Cross-linked polyethylene (XLPE) slice water tree grading method based on gray level conversion and direction weight

Abstract

The invention belongs to the field of power transmission and transformation equipment operation and maintenance, and particularly relates to an XLPE slice water tree grading method based on gray level conversion and direction weight. The method comprises the following steps: firstly, carrying out RGB-gray conversion on a methylene blue dyeing picture of an XLPE section water tree; and then, carrying out direction weight weighting on the converted gray matrix. And finally, dividing water tree grades according to a weighting result. Through special RGB-grayscale conversion, the defect that red and green in a methylene blue dyed water tree slice image are easily influenced by a shooting environment is effectively avoided, and the weight of blue as a dye color is highlighted. Through weights in different directions, characterization parameters of the water tree are quantified, the difference of internal channels of the water tree can be reflected, and the difference of the length and the width of the water tree is more reasonably reflected.

Description

Technical Field

[0001] The invention belongs to the field of power transmission and transformation equipment operation and maintenance, and specifically belongs to a method for grading XLPE slice water trees based on grayscale conversion and directional weight. Background Art

[0002] With the continuous development of urban power grids, the cabling rate has increased year by year. As the last kilometer of power supply, the reliability of 10kV cable directly affects the reliability of power supply. At present, most 10kV cables are directly buried or laid in pipes, and they are operated in a humid environment for a long time. Water tree defects will be formed under external force damage, rat and insect bites, etc., which will lead to cable breakdown and power outage accidents. Sampling and slicing XLPE cables and evaluating their internal water tree levels are the key basis for formulating targeted maintenance and return strategies. At present, the water tree grading of XLPE slices is mainly based on the length and width of the water tree after dyeing. Some methods also use the ratio of water tree length to insulation thickness for evaluation, which is actually based on the length of the water tree. However, due to the growth characteristics of water trees, when the length and width of the water tree remain unchanged, more water tree channels will grow inside, which will lead to an increase in cable dielectric loss and a decrease in insulation strength. The traditional water tree grading method cannot reflect this difference, and it is urgent to carry out research on a new XLPE slice water tree grading method. Summary of the invention

[0003] The purpose of the present invention is to solve the problem that the current XLPE slice water tree grade classification can only be based on the length and width of the water tree after dyeing, and cannot reflect the growth state of the internal water tree channel. A XLPE slice water tree grade classification method based on grayscale conversion and directional weight is proposed.

[0004] The technical solution of the present invention is: a method for classifying water trees of XLPE slices based on grayscale conversion and directional weight, characterized in that the methylene blue dyed image of the water tree of the XLPE slice is first converted from RGB to grayscale. Preferably, the conversion from RGB to grayscale is performed according to formula (1), G=0.2×R+0.3×G+0.3×B (1).

[0005] Then, the converted grayscale matrix is ​​weighted by directional weights. Preferably, the root of the water tree is used as the coordinate origin. The horizontal coordinate of the grayscale matrix is ​​the pixel position in the width direction of the water tree, the vertical coordinate is the pixel position in the length direction of the water tree, and the matrix value is the grayscale value of the pixel point. The grayscale matrix is ​​weighted according to formula (2): W=∑G(0.3H+0.7Z) (2).

[0006] Finally, the water tree levels are divided according to the weighted results.

[0007] The beneficial effects of the present invention are as follows: Compared with the existing cable water tree severity assessment method, the present invention achieves the following beneficial effects: 1. Through the special RGB-grayscale conversion, the deficiency that the red and green colors in the methylene blue dyed water tree slice image are easily affected by the shooting environment is effectively avoided, and the weight of blue as a dye color is highlighted. 2. Through different directional weights, the characterization parameters of the water tree are quantified, which can reflect the differences in the internal channels of the water tree and more reasonably reflect the differences in the length and width of the water tree. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] Figure 1 It is a schematic diagram of the grayscale matrix of the present invention. DETAILED DESCRIPTION

[0009] See also Figure 1 The present invention provides a method for classifying water trees of XLPE slices based on grayscale conversion and directional weights. The working principle is: first, the methylene blue dyed image of the water tree of the XLPE slice is converted from RGB to grayscale, then the converted grayscale matrix is ​​weighted by directional weights, and finally the water tree level is classified according to the weighted results.

