Polymer insulator corona discharge detection and evaluation system and method

Abstract

The invention provides a polymer insulator corona discharge detection and evaluation system and method, and relates to the technical field of polymer insulator detection.The method comprises the steps that an ultraviolet video of polymer insulator corona discharge is acquired based on a video acquisition module, the acquired ultraviolet video is preprocessed based on an image enhancement module, and the preprocessed ultraviolet video is acquired; and the attribute-based quantification module calculates related attributes according to the preprocessed ultraviolet video, the related attributes comprise the ultraviolet display area, the ultraviolet display perimeter, the shape factor and the persistence factor, and the attribute analysis processing module performs normalization processing on the obtained attributes so as to evaluate the operation state of the polymer insulator. According to the method, the related characteristics, including the area, the shape, the continuity and the like, of the corona discharge ultraviolet video of the polymer insulator are comprehensively extracted, the limitation of the prior art is overcome, the accuracy and the reliability of evaluation of the operating state of the polymer insulator are improved, and more effective guarantee is provided for stable operation of a power system.

Description

Technical Field

[0001] The present invention relates to the technical field of polymer insulator detection, and in particular to a polymer insulator corona discharge detection and evaluation system and method. Background Art

[0002] The reliable performance of polymer insulators plays a key role in the stable operation of power systems. However, during long-term operation, polymer insulators are susceptible to various factors, leading to corona discharge. Corona discharge not only consumes electrical energy but also damages the insulation performance of the insulator, seriously threatening the safe and stable operation of power systems. Currently, the following methods are used to detect corona discharge in polymer insulators and assess their operating status:

[0003] 1. Simple area-based evaluation method: The insulator condition is evaluated by calculating the area of the corona discharge region in the UV image. The image is first filtered for noise using thresholding or morphological operations, and then the sum of the corresponding corona discharge pixels is calculated. The area is expressed in absolute form or relative to the image size.

[0004] 2. Shape attribute approximation analysis method: Use attributes such as radius or diameter to describe the corona discharge shape and analyze the insulator state;

[0005] 3. Continuity evaluation method under arbitrary indicators: Try to calculate the repetition rate of corona discharge to reflect its changes over time.

[0006] Existing methods for detecting and evaluating corona discharge in polymer insulators have the following disadvantages:

[0007] 1. The simple evaluation method based on area only focuses on the area attribute, ignoring important attributes such as shape and continuity. It cannot fully reflect the characteristics of corona discharge, resulting in inaccurate evaluation of the insulator operating status.

[0008] 2. Shape attribute approximate analysis method: Since the actual shape of corona discharge is not a regular circle, the use of these attributes is limited and cannot accurately characterize the shape characteristics, which affects the judgment of the insulator status;

[0009] 3. Continuity evaluation method using arbitrary indicators: This method attempts to calculate the repetition rate of corona discharge to reflect its changes over time. However, the indicators used are relatively arbitrary and do not fully consider the impact of all frames of UV video. Continuity is also not considered as a key attribute. At the same time, the photon count of a single video frame cannot clearly reflect the operating status of the insulator, resulting in inaccurate and incomplete evaluation results.

[0010] Therefore, it is very necessary to design a corona discharge detection and evaluation system and method for polymer insulators. Summary of the Invention

[0011] In order to overcome the deficiencies of the prior art, the present invention aims to provide a polymer insulator corona discharge detection and evaluation system and method.

[0012] To achieve the above object, the present invention provides the following solutions:

[0013] The present invention provides a polymer insulator corona discharge detection and evaluation system, comprising: a video acquisition module for acquiring ultraviolet video of polymer insulator corona discharge;

[0014] Image enhancement module, used to pre-process the acquired UV video;

[0015] Attribute quantification module, used to calculate relevant attributes based on the preprocessed UV video;

[0016] The attribute analysis and processing module is used to normalize the acquired attributes and then evaluate the operating status of the polymer insulator.

