The classification unit is used to use the multi-level segmentation module in the target detection model to segment the second spliced ​​ice-covered image to obtain a second segmented ice-covered image, and use the lightweight classification module in the target detection model to classify the second segmented ice-covered image to obtain a corresponding ice-covered type.

[0096] Further, the second splicing unit may include:

[0097] Determine a subunit, used to determine the feature extraction layer, convolution layer and upsampling layer included in the multi-level segmentation module in the target detection model; wherein the number of the feature extraction layers is at least two;

[0098] The stitching subunit is used to use each feature extraction layer, convolution layer and upsampling layer to hierarchically perform extraction processing, convolution processing and upsampling processing on the first stitched ice-covered image to segment and obtain the position information of the power transmission line area, ice-covered area and background area in the first stitched ice-covered image, and mark them in the first stitched ice-covered image according to the position to obtain a first segmented ice-covered image.

[0099] Further, the splicing subunit is specifically used for:

[0100] The first feature extraction layer, the second feature extraction layer, the third feature extraction layer and the fourth feature extraction layer are used to extract and process the first spliced ​​ice-covered image respectively to obtain a first feature, a second feature, a third feature and a fourth feature;

[0101] Performing upsampling processing on the fourth feature, and performing addition processing and convolution processing on the fourth feature after the upsampling processing and the third feature to obtain a fifth feature;

[0102] Performing upsampling processing on the fifth feature, and performing addition processing and convolution processing on the fifth feature after the upsampling processing and the second feature to obtain a sixth feature;

[0103] Performing upsampling processing on the sixth feature, and performing addition processing on the upsampling sixth feature and the first feature to obtain a seventh feature;

[0104] The seventh feature is sequentially subjected to convolution processing, upsampling processing and secondary convolution processing to obtain the position information of the power transmission line area, the ice-covered area and the background area in the first spliced ​​ice-covered image, and marked in the first spliced ​​ice-covered image according to the position to obtain a first segmented ice-covered image.

[0105] Furthermore, the taxonomic units are specifically used for:

[0106] Determine a first classification convolution layer, an information extraction layer, a second classification convolution layer, an average pooling layer, and a fully connected layer included in a lightweight classification module in a target detection model; wherein the number of information extraction layers is at least two;

[0107] Using the first classification convolution layer, convolution processing is performed on the second segmented ice-covered image to obtain a convolution image, and each information extraction layer is used to extract the convolution image in turn to obtain an extracted image;

[0108] The second classification convolution layer, average pooling layer and fully connected layer are used to classify the extracted images and obtain the corresponding ice cover type.

[0109] Furthermore, the above device is also used for:

[0110] Construct a training sample set based on historical ice cover images and their corresponding ice cover types;

[0111] According to the training sample set, the initial detection model is trained based on the cross entropy loss function and the negative log-likelihood loss function to obtain the target detection model.

[0112] Embodiment 4

[0113] Figure 4 It is a schematic diagram of the structure of an electronic device provided in Embodiment 4 of the present invention. Figure 4 A schematic diagram of the structure of an electronic device 10 that can be used to implement an embodiment of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0114] like Figure 4 As shown, the electronic device 10 includes at least one processor 11, and a memory connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 to the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0115] A number of components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0116] The processor 11 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as a power transmission line ice treatment method.

[0117] In some embodiments, the power transmission line icing treatment method may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the power transmission line icing treatment method described above may be performed. Alternatively, in other embodiments, the processor 11 may be configured to execute the power transmission line icing treatment method in any other appropriate manner (e.g., by means of firmware).

[0118] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), system-on-chips (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0119] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0120] In the context of the present invention, a computer readable storage medium may be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, device, or equipment. A computer readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, semiconductor systems, devices, or equipment, or any suitable combination of the foregoing. Alternatively, a computer readable storage medium may be a machine readable signal medium. A more specific example of a machine readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0121] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0122] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0123] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0124] In one embodiment, the embodiment of the present invention further includes a computer program product, the computer program product includes a computer program, and when the computer program is executed by a processor, the method for processing ice coating on a power transmission line of any embodiment of the present invention is implemented.

[0125] In the process of implementation, the computer program product can be written in one or more programming languages ​​or a combination thereof to perform the computer program code of the present invention, including object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computer (for example, using an Internet service provider to connect through the Internet).

[0126] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0127] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for treating ice coating on a power transmission line, characterized in that: include: If an ice detection request for a transmission line is detected, a target ice image corresponding to the transmission line and a pre-trained target detection model are determined; According to the target ice-covered image, the multi-level segmentation module and lightweight classification module in the target detection model are used to identify and process the target ice-covered image to obtain the corresponding ice-covered type; According to the type of icing, determine the icing countermeasures for the transmission lines, and deal with the icing on the transmission lines according to the icing countermeasures.

