A method for rating adhesion of electrical steel coatings

Abstract

The present invention discloses a method for rating the coating adhesion of electrical steel. The method comprises: preparing an electrical steel sample coated with an insulating coating; performing a bending test on the sample; acquiring a coating adhesion image; processing the coating adhesion image to obtain white pixels representing areas where the coating has fallen off and black pixels representing areas where the coating has not fallen off; calculating the coating fall-off ratio, and rating the coating adhesion of the electrical steel sample. The present invention has the beneficial effect of incorporating visualization and visual analysis techniques to obtain quantitative coating adhesion data, and rating the coating adhesion based on this data. This rating method is objective, highly accurate, and reduces the errors and deviations caused by manual visual evaluation.

Description

Technical Field

[0001] The present invention relates to the field of detection technology, and in particular to a method for rating the adhesion of an electrical steel coating. Background Art

[0002] Electrical steel, also known as silicon steel sheet, is an essential soft magnetic alloy used in the power, electronics, and military industries. It is also the most produced functional metal material, primarily used as the core for various motors, generators, and transformers. In the manufacture of electrical steel sheet products, forming an insulating coating is one of the final steps in the manufacturing process. Typically, electrical steel is coated with an insulating coating. The adhesion of electrical steel coatings reflects the bond strength between the coating and the steel substrate and is a key indicator of coating performance, directly impacting the insulation performance and lifespan of electrical steel products.

[0003] Currently, the coating adhesion rating of electrical steel is typically based on a method based on the national standard (GB / T2522-2017). This testing method involves shearing and punching coated electrical steel specimens to the specified dimensions, then bending them 180° tightly around a brass cylinder. The bent specimens are then straightened and visually assessed for coating loss, resulting in a grade of A, B, C, D, E, or F. The main issues with this testing method are: 1. The manual operation is non-standard. Bending the specimen 180° around the brass cylinder can lead to the specimen not being firmly attached to the cylinder or not bending fully, affecting the test results. 2. The severity and level of coating loss are defined verbally, lacking a quantitative evaluation method. In practice, due to varying skill levels and understanding among operators, inconsistent and unstable judgments can occur, making the test results less reliable. 3. Coating loss levels are categorized by range, without providing a specific numerical value. Consequently, coating adhesion within the same grade cannot be further subdivided, making it difficult to find optimal process parameters on-site. In summary, the current mainstream method uses manual bending of specimens and then qualitative evaluation by naked eye. The measurement results fluctuate greatly and cannot be accurately evaluated or calculated. Summary of the Invention

[0004] The purpose of the present invention is to provide a highly accurate and objective method for evaluating the adhesion of electrical steel coatings in order to address the deficiencies of the prior art.

[0005] The technical solution adopted by the present invention is: a method for rating the adhesion of electrical steel coatings, the method comprising:

[0006] S1. Prepare electrical steel samples coated with insulating coatings;

[0007] S2. Perform bending test on electrical steel samples;

[0008] S3. collecting an image of the coating after the bending test to obtain an original coating adhesion image;

[0009] S4. Rotate, grayscale, crop, and binarize the coating adhesion image to obtain the number of white pixels representing coating peeling areas and black pixels representing coating non-peeling areas:

[0010] S5. Calculate the coating shedding ratio respectively, coating shedding ratio = number of white pixels / (number of black pixels + number of white pixels);

[0011] S6. Rating the coating adhesion of the electrical steel samples based on the coating shedding ratio.

[0012] According to the above solution, the electrical steel sample is a oriented electrical steel sample or a non-oriented electrical steel sample.

[0013] According to the above scheme, in S2, when the sample is a grain-oriented electrical steel sample, the method for conducting the bending test is as follows: the grain-oriented electrical steel sample is placed horizontally with both ends pressed tightly; three brass cylinders with diameters of 10 mm, 20 mm, and 30 mm, respectively, are pressed down side by side and synchronously onto the surface of the grain-oriented electrical steel sample, so that the grain-oriented electrical steel sample completes a 180° bend in the three corresponding areas where the cylinders are pressed down; and the bent 180° grain-oriented electrical steel sample is then pulled into a horizontal state.