[0017] Preferably, the video acquisition module is a special camera specifically used to detect ultraviolet emissions.

[0018] The present invention also provides a method for detecting and evaluating corona discharge of polymer insulators, comprising:

[0019] Step 1: Acquire ultraviolet video of corona discharge of polymer insulator based on the video acquisition module;

[0020] Step 2: Preprocess the acquired UV video based on the image enhancement module;

[0021] Step 3: Calculate relevant attributes based on the preprocessed UV video based on the attribute quantification module. The relevant attributes include the area of UV appearance, the perimeter of UV appearance, the shape factor, and the persistence factor.

[0022] Step 4: Based on the attribute analysis and processing module, the acquired attributes are normalized to evaluate the operating status of the polymer insulator.

[0023] Preferably, in step 1, the ultraviolet video of the corona discharge of the polymer insulator is obtained based on the video acquisition module, specifically:

[0024] During the voltage application test of 220kV polymer insulators, a special camera is aimed at key areas of defects on the polymer insulator to capture ultraviolet video of corona discharge on the polymer insulator.

[0025] Preferably, in step 2, the acquired ultraviolet video is preprocessed based on the image enhancement module, specifically:

[0026] Obtain ultraviolet video, perform insulator shape segmentation on it, and use a semi-automatic method combined with Otsu's criterion to separate the insulator shape from the ultraviolet video image;

[0027] Segment the UV image of the UV video. For each frame in the UV video, apply the Otsu criterion to segment the UV image produced by corona discharge.

[0028] The acquired UV images are integrated, and the UV effects in each frame of the video are accumulated frame by frame. To avoid bit saturation, each frame is weighted during the integration process. The specific operation is to divide the pixel value of each frame by the total number of frames to obtain the integrated image.

[0029] The noise of the integrated image is removed using morphological opening and closing operations.

[0030] Preferably, in step 3, the area of ultraviolet appearance is calculated based on the pre-processed ultraviolet video based on the attribute quantification module, specifically:

[0031] Scan the image after noise elimination, count the number of pixels corresponding to UV exposure using a counter, and calculate the area of UV exposure as:

[0032]

[0033] Where UVManifestationPixels is the ultraviolet manifestation pixel.

[0034] Preferably, in step 3, the perimeter of the ultraviolet appearance is calculated based on the pre-processed ultraviolet video based on the attribute quantification module, specifically:

[0035] Analyze the n-8 connectivity of each UV-imaged pixel to determine which pixels are edge pixels. After traversing the entire image and identifying all edge pixels, scan again and calculate the perimeter of the UV-imaged image as:

[0036]

[0037] Where UVManifestationBoundaryPixels is the UV manifestation edge pixel.

[0038] Preferably, in step 3, the shape factor is calculated based on the pre-processed ultraviolet video based on the attribute quantification module, specifically:

[0039] The formula that defines the shape factor is:

[0040]

[0041] Where Perimeter is the perimeter of the outer surface, and Area is the area of the UV surface.

[0042] Preferably, in step 3, the persistence factor is calculated based on the pre-processed ultraviolet video based on the attribute quantification module, specifically:

[0043] By scanning the image to identify the pixels corresponding to the UV exposure, a counter records the pixel value of each pixel in the red channel of the RGB system, and the duration of the UV exposure in the video is calculated as:

[0044]

[0045] Where UVManifestationPixelsValues is the UV manifestation pixel parameter value, and the duration of UV manifestation in the video is defined as the persistence factor.

[0046] Preferably, in step 4, the acquired attributes are normalized based on the attribute analysis and processing module, and then the operating status of the polymer insulator is evaluated, specifically:

[0047] Since the extracted attributes have different orders of magnitude, they are normalized as follows:

[0048]

[0049] Where x n represents the normalized attribute value, x represents the attribute to be normalized, and X is the extracted attribute set. The normalized attributes are displayed using a radar chart. The radar chart displays quantitative variables on multiple axes starting from the same point in the form of multiple variables to analyze the operating status of polymer insulators.