2. The method according to claim 1, characterized in that According to the target ice-covered image, the multi-level segmentation module and lightweight classification module in the target detection model are used to identify and process the target ice-covered image to obtain the corresponding ice-covered type, including: The target ice-covered image is self-stitched to obtain a first stitched ice-covered image; The multi-level segmentation module in the target detection model is used to segment the first spliced ​​ice-covered image to obtain a first segmented ice-covered image, and the first segmented ice-covered image is spliced ​​with the target ice-covered image to obtain a second spliced ​​ice-covered image; The multi-level segmentation module in the target detection model is used to segment the second spliced ​​ice-covered image to obtain a second segmented ice-covered image, and the lightweight classification module in the target detection model is used to classify the second segmented ice-covered image to obtain the corresponding ice-covered type.

3. The method according to claim 2, characterized in that The multi-level segmentation module in the target detection model is used to segment the first spliced ​​ice-covered image to obtain a first segmented ice-covered image, including: Determine a feature extraction layer, a convolution layer, and an upsampling layer included in a multi-level segmentation module in a target detection model; wherein the number of feature extraction layers is at least two; The first spliced ​​ice-covered image is hierarchically extracted, convolved and up-sampled by using various feature extraction layers, convolution layers and up-sampling layers to segment and obtain the position information of the power transmission line area, ice-covered area and background area in the first spliced ​​ice-covered image, and marked in the first spliced ​​ice-covered image according to the position to obtain a first segmented ice-covered image.

4. The method according to claim 3, characterized in that The first segmented ice-covered image is obtained, including: The first feature extraction layer, the second feature extraction layer, the third feature extraction layer and the fourth feature extraction layer are used to extract and process the first spliced ​​ice-covered image respectively to obtain a first feature, a second feature, a third feature and a fourth feature; Performing upsampling processing on the fourth feature, and performing addition processing and convolution processing on the fourth feature after the upsampling processing and the third feature to obtain a fifth feature; Performing upsampling processing on the fifth feature, and performing addition processing and convolution processing on the fifth feature after the upsampling processing and the second feature to obtain a sixth feature; Performing upsampling processing on the sixth feature, and performing addition processing on the upsampling sixth feature and the first feature to obtain a seventh feature; The seventh feature is sequentially subjected to convolution processing, upsampling processing and secondary convolution processing to obtain the position information of the power transmission line area, the ice-covered area and the background area in the first spliced ​​ice-covered image, and marked in the first spliced ​​ice-covered image according to the position to obtain a first segmented ice-covered image.

5. The method according to claim 2, characterized in that: The lightweight classification module in the target detection model is used to classify the second segmented ice-covered image to obtain the corresponding ice-covered type, including: Determine a first classification convolution layer, an information extraction layer, a second classification convolution layer, an average pooling layer, and a fully connected layer included in a lightweight classification module in a target detection model; wherein the number of information extraction layers is at least two; Using the first classification convolution layer, convolution processing is performed on the second segmented ice-covered image to obtain a convolution image, and each information extraction layer is used to extract the convolution image in turn to obtain an extracted image; The second classification convolution layer, average pooling layer and fully connected layer are used to classify the extracted images and obtain the corresponding ice cover type.

6. The method according to claim 1, characterized in that Before determining the pre-trained object detection model, it also includes: Construct a training sample set based on historical ice cover images and their corresponding ice cover types; According to the training sample set, the initial detection model is trained based on the cross entropy loss function and the negative log-likelihood loss function to obtain the target detection model.

7. A transmission line ice treatment device, characterized in that: include: A determination module, for determining a target ice-covered image corresponding to the transmission line and a pre-trained target detection model if an ice-covered detection request for the transmission line is detected; The recognition module is used to identify and process the target ice-covered image based on the target ice-covered image by using the multi-level segmentation module and the lightweight classification module in the target detection model to obtain the corresponding ice-covered type; The processing module is used to determine the icing countermeasures for the transmission line according to the icing type, so as to process the icing on the transmission line according to the icing countermeasures.

8. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the power transmission line icing treatment method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the power transmission line icing treatment method according to any one of claims 1 to 6 when executed.

10. A computer program product, characterized in that The computer program product comprises a computer program, and when the computer program is executed by a processor, the method for processing ice coating on a power transmission line according to any one of claims 1 to 6 is implemented.

Images