[0014] According to the above scheme, in S2, when the sample is a non-oriented electrical steel sample, the bending test method is as follows: the non-oriented electrical steel sample is placed horizontally and conveyed to the top of the tape with both ends tightened; then, a brass cylinder with a diameter of 10 mm is pressed from bottom to top against the tape and the surface of the non-oriented electrical steel sample, so that the tape is in close contact with the bottom surface of the non-oriented electrical steel sample, and the non-oriented electrical steel sample is bent 180 degrees within the pressing area of ​​the cylinder and maintained for at least 5 seconds; after removing the brass cylinder, the tape in close contact with the bottom surface of the non-oriented electrical steel sample is pulled into a straight state, and the tape falls off from the surface of the non-oriented electrical steel sample under the tension; after removing the non-oriented electrical steel sample, an image of the tape is collected.

[0015] According to the above scheme, when the sample is a grain-oriented electrical steel sample, the surface morphology of the grain-oriented electrical steel sample after being bent and then stretched into a straight state is collected as the coating adhesion image.

[0016] According to the above scheme, when the sample is a non-oriented electrical steel sample, the surface morphology of the tape after the non-oriented electrical steel sample is compacted by the tape and then released is collected as the coating adhesion image.

[0017] According to the above scheme, for the oriented electrical steel sample, the coating adhesion image is rotated, grayscaled, and cropped, and the X, Y, and Z regions corresponding to the brass cylinders with diameters of 10 mm, 20 mm, and 30 mm are framed in the image. The coating shedding ratios in the three regions are calculated, and the coating shedding ratio in the X region is α1, the coating shedding ratio in the Y region is α2, and the coating shedding ratio in the Z region is α3. The three regions are rated separately, and the rating levels are divided into six grades: A, B, C, D, E, and F, among which grade A is the best and grade F is the worst. The grading of each region can include all of the above grades or only include some of them (for example, the X region is only divided into grades A, B, C, and D). Finally, the coating adhesion rating of the entire electrical steel sample is obtained based on the rating of each region. The rating method is:

[0018] In the X-chart area: when α1 is less than 5.0%, the grade of the X-chart area is A; when 5.0% ≤ α1 < 30.0%, the grade of the X-chart area is B; when 30.0% ≤ α1 < 50.0%, the grade of the X-chart area is C; when α1 ≥ 50.0%, the grade of the X-chart area is D;

[0019] In the Y-graph area: when α2 < 1.0%, the grade of the Y-graph area is A; when 1.0% ≤ α2 < 20.0%, the grade of the Y-graph area is C; when α2 ≥ 20.0%, the grade of the Y-graph area is D;

[0020] In the Z area, when α3 < 1.0%, the grade of the Y area is A; when 1.0% ≤ α2 < 10.0%, the grade of the Y area is E; when α3 ≥ 10.0%, the grade of the Y area is F;

[0021] When the ratings of the three image areas are all A, the coating adhesion rating of the oriented electrical steel sample is A; when the ratings of the three image areas are inconsistent, the rating of the lowest-grade image area is used as the coating adhesion rating of the oriented electrical steel sample.

[0022] According to the above scheme, for non-oriented electrical steel samples, α is the coating shedding ratio of the non-oriented electrical steel samples, and the rating grades are divided into four grades: A, B, C and D. The specific rating method is:

[0023] When α is less than 5.0%, the coating adhesion rating of the non-oriented electrical steel sample is A;

[0024] When 5.0%≤α<30.0%, the coating adhesion rating of the non-oriented electrical steel sample is B;

[0025] When 30.0%≤α<50.0%, the coating adhesion rating of the non-oriented electrical steel sample is C;

[0026] When α≥50.0%, the coating adhesion rating of the non-oriented electrical steel sample is D.

[0027] According to the above scheme, in S2, the method of image binarization is: converting the image into pixel values, calculating the dynamic threshold to distinguish the image, setting the pixels above the dynamic threshold as the maximum value, the pixel with the maximum value is white, representing the area where the coating has fallen off; setting the pixels below the dynamic threshold as the minimum value, the pixel with the minimum value is black, representing the area where the coating has not fallen off.

[0028] According to the above scheme, the method for obtaining the dynamic threshold is as follows:

[0029] 1) Calculate the grayscale histogram of the image, and the grayscale value of each pixel is counted into the histogram;

[0030] 2) Calculate the grayscale value and probability of each type of pixel. For each possible threshold T, divide the image into two categories: pixels with grayscale values ​​less than T and pixels with grayscale values ​​greater than or equal to T. Calculate the average grayscale value of the two categories of pixels respectively.