[0050] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0051] The present invention provides a polymer insulator corona discharge detection and evaluation system and method. The system includes a video acquisition module, an image enhancement module, an attribute quantification module, and an attribute analysis and processing module. The method includes acquiring ultraviolet video of the polymer insulator corona discharge using the video acquisition module, preprocessing the acquired ultraviolet video using the image enhancement module, calculating relevant attributes based on the preprocessed ultraviolet video using the attribute quantification module. The relevant attributes include the ultraviolet visible area, the ultraviolet visible perimeter, the shape factor, and the persistence factor. The acquired attributes are normalized using the attribute analysis and processing module to evaluate the operating status of the polymer insulator. The present invention has the following beneficial effects:

[0052] 1. Comprehensive display of multiple attributes: Traditional methods often focus only on a single attribute, such as the area of the corona discharge region. However, the radar chart of the present invention can simultaneously display multiple attributes, including area, perimeter, shape factor, and persistence. For example, for new insulators, the radar chart shows that their attribute values are relatively low; for insulators with tracking and erosion, the area and perimeter attribute values are increased, and the persistence is also enhanced. This multi-attribute comparison can fully understand the corona discharge characteristics of the insulator, overcome the one-sidedness of traditional methods, and more accurately assess the operating status of the insulator.

[0053] 2. Visually display attribute differences: Insulators in different operating states appear distinctly different on the radar chart. New insulators and cracked insulators have weaker corona discharge, resulting in shorter attribute radii on the radar chart. Insulators with corroded and eroded end fittings and exposed cores have higher attribute values and longer corresponding attribute radii on the radar chart. This intuitive display of differences allows maintenance personnel to quickly identify insulators in different states, improving inspection efficiency.

[0054] 3. Standardized data facilitates analysis: After calculating the attributes, normalization is performed so that attributes of different dimensions can be compared in the same radar chart. Through this standardization, the attributes of different insulators are under the same measurement standard, eliminating the analysis difficulties caused by different attribute units and magnitudes, facilitating more accurate quantitative analysis, and facilitating the subsequent use of classification tools such as machine learning and artificial intelligence, thereby enhancing the practicality and scalability of the algorithm. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in 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 paying any creative work.

[0056] Figure 1 This is a flow chart of a method for detecting and evaluating corona discharge of polymer insulators according to an embodiment of the present invention;

[0057] Figure 2 is a schematic diagram of pixel connectivity;

[0058] Figure 3 Radar diagram of attribute quantification and analysis results. DETAILED DESCRIPTION

[0059] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0060] Traditional algorithms usually only perform segmentation and attribute calculations on a single frame of image, such as calculating the area after segmentation using only the threshold method, or filtering noise only through morphological operations. In terms of image integration, existing methods may ignore the temporal dimension information of multiple frames of data and fail to perform equal-weighted averaging.

[0061] The present invention aims to provide a system and method for detecting and evaluating corona discharge in polymer insulators. The system uses multi-frame integration to enhance the temporal persistence of corona discharge by integrating images. This is in contrast to existing technologies that rely on single-frame analysis and are susceptible to random noise interference.

[0062] The present invention adopts a systematic process, combining semi-automated segmentation, multi-step morphological processing and multi-frame integration to form a complete enhancement process. In contrast to the existing technology, the existing algorithm steps are scattered and have not formed a standardized system.

[0063] This paper utilizes a multi-attribute combination, introducing shape factors (non-circular characteristics) and persistence (the time dimension) to address the shortcomings of using only area. Visual analysis and radar charts provide intuitive comparisons of multidimensional data. However, most existing studies use only area as a single evaluation metric, or incorporate circular parameters such as radius and diameter, without considering shape diversity and time persistence. Furthermore, existing analysis methods lack multi-attribute visualization and rely on numerical comparisons, which are highly subjective.

