Hydrophobicity grade evaluation method and equipment based on composite insulator water drop image

The composite insulator water droplet image is evaluated through the DynaMask-based neural network model, and combined with temperature and humidity information, the problem of low detection accuracy in the prior art is solved, and the efficiency, accuracy and stability of the composite insulator hydrophobicity evaluation is achieved.

CN119090805APending Publication Date: 2024-12-06ELECTRIC POWER SCI RES INST OF STATE GRID XINJIANG ELECTRIC POWER CO LTD +2
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202411022139.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-29
Publication Date
2024-12-06

AI Technical Summary

Technical Problem

The existing composite insulator hydrophobicity evaluation method has low detection accuracy and is greatly affected by image quality and environmental factors, resulting in increased uncertainty in the result.

Method used

A neural network model based on DynaMask is adopted, and the composite insulator water droplet images are obtained for preprocessing and annotation, the training set and test set are constructed, the neural network model is trained, and the temperature and humidity information is combined to compensate, and the hydrophobicity level is finally output.

Benefits of technology

It improves the robustness and accuracy of detection, can quickly and accurately evaluate the hydrophobicity of composite insulators, reduces the influence of environmental factors, and provides long-distance non-contact judgment ability, reducing leakage risks.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119090805A_ABST
    Figure CN119090805A_ABST
Patent Text Reader

Abstract

The invention provides a hydrophobicity grade evaluation method and device based on a composite insulator water drop image, and relates to the field of digital image technology processing, and the method comprises the steps: obtaining and marking the composite insulator water drop image, and constructing a training set and a test set; constructing a neural network model based on the DynaMask; training the neural network model by using the training set and the test set; obtaining a to-be-detected composite insulator water drop image and inputting the to-be-detected composite insulator water drop image into the trained neural network model to obtain a first hydrophobicity grade; and performing temperature and humidity compensation on the first hydrophobicity grade in combination with the temperature and humidity information, and outputting a final hydrophobicity grade. Compared with a traditional mean entropy method and a shape factor method, a DynaMask neural network model can neglect the influence of image quality and the like on a final detection result, temperature and humidity factors are considered, grade adjustment is performed according to temperature and humidity after the neural network outputs a hydrophobicity grade result, and the accuracy of composite insulator hydrophobicity grade evaluation is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of digital image technology processing, and in particular to a method and device for evaluating the hydrophobicity level of a composite insulator based on a water droplet image. Background Art

[0002] Composite insulators play an important role in the power system, especially in the field of high-voltage power transmission. Composite insulators have become the preferred material due to their light weight, corrosion resistance, high strength and good insulation performance. Among them, surface hydrophobicity is a key indicator to measure the performance of composite insulators, which is directly related to the safety and stability of power facilities. The surface of composite insulators with good hydrophobicity can keep water in the form of small water droplets, avoiding the formation of a continuous water film, thereby reducing surface leakage current, increasing flashover voltage, and ensuring the safe and stable operation of the power grid.

[0003] Earlier image processing technologies mainly relied on the mean entropy method and the shape factor method to evaluate hydrophobicity. Although these two methods have certain practicality, they are highly dependent on the quality of the image itself. Factors such as lighting, resolution, and focus will affect the accuracy of detection. In addition, environmental factors, such as humidity changes and temperature fluctuations, will also significantly interfere with the detection accuracy, resulting in increased uncertainty in the results. Summary of the invention

[0004] The purpose of the present invention is to provide a method and device for evaluating the hydrophobicity level of a composite insulator based on a water drop image in order to solve the problem of low detection accuracy in the existing method for evaluating the hydrophobicity of a composite insulator.

[0005] The above-mentioned purpose of the present application is achieved through the following technical solutions: S1: Obtain and annotate water drop images of composite insulators, and construct training and test sets; S2: Build a neural network model based on DynaMask; S3: Use the training set and the test set to train the neural network model; S4: obtaining a water drop image of the composite insulator to be tested and inputting it into the trained neural network model to obtain a first hydrophobicity grade; S5: Perform temperature and humidity compensation on the first hydrophobicity level in combination with the temperature and humidity information, and output a final hydrophobicity level.