[0031] 3) Calculate the inter-class variance, which is the square of the difference between the average grayscale values ​​of the two classes of pixels multiplied by the sum of the probabilities of the two classes of pixels;

[0032] 4) Find the optimal threshold, traverse all possible thresholds and calculate their corresponding inter-class variance. The threshold with the largest inter-class variance is the optimal global threshold, which is also the dynamic threshold sought.

[0033] The beneficial effects of the present invention are as follows: the present invention introduces visualization and visual analysis technology, collects coating adhesion images through industrial cameras, binarizes the images, calculates and obtains quantitative data of coating adhesion, and rates the coating adhesion based on the data. This rating method is objective and greatly reduces the errors and deviations caused by manual visual evaluation. The rating results are highly accurate and can provide a reference for improving the quality of coating adhesion. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 This is an image of a oriented electrical steel sample after bending.

[0035] Figure 2 Images of coating adhesion collected for oriented electrical steel samples.

[0036] Figure 3 Schematic diagram of image cropping and framing.

[0037] Figure 4 This is a schematic diagram of the image after binarization processing.

[0038] Figure 5 Schematic diagram for determining the dynamic threshold of an image. DETAILED DESCRIPTION

[0039] In order to better understand the present invention, the present invention is further described below with reference to the accompanying drawings and specific embodiments.

[0040] like Figure 1 The electrical steel coating adhesion rating method shown is specifically a quantitative rating method for electrical steel coating adhesion, which is:

[0041] S1. Prepare electrical steel samples coated with insulating coatings;

[0042] S2. Perform a bending test on the sample;

[0043] S3. collecting an image of the coating after the bending test to obtain an original coating adhesion image;

[0044] S4. Rotate, grayscale, crop, and binarize the coating adhesion image to obtain the number of white pixels representing coating peeling areas and black pixels representing coating non-peeling areas;

[0045] The binarization method is to convert the image into pixel values, calculate the dynamic threshold to distinguish the image, set the pixels above the dynamic threshold to the maximum value, the pixels with the maximum value are white, representing the coating peeling area; set the pixels below the dynamic threshold to the minimum value, the pixels with the minimum value are black, representing the coating not peeling area;

[0046] S5. Calculate the coating shedding ratio respectively, coating shedding ratio = number of white pixels / (number of black pixels + number of white pixels);

[0047] S6. Rating the coating adhesion of the electrical steel samples based on the coating shedding ratio.

[0048] The present invention can quantitatively grade the coating adhesion performance by taking oriented electrical steel and non-oriented electrical steel as samples.

[0049] In S1, the oriented electrical steel samples used were 0.10-0.35 mm thick and 7.65 g / cm 3 The w(Si+Al) content of the electrical steel sheet is generally below 6.5%, and the sample size is 320×30mm (length×width). The non-oriented electrical steel sample used is 0.10-0.65mm thick and has an agreed density of 7.60-7.85g / cm 3 The electrical steel sheet has a w (Si + Al) content generally below 4.5% and a size of 320mm × 30mm (length × width).

[0050] In S2, the bending tests of the oriented electrical steel samples were different from those of the non-oriented electrical steel samples.

[0051] When the sample is a grain-oriented electrical steel sample, a first bending device is used to perform a bending test on the grain-oriented electrical steel sample. The bending method is as follows: the grain-oriented electrical steel sample is placed horizontally with both ends pressed tightly; three brass cylinders with diameters of 10 mm, 20 mm, and 30 mm, respectively, are pressed down side by side and synchronously onto the surface of the grain-oriented electrical steel sample, so that the grain-oriented electrical steel sample completes a 180° bend in the three corresponding areas where the cylinders are pressed; and the bent 180° grain-oriented electrical steel sample is then pulled into a horizontal state.

[0052] The first bending device comprises a bending mechanism, a pneumatic clamping mechanism, a servo control mechanism, and a sample clamping and stretching mechanism. During operation, a GO-steel sample is moved to the bending area of ​​the bending mechanism. The pneumatic clamping mechanism clamps the horizontal GO-steel sample at both ends. The servo control mechanism simultaneously controls three parallel brass cylinders with diameters of 10mm, 20mm, and 30mm to press down on the surface of the GO-steel sample, completing a 180° bend. The servo control mechanism then controls the upward movement of the three brass cylinders, while the sample clamping and stretching mechanism straightens the GO-steel sample. In this method, the diameter tolerance of the brass cylinders fluctuates within a range of -0.5mm to 0.1mm.