[0011] A method for grading water trees in XLPE slices based on grayscale conversion and directional weights is proposed. First, the methylene blue dyed image of the water tree in the XLPE slice is converted from RGB to grayscale. Since the methylene blue dyed image is in RGB (red, green, and blue) color, its value is easily affected by factors such as the shooting environment, which affects the accuracy of the weighted result. After conversion to grayscale, the image only contains grayscale information, without color information. Figure 1 In the image, the value of each pixel represents the brightness of the pixel, which is represented by an integer from 0 to 255, where 0 represents black, 255 represents white, and the values ​​in between represent different gray levels. This can effectively avoid the influence of factors such as the shooting environment and perform weighted calculations more accurately.

[0012] Preferably, the conversion from RGB to grayscale is performed according to formula (1): G=0.2×R+0.3×G+0.3×B (1).

[0013] Compared with the traditional weighting coefficient based on the human eye's color sensitivity (G=0.299×R+0.587×G+0.114×B), the weights of R (red) and G (green) are reduced, and the weight of B (blue) is increased, thereby further avoiding the deficiency that the red and green colors in the methylene blue-dyed water tree slice image are easily affected by the shooting environment, and highlighting the weight of blue as a dye color.

[0014] Then, the transformed grayscale matrix is ​​weighted by direction weight. Preferably, the root of the water tree is taken as the origin of the coordinate. The horizontal coordinate of the grayscale matrix is ​​the pixel position in the width direction of the water tree, the vertical coordinate is the pixel position in the length direction of the water tree, and the matrix value is the grayscale value of the pixel point. The grayscale matrix is ​​as follows: Figure 1 As shown. Since the water tree length has a greater impact on insulation, the weight coefficient assigned to the ordinate is higher. Preferably, the weight of the abscissa is 0.3 and the weight of the ordinate is 0.7. The calculation method of the corresponding water tree weighting is shown in formula (2).

[0015] W=∑G(0.3H+0.7Z) (2).

[0016] Where: G is the gray value of the pixel calculated according to formula (1), H is the absolute value of the horizontal coordinate of the pixel, and Z is the vertical coordinate of the pixel.

[0017] Finally, the water tree levels are divided according to the weighted results. The advantages of the present invention are: 1) through the special RGB-grayscale conversion, the deficiency that the red and green colors in the methylene blue dyed water tree slice image are easily affected by the shooting environment is effectively avoided, and the weight of blue as a dye color is highlighted. 2) through different directional weights, the characterization parameters of the water tree are quantified, which can reflect the differences in the internal channels of the water tree and more reasonably reflect the differences in the length and width of the water tree.

[0018] The above description is only a specific implementation mode of the present invention, and various examples do not constitute a limitation on the essential content of the present invention.

Claims

1. A water tree classification method for XLPE slices based on grayscale conversion and directional weight, characterized in that First, the methylene blue dyed image of the water tree of the XLPE slice is converted from RGB to grayscale. Preferably, the conversion from RGB to grayscale is performed according to formula (1): G=0.2×R+0.3×G+0.3×B (1); Then, the transformed grayscale matrix is ​​weighted by directional weights. Preferably, the root of the water tree is taken as the coordinate origin; the horizontal coordinate of the grayscale matrix is ​​the pixel position in the width direction of the water tree, the vertical coordinate is the pixel position in the length direction of the water tree, and the matrix value is the grayscale value of the pixel point; the grayscale matrix is ​​weighted according to formula (2): W=∑G(0.3H+0.7Z) (2); Finally, the water tree levels are divided according to the weighted results.

Images