[0006] Optionally, step S1 includes: S11: Obtain a water drop image of the composite insulator, evenly cut N sections, and obtain N cross-sectional water drop height maps; S12: determining the height interval of each water drop in the cross-sectional water drop height diagram; and marking the hydrophobicity level in sequence according to the height interval; S13: Calculate the average value of the hydrophobicity levels of the N cross-sectional water drop height images, which is the hydrophobicity level of the composite insulator image, and complete the hydrophobicity labeling of the composite insulator water drop image; S14: Divide the labeled composite insulator water droplet images according to a preset ratio to construct a training set and a test set.

[0007] Optionally, step S11 includes: The acquired composite insulator water drop image is preprocessed, and the specific steps of the preprocessing are as follows: S11a: converting the composite insulator water drop image into a grayscale image; S11b: Use bilinear interpolation to adjust the size of the composite insulator water drop image as follows:

[0008] in: Represents the integer coordinates Interpolate the four nearby points to get the coordinates The function value at ; Relative to integer coordinates The offset used to specify the position of the interpolation point. The function values ​​representing four adjacent points are weighted according to preset weights.

[0009] Optionally, the neural network model adopts a joint loss function as follows:

[0010]

[0011] In the formula, represents the mask loss function, which is used to evaluate the mask accuracy of the neural network; represents the edge loss function, which is used to evaluate the similarity between the edge of the neural network output image and the edge of the real image; Represents the calculation constraint loss function and The sum of information entropy loss functions; represents the weight coefficient; Represents the probability that the model prediction is correct; represents an adjustable weight factor; is a hyperparameter that controls the smoothness of the loss function.

[0012] Optionally, step S4 includes: The neural network model includes: a pyramid feature extraction network, a target detection network and a segmentation network; S41: preprocessing the water drop image of the actual composite insulator; S42: the pre-processed water drop image of the actual composite insulator is passed through a pyramid feature extraction network to obtain a multi-level feature image; S43: After the multi-level feature image passes through the target detection network, the water drop target classification result and the bounding box are obtained; S44: After the water drop target classification result and the bounding box are input into the segmentation network, the texture features of the water drop region are refined to generate a water drop segmentation result; S45: Calculate the water drop target classification result and the water drop segmentation result, and divide the composite insulator into hydrophobicity levels, the first hydrophobicity level.

[0013] Optionally, step S5 includes: The temperature and humidity information includes: relative humidity RH and ambient temperature; If the relative humidity RH is less than A% or the ambient temperature is greater than the preset temperature, correction is made at the first hydrophobicity level according to the empirical formula; If the relative humidity RH is greater than B% or the ambient temperature is less than the preset temperature, correction is made at the first hydrophobicity level according to the empirical formula.

[0014] An electronic device comprises a processor, a memory, a user interface and a network interface, wherein the memory is used to store instructions, the user interface and the network interface are used to communicate with other devices, and the processor is used to execute the instructions stored in the memory so that the electronic device executes a method for evaluating the hydrophobicity level based on water droplet images of composite insulators.

[0015] A computer-readable storage medium stores instructions. When the instructions are executed, a method for evaluating the hydrophobicity level based on a water drop image of a composite insulator is performed.

[0016] The beneficial effects of the technical solution provided by this application are: 1. Construct a neural network model based on DynaMask. Compared with the traditional mean entropy method and shape factor method, the influence of image quality on the final detection results can be ignored. Compared with the traditional method, the neural network model of DynaMask combines classification, bounding box and partial segmentation results to comprehensively evaluate the hydrophobicity level of composite insulators. The robustness of the model is increased by preprocessing the image. It effectively improves the problems of poor detection robustness caused by factors such as image quality and environment under the traditional algorithm, and can quickly and accurately detect the hydrophobicity of composite insulators, providing a basis for subsequent data analysis and processing. Considering the temperature and humidity factors, after the neural network outputs the hydrophobicity level result, it is adjusted according to the temperature and humidity. The technical solution of the present application can construct a digital twin of water droplets on the surface of a composite insulator based on the image, and the image after segmentation and drawing is input into the neural network to output the hydrophobicity level of the corresponding composite insulator, which can quickly judge the hydrophobicity of the composite insulator in a non-contact manner at a long distance, and thereby judge whether the composite insulator has hidden dangers such as leakage.

[0017] 2. In the DynaMask network model, the ROIAlign technology is introduced, and the bilinear interpolation method is used to accurately calculate the eigenvalue of each sampling point, avoiding information loss and improving the accuracy of the water drop segmentation results.