[0053] When the sample is a non-oriented electrical steel sample, a second bending device is used to conduct a bending test on the non-oriented electrical steel sample. The specific bending method is as follows: the non-oriented electrical steel sample is placed horizontally and conveyed above the tape with both ends tightened; then, a brass cylinder with a diameter of 10 mm is pressed from bottom to top against the tape and the surface of the non-oriented electrical steel sample, so that the tape is in close contact with the lower surface of the non-oriented electrical steel sample, and the non-oriented electrical steel sample is bent 180° within the pressing area of ​​the cylinder for at least 5 seconds; after removing the brass cylinder, the tape in close contact with the lower surface of the non-oriented electrical steel sample is pulled into a straight state, and the tape falls off the surface of the non-oriented electrical steel sample under the tension. After the non-oriented electrical steel sample is removed, an image of the tape is collected.

[0054] In S2, the second bending device includes a bending mechanism, a pneumatic clamping mechanism, a servo control mechanism, a tape-pulling mechanism, and a sample clamping and stretching mechanism. The sample clamping and stretching mechanism straightens the bent non-oriented electrical steel sample and moves it to the bending area of ​​the bending mechanism. The servo control mechanism controls the bending mechanism (a 10mm diameter brass cylinder) to press upward against the tape and the surface of the non-oriented electrical steel sample, completing a 180° bend and maintaining the bend for at least 5 seconds. After the servo control mechanism controls the brass cylinder to move downward, the tape-pulling mechanism straightens the tape from the bottom surface of the non-oriented electrical steel sample and removes it from the sample. The sample clamping and stretching mechanism removes the non-oriented electrical steel sample and then captures an image of the tape. In this method, the diameter error of the brass cylinder should fluctuate within -0.5mm to 0.1mm.

[0055] In the present invention, the various mechanisms in the first bending device and the second bending device, such as the bending mechanism, the pneumatic pressing mechanism, the servo control mechanism, the tape pressing mechanism and the sample clamping and stretching mechanism, are all conventional mechanisms existing in the industry; the structure and function of the bending device are all existing and are not the invention point of this application, and will not be repeated here.

[0056] In S3, an industrial camera system is used to collect coating images to obtain the original coating adhesion image. When the sample is an oriented electrical steel sample, the surface morphology of the sample after being bent and then stretched into a flat state is collected as the original coating adhesion image, such as Figure 1 When the sample is a non-oriented electrical steel sample, the surface morphology of the tape after it is pressed on the surface for 5 seconds and then removed is collected as the original coating adhesion image.

[0057] In S3, the industrial camera system includes a camera, a lens, and a light source. The camera is a color camera with a resolution of not less than 4024×3036, a maximum frame rate of not less than 29.2fps, an exposure time of not less than 20μs, and a single pixel accuracy of not less than 0.08mm. The focal length of the lens should not exceed 16mm, the image plane diameter should be at least 17.6mm, the working distance should not be less than 400mm, and the field of view should not be less than 350×258mm. The light source is an LED ring light source with an operating voltage of 24V, a working distance of not less than 120mm, and a spacing of not less than 200mm; the light source is positioned on both sides of the sample to avoid direct reflection, ensure uniform lighting, and effectively highlight the texture and details of the sample. In the present invention, the industrial camera systems are all existing and will not be described here.

[0058] In S4, the collected coating adhesion image is rotated, grayscaled, cropped, and binarized in sequence to obtain the number of white pixels representing the coating peeling area and the number of black pixels representing the coating non-peeling area.

[0059] When the sample is an oriented electrical steel sample, the collected coating adhesion image is rotated, grayscaled, and cropped in sequence. Three pressing areas corresponding to brass cylinders with diameters of 10 mm, 20 mm, and 30 mm are framed: the X-area, the Y-area, and the Z-area. The images of these three areas are then binarized. When the sample is a non-oriented electrical steel sample, the collected coating adhesion image is directly rotated, grayscaled, cropped, and binarized without marking the pressing area.

[0060] The image binarization processing method is as follows: convert the image into pixel values, distinguish the image based on the dynamic threshold, set the pixel values ​​above the dynamic threshold to the maximum value, and the pixel points of the maximum value are white; set the pixel values ​​below the dynamic threshold to the minimum value, and the pixel points of the minimum value are black.