[0064] The present invention adopts full-frame accumulation and directly utilizes the pixel values of all frames, avoiding the influence of subjective thresholds and random frames, and accurately reflecting its long-term discharge state. The physical meaning of the present invention is clear, and the persistence is positively correlated with the total discharge energy. The existing technology evaluates persistence through subjective indicators such as "repetition rate", such as manually setting thresholds to count the proportion of frames in which discharge occurs, or only analyzing the photon count of a single frame. These methods do not fully utilize all frame data, rely on manually set parameters, and lack objectivity.

[0065] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0066] The present invention provides a polymer insulator corona discharge detection and evaluation system, comprising:

[0067] Video acquisition module, used to obtain UV videos of corona discharge on polymer insulators. These videos serve as the raw data source for subsequent analysis.

[0068] The image enhancement module is used to pre-process the acquired UV video to implement functions such as insulator shape segmentation, UV appearance segmentation, image integration, and noise removal, thereby improving image quality and providing more accurate data for attribute quantification.

[0069] The attribute quantification module is used to calculate relevant attributes based on the pre-processed UV video, including the area, perimeter, shape factor and persistence of the corona discharge;

[0070] The attribute analysis and processing module is used to normalize the acquired attributes and display and analyze them through radar charts and other methods, thereby evaluating the operating status of the polymer insulator.

[0071] The video acquisition module is a special camera designed to detect ultraviolet emissions. It captures ultraviolet video during the voltage application test of 220kV polymer insulators. During the test, the camera is aimed at key areas of defects on the insulator to ensure that valid video containing corona discharge information is obtained.

[0072] like Figure 1 As shown, the present invention also provides a method for detecting and evaluating corona discharge of polymer insulators, comprising the following steps:

[0073] Step 1: Acquire ultraviolet video of corona discharge of polymer insulator based on the video acquisition module;

[0074] Step 2: Preprocess the acquired UV video based on the image enhancement module;

[0075] Step 3: Calculate relevant attributes based on the preprocessed UV video using the attribute quantification module. The relevant attributes include the UV-visible area, UV-visible perimeter, shape factor, and persistence factor. These attributes are described below:

[0076] UV-visible area: the total number of pixels in the UV region, reflecting the discharge intensity;

[0077] UV-visible perimeter: the number of pixels at the edge of the UV area, which helps determine the complexity of the shape;

[0078] Shape factor: area / perimeter, which overcomes the limitations of traditional circular assumptions (such as radius and diameter) and is applicable to discharge areas of arbitrary shapes;

[0079] Persistence factor: The pixel values of all frames are accumulated to quantify the discharge time distribution;

[0080] Step 4: Based on the attribute analysis and processing module, the acquired attributes are normalized to evaluate the operating status of the polymer insulator.

[0081] In step 1, the ultraviolet video of the corona discharge of the polymer insulator is obtained based on the video acquisition module, specifically:

[0082] During the voltage application test of 220kV polymer insulators, a special camera is aimed at key areas of defects on the polymer insulator to capture ultraviolet video of corona discharge on the polymer insulator.

[0083] In step 2, the acquired UV video is preprocessed based on the image enhancement module, specifically:

[0084] Preprocessing is mainly used to perform image enhancement processing on the acquired UV video, which includes insulator shape segmentation processing, UV appearance segmentation processing, image integration processing, and noise elimination processing. Each of them is introduced in detail:

[0085] 1. Insulator shape segmentation

[0086] A semi-automatic method is used in combination with the Otsu criterion to separate the insulator shape from the image of the ultraviolet video. The Otsu criterion is based on the grayscale image and can determine an ideal threshold value to divide the background and foreground elements in the image into two different clusters and assign them different colors, so as to achieve accurate segmentation of the insulator shape. Specifically, the threshold value is determined by maximizing the inter-class variance to separate the insulator and the background. Based on the Otsu criterion, the variance σ between the background and the foreground is calculated by traversing the threshold t. 2 (t), choose σ 2 (t) The largest t is used as the segmentation threshold;

[0087] 2. UV display segmentation processing

[0088] For each frame of the UV video, the Otsu criterion is applied separately to segment the UV manifestation produced by the corona discharge. This method can accurately separate the UV manifestation of the corona discharge from the other parts of the image.