[0018] 3. Add a focal loss function based on the original loss function of the DynaMask network model, so that the recognition and segmentation focus on difficult-to-separate samples, increase the difference in the number of balanced samples, and make the model focus on difficult-to-separate samples. At the same time, introduce a weight coefficient to adjust the importance of the loss function and balance the problem of uneven sample number. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The present application will be further described below with reference to the accompanying drawings and embodiments, in which: Figure 1 is a step diagram of a method for evaluating the hydrophobicity level of a composite insulator based on a water drop image in an embodiment of the present application; Figure 2 is a cross-sectional water drop height map of a water repellency grade evaluation method based on a water drop image of a composite insulator in an embodiment of the present application; Figure 3 It is a partial enlarged view of the height of water drops in a cross section of the method for evaluating the hydrophobicity level based on the water drop image of the composite insulator in the embodiment of the present application; Figure 4 is a DynaMask network model diagram of a method for evaluating the hydrophobicity level of a composite insulator based on a water drop image in an embodiment of the present application; Figure 5 is a ROIAlign explanation diagram of the hydrophobicity grade evaluation method based on the water drop image of the composite insulator in the embodiment of the present application; Figure 6 It is a schematic diagram of the structure of an electronic device according to an embodiment of the present application, which is a method for evaluating the hydrophobicity level of a composite insulator based on a water droplet image. DETAILED DESCRIPTION

[0020] In order to have a clearer understanding of the technical features, purposes and effects of the present application, the specific implementation methods of the present application are now described in detail with reference to the accompanying drawings.

[0021] The embodiment of the present application provides a method for evaluating the hydrophobicity level of a composite insulator based on a water drop image.

[0022] Please refer to Figure 1 , Figure 1 : is a step diagram of a method for evaluating the hydrophobicity level of a composite insulator based on a water drop image in an embodiment of the present application, comprising: S1: Obtain and annotate water drop images of composite insulators, and construct training and test sets; Step S1 includes: S11: Obtain a water drop image of the composite insulator, evenly cut N sections, and obtain N cross-sectional water drop height maps; Step S11 includes: The acquired composite insulator water drop image is preprocessed, and the specific steps of the preprocessing are as follows: S11a: converting the composite insulator water drop image into a grayscale image; S11b: Use bilinear interpolation to adjust the size of the composite insulator water drop image as follows:

[0023] in: Represents the integer coordinates Interpolate the four nearby points to get the coordinates The function value at ; Relative to integer coordinates The offset used to specify the position of the interpolation point. The function values ​​representing four adjacent points are weighted according to preset weights.

[0024] Specifically, the preprocessing operation aims to enhance the robustness of the model, and image grayscale is used to eliminate redundant information.

[0025] S12: determining the height interval of each water drop in the cross-sectional water drop height diagram; and marking the hydrophobicity level in sequence according to the height interval; S13: Calculate the average value of the hydrophobicity levels of the N cross-sectional water drop height images, which is the hydrophobicity level of the composite insulator image, and complete the hydrophobicity labeling of the composite insulator water drop image; S14: Divide the labeled composite insulator water droplet images according to a preset ratio to construct a training set and a test set.

[0026] Specifically, the composite insulator water drop image data includes: water drop shape information and contour information. The water repellency level of the composite insulator water drop image is determined by human visual observation. N sections of composite insulator surface water drop cross-sections are uniformly intercepted and drawn as follows: Figure 2 The cross-sectional water drop height diagram is shown in Figure 1, and the local magnification is shown in Figure 2. Figure 3 As shown, the height of each water drop can be obtained; by counting the number of water droplets in different height intervals in each cross-sectional water drop height map and converting them into corresponding hydrophobicity levels (referring to the artificially defined water drop-hydrophobicity level comparison table), finally averaging the hydrophobicity levels of N cross-sectional images, the hydrophobicity level of the composite insulator image can be approximately obtained.

[0027] S2: Build a neural network model based on DynaMask; Specifically, the network structure can be divided into three main components: two-layer FPN, MSM module and loss function. The DynaMask network is based on the Faster R-CNN network framework, on which the ROI Pooling is changed and ROIAlign is added, which uses bilinear interpolation to accurately calculate the feature value of each sampling point. In addition, FCN is used to predict an M*M mask for each ROI, mark the pixel-level area of ​​the target object, and further complete the water drop segmentation task.