[0061] In the present invention, an image recognition system is used for image processing. The image recognition system is an image processing and visual recognition system developed based on OpenCV, including image rotation, grayscale, cropping, and binarization processing.

[0062] The specific method of image rotation processing is: using the image rotation function, adjust the function's parameter angle so that the image position is displayed correctly. The specific method of image grayscale processing is: using the grayscale algorithm to convert the color image into a grayscale image, that is, converting the RGB value into a brightness value. The specific method of image cropping is: using Canny edge detection (an algorithm used for image edge detection) to obtain the top, bottom, left and right boundary coordinates of the image area, using the image array slicing function to crop the coordinate edge area, and based on the actual physical size, according to the ratio of the actual distance to the image pixel; for oriented electrical steel samples, use the OpenCV function drawing library on the image to frame the downward pressure areas corresponding to brass cylinders with diameters of 10mm, 20mm, and 30mm respectively (that is, the areas corresponding to bending diameters of 10mm, 20mm, and 30mm), such as the X area, Y area, and Z area in the figure. Figure 2 As shown, the three regions are half of the circumference of the brass cylinder cross section.

[0063] The image binarization processing method is as follows: use the OpenCV image processing library cv2.threshold() function to convert the grayscale images of the three areas into simplified pixel values, and the pixel values ​​fluctuate between 0 and 255; distinguish the pixel values ​​based on the dynamic threshold, and assign the maximum pixel value of 255 to the pixel points with pixel values ​​greater than the dynamic threshold, and the corresponding pixel points are white; the pixel points with pixel values ​​less than the dynamic threshold are assigned the minimum pixel value of 0, and the corresponding pixel points are black.

[0064] In the present invention, the dynamic threshold is obtained by the Ostu algorithm, and the specific acquisition method is as follows:

[0065] 1) Calculate the grayscale histogram of the image. The sample size of the histogram is usually equal to the total number of pixels in the image. The grayscale value of each pixel (usually an integer between 0 and 255) is counted into the histogram;

[0066] 2) Calculate the grayscale value and probability of each type of pixel. For each possible threshold T, divide the image into two categories: pixels with grayscale values ​​less than T and pixels with grayscale values ​​greater than or equal to T. Calculate the average grayscale value of the two categories of pixels respectively.

[0067] 3) Calculate the inter-class variance, which is the square of the difference between the average grayscale values ​​of the two classes of pixels multiplied by the sum of the probabilities of the two classes of pixels;

[0068] 4) Find the optimal threshold, traverse all possible thresholds and calculate their corresponding inter-class variance. The threshold with the largest inter-class variance is the optimal global threshold, which is also the dynamic threshold sought. The dynamic threshold is usually between 120 and 180.

[0069] In S5 , the coating shedding ratio is calculated, where coating shedding ratio=number of white pixels / (number of black pixels+number of white pixels).

[0070] For the oriented electrical steel samples, the coating shedding ratios in the three regions were calculated respectively.

[0071] The quantitative calculation method of the peeling area is as follows: the number of pixels with a pixel value of 255 and a pixel value of 0 in the statistical area is counted. The white pixel with a pixel value of 255 represents the coating peeling area, and the black pixel with a pixel value of 0 represents the coating non-peeling area. The peeling area = the number of white pixels / (the number of black pixels + the number of white pixels).

[0072] S6. Determine the coating adhesion level of the electrical steel sample based on the coating shedding ratio.

[0073] In the present invention, a bending test is performed on a grain-oriented electrical steel sample using three brass cylinders with diameters of 10 mm, 20 mm, and 30 mm. The coating adhesion image is rotated, grayscaled, and cropped, and the regions X, Y, and Z corresponding to the brass cylinders with diameters of 10 mm, 20 mm, and 30 mm are framed in the image. The coating shedding ratios in the three regions are calculated, and the coating shedding ratio in the X region is α1, the coating shedding ratio in the Y region is α2, and the coating shedding ratio in the Z region is α3. The three regions are rated, and the ratings are divided into six grades, namely A, B, C, D, E, and F, according to the coating adhesion performance, with A being the best and F being the worst. The grading of each region may include all of the above grades or only include some of them (for example, the X region is only divided into A, B, C, and D). Finally, the coating adhesion rating of the entire electrical steel sample is obtained based on the ratings of each region, and the rating method is as follows:

[0074] In the X-chart area: when α1 is less than 5.0%, the grade of the X-chart area is A; when 5.0% ≤ α1 < 30.0%, the grade of the X-chart area is B; when 30.0% ≤ α1 < 50.0%, the grade of the X-chart area is C; when α1 ≥ 50.0%, the grade of the X-chart area is D;

[0075] In the Y-graph area: when α2 < 1.0%, the grade of the Y-graph area is A; when 1.0% ≤ α2 < 20.0%, the grade of the Y-graph area is C; when α2 ≥ 20.0%, the grade of the Y-graph area is D;

[0076] In the Z area, when α3 < 1.0%, the grade of the Y area is A; when 1.0% ≤ α2 < 10.0%, the grade of the Y area is E; when α3 ≥ 10.0%, the grade of the Y area is F;

[0077] When the ratings of the three image areas are all A, the coating adhesion rating of the oriented electrical steel sample is A; when the ratings of the three image areas are inconsistent, the rating of the lowest image area is used as the coating adhesion rating of the oriented electrical steel. For example, when the X image area is A, the Y image area is D, and the Z image area is E, the coating adhesion rating of the oriented electrical steel sample is E.

[0078] For non-oriented electrical steel samples, α is the coating shedding ratio of the non-oriented electrical steel sample. According to the coating adhesion performance, the rating levels are divided into four grades: A, B, C and D, among which A is the best and D is the worst. The specific rating method is:

[0079] When α is less than 5.0%, the coating adhesion rating of the non-oriented electrical steel sample is A;

[0080] When 5.0%≤α<30.0%, the coating adhesion rating of the non-oriented electrical steel sample is B;

[0081] When 30.0%≤α<50.0%, the coating adhesion rating of the non-oriented electrical steel sample is C;

[0082] When α≥50.0%, the coating adhesion rating of the non-oriented electrical steel sample is D.

[0083] In the present invention, due to the differences in the characteristics of different electrical steel coatings, the thresholds of different levels in a certain area can be appropriately adjusted.

[0084] The contents not described in detail in this specification belong to the prior art known to those skilled in the art.

[0085] Finally, it should be noted that the above are only preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or make equivalent replacements for some of the technical features therein. However, any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for rating the adhesion of electrical steel coatings, characterized in that: The method is: S1. Prepare electrical steel samples coated with insulating coatings; S2. Perform bending test on electrical steel samples; S3. collecting an image of the coating after the bending test to obtain an original coating adhesion image; S4. Rotate, grayscale, crop, and binarize the coating adhesion image to obtain the number of white pixels representing coating peeling areas and black pixels representing coating non-peeling areas: S5. Calculate the coating shedding ratio respectively, coating shedding ratio = number of white pixels / (number of black pixels + number of white pixels); S6. Rating the coating adhesion of electrical steel samples based on the coating shedding ratio; The electrical steel sample is a oriented electrical steel sample or a non-oriented electrical steel sample; For the grain-oriented electrical steel sample, the coating adhesion image was rotated, grayscaled, and cropped. The X, Y, and Z regions corresponding to the brass cylinders with diameters of 10 mm, 20 mm, and 30 mm were framed in the image. The coating shedding ratios in the three regions were calculated, with the coating shedding ratio in the X region being α1, the coating shedding ratio in the Y region being α2, and the coating shedding ratio in the Z region being α3. The three regions were rated separately, with the ratings divided into six grades: A, B, C, D, E, and F, with A being the best and F being the worst. The coating adhesion rating of the entire electrical steel sample was obtained based on the ratings of each region. The rating method is as follows: In the X-chart area: when α1 is less than 5.0%, the grade of the X-chart area is A; when 5.0%≤α1<30.0%, the grade of the X-chart area is B; when 30.0%≤α1<50.0%, the grade of the X-chart area is C; when α1≥50.0%, the grade of the X-chart area is D; In the Y-graph area: when α2 < 1.0%, the grade of the Y-graph area is A; when 1.0% ≤ α2 < 20.0%, the grade of the Y-graph area is C; when α2 ≥ 20.0%, the grade of the Y-graph area is D; In the Z-chart area, when α3 is less than 1.0%, the grade of the Z-chart area is A; when 1.0%≤α3<10.0%, the grade of the Z-chart area is E; when α3≥10.0%, the grade of the Z-chart area is F; When the ratings of the three image areas are all A, the coating adhesion rating of the oriented electrical steel sample is A; when the ratings of the three image areas are inconsistent, the rating of the lowest-grade image area is used as the coating adhesion rating of the oriented electrical steel sample.