[0089] 3. Image integration processing

[0090] The UV effect of each frame of video is accumulated frame by frame to form a comprehensive image. To avoid bit saturation, each frame is weighted during the integration process. The specific operation is to divide the pixel value of each frame by the total number of frames to obtain the integrated image. Taking 1800 frames as an example, the formula is:

[0091]

[0092] Where Dn For the nth frame image, weighting is used to avoid bit depth saturation;

[0093] 4. Noise cancellation processing

[0094] The integrated image is subjected to noise elimination processing using morphological opening and closing operations. The structural element used is composed of a central pixel and its n-4 connected neighborhood. Through this processing step, those ultraviolet manifestations that are considered to be false, poorly persistent, and cannot truly reflect the corona discharge effect can be removed, thereby obtaining a clear image without noise interference, providing a reliable basis for subsequent attribute quantification.

[0095] In step 3, the area of UV exposure is calculated based on the pre-processed UV video using the attribute quantification module, specifically:

[0096] Scan the image after noise elimination, count the number of pixels corresponding to UV exposure using a counter, and calculate the area of UV exposure as:

[0097]

[0098] Where UVManifestationPixels is the ultraviolet manifestation pixel.

[0099] In step 3, the perimeter of the UV display is calculated based on the pre-processed UV video using the attribute quantification module, specifically:

[0100] Analyze the n-8 connectivity of each UV-imaged pixel to determine which pixels are edge pixels. After traversing the entire image and identifying all edge pixels, scan again and calculate the perimeter of the UV-imaged image as:

[0101]

[0102] Where UVManifestationBoundaryPixels is the UV manifestation edge pixel.

[0103] In step 3, the shape factor is calculated based on the preprocessed UV video based on the attribute quantification module, specifically:

[0104] The formula that defines the shape factor is:

[0105]

[0106] Where Perimeter is the perimeter of the outer surface, and Area is the area of the UV surface.

[0107] In step 3, the persistence factor is calculated based on the pre-processed UV video based on the attribute quantification module, specifically:

[0108] By scanning the image to identify the pixels corresponding to the UV exposure, a counter records the pixel value of each pixel in the red channel of the RGB system, and the duration of the UV exposure in the video is calculated as:

[0109]

[0110] Where UVManifestationPixelsValues is the UV manifestation pixel parameter value, and the duration of UV manifestation in the video is defined as the persistence factor.

[0111] In step 4, the acquired attributes are normalized based on the attribute analysis and processing module to evaluate the operating status of the polymer insulator, specifically:

[0112] Since the extracted attributes have different orders of magnitude, they are normalized to facilitate comparison, as follows:

[0113]

[0114] Where x n represents the normalized attribute value, x represents the attribute to be normalized, and X is the set of extracted attributes. The normalized attributes are displayed using a radar chart. Radar charts use multivariate representations, displaying quantitative variables on multiple axes starting from the same point, to analyze the operating status of polymer insulators. For example, new polymer insulators have relatively low corona discharge-related attribute values; whereas insulators with defects such as tracking and erosion, end fitting corrosion, and erosion with exposed cores will have corresponding attribute values that increase to varying degrees. By comparing these attribute value differences, the operating status of the insulator can be accurately identified. The present invention ultimately uses a radar chart to display the four normalized attributes, mapping them to a four-dimensional radar chart, allowing the insulator status to be intuitively distinguished through the graphical outline.