[0028] Specifically, after the image input is preprocessed and input into the neural network, the original image layer FPN is first used. The i-FPN (i-Feature Pyramid Network) uses the pyramid structure to extract image features at different levels, and these feature images are input into ROI Align to obtain the region of interest. Then, the r-FPN module is used to perform multi-scale feature fusion in the r-FPN ( Figure 4 In the middle path, the fused multi-scale features come from the output of i-FPN, which is aggregated at different feature scales and dynamically masked by the mask size selection module (MSM), and then hierarchically fused into the feature maps of r-FPN layers according to the mask size ( Figure 4 The other part comes directly from the feature maps of each layer of i-FPN ( Figure 4 The neural network will then output the classification of each water drop and the object box in the image.

[0029] Regarding the above ROI Align, in the target detection network model, because there is an fc layer in the output stage, it needs to be input to the fc layer with a fixed size. Therefore, in order to solve the problem that the input of feature maps corresponding to ROIs of different sizes must be input to the fc layer with a fixed size, the ROI Pooling method is proposed. However, target segmentation requires the output of a mask that accurately corresponds to pixel to pixel. The ROI Pooling method will cause the ROI on the original image to be misaligned with the corresponding area on the feature map. In the traditional Faster RCNN, there are two integerization processes, that is, the x, y, w, and h of the Region proposal are usually decimals, but they will be integerized for ease of operation. The integerized boundary area is evenly divided into k*k units, and the boundary of each unit is integerized. In fact, after the above two integerizations, the candidate box at this time has a certain deviation from the position regressed at the beginning. Because the classification task is translation invariant, the impact is relatively small, but this will have a very large negative impact when predicting a mask with pixel-level accuracy.

[0030] In order to solve the above-mentioned problems, the ROI Align method cancels the integer operation, retains the decimal, and uses the bilinear interpolation method to obtain the image value of the pixel point whose coordinates are floating point numbers. However, in actual operation, ROI Align does not simply supplement the coordinate points on the boundary of the candidate region and then perform pooling, but is redesigned.

[0031] like Figure 5 As shown in the figure, the dotted part represents the feature map, and the solid line represents the ROI. Here, the ROI is divided into 2×2 cells. If the number of sampling points is 4, then we first divide each cell into four small squares (as shown by the red line), and the center of each small square is the sampling point. The coordinates of these sampling points are usually floating point numbers, so it is necessary to perform bilinear interpolation on the pixels of the sampling points (as shown by the four arrows) to get the value of the pixel. Then maxpooling is performed on the four sampling points in each cell to get the final ROI Align result.

[0032] S3: Use the training set and the test set to train the neural network model; S4: obtaining a water drop image of the composite insulator to be tested and inputting it into the trained neural network model to obtain a first hydrophobicity grade; S5: Perform temperature and humidity compensation on the first hydrophobicity level in combination with the temperature and humidity information, and output a final hydrophobicity level.

[0033] The neural network model adopts a joint loss function as follows:

[0034]

[0035] In the formula, represents the mask loss function, which is used to evaluate the mask accuracy of the neural network; represents the edge loss function, which is used to evaluate the similarity between the edge of the neural network output image and the edge of the real image; Represents the calculation constraint loss function and The sum of information entropy loss functions; represents the weight coefficient; Represents the probability that the model prediction is correct; represents an adjustable weight factor; is a hyperparameter that controls the smoothness of the loss function.

[0036] Specifically, the adjustable weight factor helps deal with the problem of data imbalance by giving smaller weights to large amounts of data and larger weights to small amounts of data; the hyperparameter (artificially set) that controls the smoothness of the loss function, the larger the value, the smaller the loss of well-classified samples, making the network pay more attention to difficult-to-train samples.

[0037] Step S4 includes: The neural network model includes: a pyramid feature extraction network, a target detection network and a segmentation network; S41: preprocessing the water drop image of the actual composite insulator; S42: the pre-processed water drop image of the actual composite insulator is passed through a pyramid feature extraction network to obtain a multi-level feature image; S43: After the multi-level feature image passes through the target detection network, the water drop target classification result and the bounding box are obtained; S44: After the water drop target classification result and the bounding box are input into the segmentation network, the texture features of the water drop region are refined to generate a water drop segmentation result; S45: Calculate the water drop target classification result and the water drop segmentation result, and divide the composite insulator into hydrophobicity levels, the first hydrophobicity level.