2. The method for evaluating the adhesion of electrical steel coatings according to claim 1, wherein: In S2, when the sample is a grain-oriented electrical steel sample, the method for conducting the bending test is as follows: the grain-oriented electrical steel sample is placed horizontally with both ends pressed tightly; three brass cylinders with diameters of 10 mm, 20 mm, and 30 mm, respectively, are pressed down side by side and synchronously onto the surface of the grain-oriented electrical steel sample, so that the grain-oriented electrical steel sample completes a 180° bend in the three corresponding areas where the cylinders are pressed; and the bent 180° grain-oriented electrical steel sample is then pulled into a horizontal state.

3. The method for evaluating the adhesion of electrical steel coatings according to claim 1, wherein: In S2, when the sample is a non-oriented electrical steel sample, the method for conducting the bending test is as follows: the non-oriented electrical steel sample is placed horizontally and the two ends are tightened to be transported above the tape; then, a brass cylinder with a diameter of 10 mm is pressed from bottom to top against the tape and the surface of the non-oriented electrical steel sample, so that the tape is tightly attached to the lower surface of the non-oriented electrical steel sample, and the non-oriented electrical steel sample is bent 180° within the pressing area of ​​the cylinder and maintained for at least 5 seconds; after removing the brass cylinder, the tape tightly attached to the lower surface of the non-oriented electrical steel sample is pulled into a straight state, and the tape falls off from the surface of the non-oriented electrical steel sample under the tension, and an image of the tape is collected after the non-oriented electrical steel sample is removed.

4. The method for evaluating the adhesion of electrical steel coatings according to claim 1, wherein: When the sample is a grain-oriented electrical steel sample, the surface morphology of the grain-oriented electrical steel sample after being bent and then stretched into a straight state is collected as the coating adhesion image.

5. The method for evaluating the adhesion of electrical steel coatings according to claim 1, wherein: When the sample is a non-oriented electrical steel sample, the surface morphology of the tape after the non-oriented electrical steel sample is compacted by the tape and then released is collected as the coating adhesion image.

6. The method for evaluating the adhesion of electrical steel coatings according to claim 1, wherein: For non-oriented electrical steel samples, α is the coating shedding ratio of the non-oriented electrical steel sample. The rating grades are divided into four grades: A, B, C and D. The specific rating method is as follows: When α is less than 5.0%, the coating adhesion rating of the non-oriented electrical steel sample is A; When 5.0%≤α<30.0%, the coating adhesion rating of the non-oriented electrical steel sample is B; When 30.0%≤α<50.0%, the coating adhesion rating of the non-oriented electrical steel sample is C; When α≥50.0%, the coating adhesion rating of the non-oriented electrical steel sample is D.

7. The method for evaluating the adhesion of an electrical steel coating according to any one of claims 1 to 6, wherein: In S2, the image binarization process is as follows: convert the image into pixel values, calculate the dynamic threshold to distinguish the image, set the pixel points above the dynamic threshold as the maximum value, and the pixel point with the maximum value is white, representing the coating peeling area; The pixels below the dynamic threshold are set to the minimum value, and the pixels with the minimum value are black, representing the area where the coating has not fallen off.

8. The method for evaluating the adhesion of electrical steel coatings according to claim 7, wherein: The method for obtaining the dynamic threshold is as follows: 1) Calculate the grayscale histogram of the image, and the grayscale value of each pixel is counted into the histogram; 2) Calculate the grayscale value and probability of each type of pixel. For each possible threshold T, divide the image into two categories: pixels with grayscale values ​​less than T and pixels with grayscale values ​​greater than or equal to T. Calculate the average grayscale value of the two categories of pixels respectively. 3) Calculate the inter-class variance, which is the square of the difference between the average grayscale values ​​of the two classes of pixels multiplied by the sum of the probabilities of the two classes of pixels; 4) Find the optimal threshold, traverse all possible thresholds and calculate their corresponding inter-class variance. The threshold with the largest inter-class variance is the optimal global threshold, which is also the dynamic threshold sought.

Images