[0115] The present invention provides an embodiment, which conducts comparative tests on five types of typical defective insulators (IL01-IL05) on a 220kV test platform:

[0116] Test plan:

[0117] 1. Synchronously adopt the traditional area method, commercial software (CoronaScanPro 3.0) and the algorithm of the present invention;

[0118] 2. Take the measurement value of the high-voltage laboratory partial discharge detector as the benchmark (PD value);

[0119] 3. Test 100 groups of samples for each type;

[0120] The final experimental results are shown in Table 1 and Table 2;

[0121] Table 1 Experimental results

[0122] Defect Type Correlation coefficient (traditional) Correlation coefficient (business) Correlation coefficient (present invention) End corrosion 0.672 0.785 0.932 surface erosion 0.553 0.703 0.891 Core exposure 0.634 0.812 0.947 Crack defects 0.412 0.523 0.835 New product 0.387 0.465 0.796

[0123] Table 2 Comparison of misjudgment rates

[0124] Test Group Traditional methods Business Software The present invention Normal → Defective 28.7％ 15.2％ 3.4％ Defect → Normal 16.3％ 9.7％ 1.8％

[0125] It can be seen from Table 1 and Table 2 that the technical effect achieved by the present invention is the best.

[0126] It should also be noted that if Figure 2 As shown, the present invention provides a schematic diagram of pixel connectivity, which is introduced respectively;

[0127] n-4 connectivity: In digital images, n-4 connectivity means that a pixel (m,n) is connected to its adjacent pixels in the vertical and horizontal directions. In the figure, if only n-4 connectivity is considered, the pixel (m,n) is only connected to the four pixels (m-1,n), (m+1,n), (m,n-1), and (m,n+1). This connectivity definition is relatively simple and mainly focuses on the horizontal and vertical adjacency of pixels. It is often used in some image processing operations, such as simple edge detection and region filling algorithms, because it is relatively easy to calculate and can quickly determine the positions of adjacent pixels;

[0128] n-8 connectivity: n-8 connectivity is more comprehensive. It connects pixel (m,n) not only to its vertical and horizontal neighbors, but also to pixels along the diagonal line, namely (m-1,n-1), (m-1,n+1), (m+1,n-1), and (m+1,n+1). The figure fully demonstrates these connections. n-8 connectivity is advantageous in scenarios such as complex object recognition and image segmentation. It can describe the relationships between pixels in more detail and capture more complex structures and edge information in images.

[0129] m and n are coordinate values used to determine the position of pixels in a two-dimensional image. In the pixel matrix of a two-dimensional image, the position of the pixel in the horizontal direction is determined by m, and the position of the pixel in the vertical direction is determined by n. Through the coordinate combination of (m,n), every pixel in the image can be accurately located. Taking the common image coordinate system as an example, the upper left corner of the image can be regarded as the coordinate origin (0,0). The value of m increases from left to right, and the value of n increases from top to bottom. When describing the (n–4) connectivity and (n-8) connectivity of pixels, (m,n) is used as the center pixel, and the range and specific connection status of the connectivity are determined based on its coordinate relationship with the surrounding pixels.

[0130] The present invention provides a radar diagram of attribute quantification and analysis results. The radar diagram displays data using multiple quantitative variables on axes starting from the same point. In the radar diagram used to study corona discharge of polymer insulators, there are usually four attribute axes, representing area, perimeter, shape factor, and persistence. The radius length of each axis is proportional to the normalized value of the corresponding attribute. The quadrilateral shape formed by these axes is used to display the attribute characteristics of different polymer insulators.

[0131] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0132] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The above examples are only intended to help understand the method and core concept of the present invention. At the same time, those skilled in the art will find that the specific implementation methods and application scopes may vary based on the concept of the present invention. In summary, the contents of this specification should not be construed as limiting the present invention.

Claims

1. A polymer insulator corona discharge detection and evaluation system, characterized in that: include: Video acquisition module, used to obtain UV video of corona discharge of polymer insulators; Image enhancement module, used to pre-process the acquired UV video; Attribute quantification module, used to calculate relevant attributes based on the preprocessed UV video; The attribute analysis and processing module is used to normalize the acquired attributes and then evaluate the operating status of the polymer insulator.