[0038] Step S5 includes: The temperature and humidity information includes: relative humidity RH and ambient temperature; If the relative humidity RH is less than A% or the ambient temperature is greater than the preset temperature, correction is made at the first hydrophobicity level according to the empirical formula; If the relative humidity RH is greater than B% or the ambient temperature is less than the preset temperature, correction is made at the first hydrophobicity level according to the empirical formula. Specifically, as an embodiment, the above process is: 1. Set up the experimental environment as described in the following table (for example):

[0039] Measure the same water repellency level under different temperature and humidity conditions The hydrophobicity level of the composite insulator obtained by the technical solution of the present application .

[0040] 2. Use the least squares method to Fitting is performed, where is the fitted function, which is related to two variables: temperature T and relative humidity RH.

[0041] 3. For the fitted function, if the relative humidity RH is less than A% or the ambient temperature is greater than the preset temperature, correction is made at the first hydrophobicity level according to the above fitting formula; if the relative humidity RH is greater than B% or the ambient temperature is less than the preset temperature, correction is made at the first hydrophobicity level according to the fitting formula.

[0042] It should be noted that the selection of the threshold values ​​of temperature and relative humidity needs to be based on empirical fitting formulas. Within the circular range of T and RH where |ab|<1, no correction is required, but T and RH outside the circular range need to be corrected.

[0043] Specifically, temperature and humidity compensation is to label the water drop images of composite insulators under different temperature and humidity environments, so as to optimize the abnormal judgment of hydrophobicity level caused by the temperature and humidity affecting the evaporation rate of water droplets and thus indirectly affecting the shape of water droplets.

[0044] It should be noted that when the humidity is low and the temperature is high, the evaporation rate of water droplets on the composite insulator surface is high, which indirectly makes the water droplets smaller and more dispersed. At this time, whether to compensate for the hydrophobicity level is determined based on the artificially set threshold. The same applies when the temperature and humidity conditions change.

[0045] As an example, the whole process is as follows: The water droplet image on the composite insulator surface is reconstructed in three dimensions to obtain the digital twin of the water droplet on the composite insulator surface. The digital twin is divided into equal parts and the viewing angle is adjusted to obtain the following Figure 2 The N cross-sectional views shown are integrated.

[0046] Taking the processing of a single cross-sectional image as an example, the hydrophobicity is divided according to the height ranges of the water droplet height sets {h1, h2, h3, ..., hi} in the figure, which are respectively located in the height intervals in the table below to obtain the hydrophobicity level set {z1, z2, z3, ...zi}. The same is true for other cross-sectional images, and the total hydrophobicity level set {z1, z2, z3, ...zn} can be obtained. Finally, the average value of the total hydrophobicity level set {z1, z2, z3, ...zn} is calculated to obtain the hydrophobicity level of the water droplet image on the composite insulator surface and mark it.

[0047]

[0048] The water droplet images on the surface of the composite insulator that have been annotated as described above constitute a neural network data set, and are divided into a training set and a validation set in a ratio of 8:2 (for example) for training and validating the DynaMask neural network model.

[0049] The present application also discloses an electronic device. Figure 6 , Figure 6 The electronic device 500 may include: at least one processor 501 , at least one network interface 504 , a user interface 503 , a memory 505 , and at least one communication bus 502 .

[0050] The communication bus 502 is used to realize the connection and communication between these components.

[0051] The user interface 503 may include a display screen and a camera, and the optional user interface 503 may also include a standard wired interface and a wireless interface.

[0052] The network interface 504 may optionally include a standard wired interface or a wireless interface (such as a WI-FI interface).

[0053] The present application also discloses a computer-readable storage medium storing a plurality of instructions, wherein the instructions are suitable for a processor to load to execute the above-mentioned method for evaluating the hydrophobicity level based on the water droplet image of a composite insulator.

[0054] The above are only exemplary embodiments of the present disclosure and cannot be used to limit the scope of the present disclosure. That is, any equivalent changes and modifications made according to the teachings of the present disclosure are still within the scope of the present disclosure. After considering the disclosure of the specification and the truth of practice, those skilled in the art will easily think of other embodiments of the present disclosure.

[0055] This application is intended to cover any variation, use or adaptation of the present disclosure, which follows the general principles of the present disclosure and includes common knowledge or customary technical means in the art not described in the present disclosure. The description and examples are to be regarded as exemplary only, and the scope and spirit of the present disclosure are defined by the claims.