2. The system according to claim 1, wherein: The video acquisition module is a special camera specifically used to detect ultraviolet radiation.

3. A method for detecting and evaluating corona discharge of polymer insulators, characterized in that: include: Step 1: Acquire ultraviolet video of corona discharge of polymer insulator based on the video acquisition module; Step 2: Preprocess the acquired UV video based on the image enhancement module; Step 3: Calculate relevant attributes based on the preprocessed UV video based on the attribute quantification module. The relevant attributes include the area of UV appearance, the perimeter of UV appearance, the shape factor, and the persistence factor. Step 4: Based on the attribute analysis and processing module, the acquired attributes are normalized to evaluate the operating status of the polymer insulator.

4. The method according to claim 3, characterized in that In step 1, the ultraviolet video of the corona discharge of the polymer insulator is obtained based on the video acquisition module, specifically: During the voltage application test of 220kV polymer insulators, a special camera is aimed at key areas of defects on the polymer insulator to capture ultraviolet video of corona discharge on the polymer insulator.

5. The method according to claim 4, characterized in that In step 2, the acquired UV video is preprocessed based on the image enhancement module, specifically: Obtain ultraviolet video, perform insulator shape segmentation on it, and use a semi-automatic method combined with Otsu's criterion to separate the insulator shape from the ultraviolet video image; Segment the UV image of the UV video. For each frame in the UV video, apply the Otsu criterion to segment the UV image produced by corona discharge. The acquired UV images are integrated, and the UV effects in each frame of the video are accumulated frame by frame. To avoid bit saturation, each frame is weighted during the integration process. The specific operation is to divide the pixel value of each frame by the total number of frames to obtain the integrated image. The noise of the integrated image is removed using morphological opening and closing operations.

6. The method according to claim 5, characterized in that In step 3, the area of UV exposure is calculated based on the pre-processed UV video using the attribute quantification module, specifically: Scan the image after noise elimination, count the number of pixels corresponding to UV exposure using a counter, and calculate the area of UV exposure as: Where UVManifestationPixels is the ultraviolet manifestation pixel.

7. The method according to claim 6, characterized in that In step 3, the perimeter of the UV display is calculated based on the pre-processed UV video using the attribute quantification module, specifically: Analyze the n-8 connectivity of each UV-imaged pixel to determine which pixels are edge pixels. After traversing the entire image and identifying all edge pixels, scan again and calculate the perimeter of the UV-imaged image as: Where UVManifestationBoundaryPixels is the UV manifestation edge pixel.

8. The method according to claim 7, characterized in that In step 3, the shape factor is calculated based on the preprocessed UV video based on the attribute quantification module, specifically: The formula that defines the shape factor is: Where Perimeter is the perimeter of the outer surface, and Area is the area of the UV surface.

9. The method according to claim 8, characterized in that In step 3, the persistence factor is calculated based on the pre-processed UV video based on the attribute quantification module, specifically: By scanning the image to identify the pixels corresponding to the UV exposure, a counter records the pixel value of each pixel in the red channel of the RGB system, and the duration of the UV exposure in the video is calculated as: Where UVManifestationPixelsValues is the UV manifestation pixel parameter value, and the duration of UV manifestation in the video is defined as the persistence factor.

10. The method according to claim 9, characterized in that In step 4, the acquired attributes are normalized based on the attribute analysis and processing module to evaluate the operating status of the polymer insulator, specifically: Since the extracted attributes have different orders of magnitude, they are normalized as follows: Where x n represents the normalized attribute value, x represents the attribute to be normalized, and X is the extracted attribute set. The normalized attributes are displayed using a radar chart. The radar chart displays quantitative variables on multiple axes starting from the same point in the form of multiple variables to analyze the operating status of polymer insulators.

Images