Claims

1. A method for evaluating the hydrophobicity level of composite insulators based on water droplet images, characterized in that: The method comprises the following steps: S1: Obtain and annotate water drop images of composite insulators, and construct training and test sets; S2: Build a neural network model based on DynaMask; S3: Use the training set and the test set to train the neural network model; S4: obtaining a water drop image of the composite insulator to be tested and inputting it into the trained neural network model to obtain a first hydrophobicity grade; S5: Perform temperature and humidity compensation on the first hydrophobicity level in combination with the temperature and humidity information, and output a final hydrophobicity level.

2. The method for evaluating the hydrophobicity level of a composite insulator based on a water drop image according to claim 1, characterized in that: Step S1 includes: S11: Obtain a water drop image of the composite insulator, evenly cut N sections, and obtain N cross-sectional water drop height maps; S12: determining the height interval of each water drop in the cross-sectional water drop height diagram; and marking the hydrophobicity level in sequence according to the height interval; S13: Calculate the average value of the hydrophobicity levels of the N cross-sectional water drop height images, which is the hydrophobicity level of the composite insulator image, and complete the hydrophobicity labeling of the composite insulator water drop image; S14: Divide the labeled composite insulator water droplet images according to a preset ratio to construct a training set and a test set.

3. The method for evaluating the hydrophobicity level of a composite insulator based on a water drop image according to claim 2, characterized in that: Step S11 includes: The acquired composite insulator water drop image is preprocessed, and the specific steps of the preprocessing are as follows: S11a: converting the composite insulator water drop image into a grayscale image; S11b: Use bilinear interpolation to adjust the size of the composite insulator water drop image as follows: , in Represents the integer coordinates Interpolate the four nearby points to get the coordinates The function value at ; Relative to integer coordinates The offset used to specify the position of the interpolation point. The function values ​​representing four adjacent points are weighted according to preset weights.

4. The method for evaluating the hydrophobicity level of a composite insulator based on a water drop image according to claim 1, characterized in that: The neural network model adopts a joint loss function as follows: , , In the formula, represents the mask loss function, which is used to evaluate the mask accuracy of the neural network; represents the edge loss function, which is used to evaluate the similarity between the edge of the neural network output image and the edge of the real image; Represents the calculation constraint loss function and The sum of information entropy loss functions; represents the weight coefficient; Represents the probability that the model prediction is correct; represents an adjustable weight factor; is a hyperparameter that controls the smoothness of the loss function.

5. The method for evaluating the hydrophobicity level of a composite insulator based on a water drop image according to claim 1, characterized in that: Step S4 includes: The neural network model includes: a pyramid feature extraction network, a target detection network and a segmentation network; S41: preprocessing the water drop image of the actual composite insulator; S42: the pre-processed water drop image of the actual composite insulator is passed through a pyramid feature extraction network to obtain a multi-level feature image; S43: After the multi-level feature image passes through the target detection network, the water drop target classification result and the bounding box are obtained; S44: After the water drop target classification result and the bounding box are input into the segmentation network, the texture features of the water drop region are refined to generate a water drop segmentation result; S45: Calculate the water drop target classification result and the water drop segmentation result, and divide the composite insulator into hydrophobicity levels, the first hydrophobicity level.

6. The method for evaluating the hydrophobicity level of a composite insulator based on a water drop image according to claim 1, characterized in that: Step S5 includes: The temperature and humidity information includes: relative humidity RH and ambient temperature; If the relative humidity RH is less than A% or the ambient temperature is greater than the preset temperature, correction is made at the first hydrophobicity level according to the empirical formula; If the relative humidity RH is greater than B% or the ambient temperature is less than the preset temperature, correction is made at the first hydrophobicity level according to the empirical formula.

7. An electronic device, characterized in that: The electronic device comprises a processor (501), a memory (505), a user interface (503) and a network interface (504), wherein the memory (505) is used to store instructions, the user interface (503) and the network interface (504) are used to communicate with other devices, and the processor (501) is used to execute the instructions stored in the memory (505) so that the electronic device executes the method according to any one of claims 1 to 6.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores instructions, and when the instructions are executed by a computer, the method steps according to any one of claims 1 to 6 are performed.

Citation Information

Cited By

  • Composite insulator dirt hydrophobicity imaging diagnosis system and method

    CN121186050A