Enameled flat wire on-line detection system and method

Abstract

The invention discloses an enamelled rectangular wire online detection system and method, relates to the technical field of enamelled rectangular wire quality inspection, solves the problems of limited range and poor precision of a current enamelled wire detection mode, and comprises a laser data acquisition unit, a laser initial inspection unit, an image acquisition unit, an image monitoring unit and a defect identification module, the laser initial detection unit is used for detecting and analyzing external monitoring data of the enameled rectangular wire, abnormal laser monitoring data, obtained through analysis, of the enameled rectangular wire are sent to the database module, and the image monitoring unit is used for detecting and analyzing surface image data of the enameled rectangular wire, obtaining abnormal image monitoring data of the enameled rectangular wire and sending the abnormal laser monitoring data to the database module. The defect identification module is used for analyzing the defect condition of the enameled rectangular wire in combination with the abnormal laser monitoring data and the abnormal image monitoring data, the abnormal degree of the enameled rectangular wire is obtained through analysis, and comprehensive detection and accurate detection of the enameled rectangular wire are achieved.

Description

Technical Field

[0001] The invention belongs to the technical field of online detection of enameled flat wires, and in particular relates to an online detection system and method for enameled flat wires. Background Art

[0002] Enameled rectangular wire is a common electrical material. Its core consists of a conductive metal wire (usually copper or aluminum) covered with a uniform insulating varnish. Enameled rectangular wire combines excellent electrical conductivity with effective current isolation, making it a key component in the manufacture of electrical equipment such as motors, transformers, and electromagnetic coils. Its heat, wear, and voltage resistance make it suitable for the complex operating environments of various industrial and electronics fields.

[0003] The Chinese invention patent application with application number 20211107597 proposes a method and system for detecting paint nodules on enameled wires. The method includes the following steps: passing the enameled wire through an online detection device to detect the current signal in real time; sequentially capturing the instantaneous large current signals in the current signal to construct a large current sequence; identifying the large current sequence and determining whether there are continuous paint nodules.

[0004] However, the above technical solution relies solely on detecting the surface of the enameled wire through electrical signals, ignoring the quality issues that may be caused by minor defects on the surface of the enameled wire. Furthermore, using only electrical signals for detection makes it difficult to fully identify protrusions, depressions, and color differences on the outer surface of the enameled wire, and the detection coverage and accuracy are limited.

[0005] To this end, the present invention provides an enameled flat wire online detection system and method. Summary of the Invention

[0006] The purpose of the present invention is to provide an online detection system and method for enameled flat wires to solve the problems raised in the above background technology.

[0007] The technical problems to be solved by the present invention are:

[0008] How to achieve comprehensive and accurate inspection of enameled flat wire through online inspection.

[0009] In order to achieve the above object, the present invention adopts the following technical solutions:

[0010] An enameled flat wire online detection system includes a laser monitoring module, an image detection module, a database module, a defect identification module, an integration module and a management terminal;

[0011] The laser monitoring module includes a laser data acquisition unit and a laser initial inspection unit. The laser data acquisition unit is used to collect external monitoring data of the enameled flat wire and send the collected external monitoring data of the enameled flat wire to the laser initial inspection unit. The laser initial inspection unit is used to detect and analyze the external monitoring data of the enameled flat wire and send abnormal laser monitoring data of the enameled flat wire obtained through analysis to the database module.

[0012] The image monitoring module includes an image acquisition unit and an image monitoring unit. The image acquisition unit is used to acquire surface image data of the enameled flat wire and send the surface image data of the enameled flat wire to the image monitoring unit. The image monitoring unit is used to detect and analyze the surface image data of the enameled flat wire, and send abnormal image monitoring data of the enameled flat wire obtained through analysis to the database module.

[0013] The database module is used to store abnormal laser monitoring data and abnormal image monitoring data; the defect identification module is used to analyze the defect conditions of the enameled flat wire in combination with the abnormal laser monitoring data and the abnormal image monitoring data, and the abnormal degree of the enameled flat wire is obtained by analysis and sent to the integration module. The integration module is used to integrate the alarm information corresponding to the enameled flat wire according to the abnormal degree, and send the alarm information of the enameled flat wire to the management terminal. The management terminal is used to view the online detection results of the enameled flat wire.

[0014] Preferably, the external monitoring data of the enameled flat wire is the acquisition coordinates of any surface of the enameled flat wire, and the external monitoring data acquisition process is specifically as follows:

[0015] A three-dimensional coordinate system is established, with the starting point of the conveyor belt as the origin, the horizontal plane parallel to the ground as the plane formed by the x-axis and the y-axis, the forward direction of the conveyor belt as the positive direction of the x-axis, and the direction perpendicular to the x-axis in the plane formed by the x-axis and the y-axis as the y-axis. The direction corresponding to the plane perpendicular to the x-axis and the y-axis is used as the z-axis to obtain the collection coordinates of the surface of any enameled flat wire, and then the collection coordinates of the surface of the enameled flat wire are merged into the external monitoring data of the enameled flat wire.

[0016] Preferably, the process of the laser initial inspection unit is as follows:

[0017] Obtaining external monitoring data of the enameled flat wire, and then preprocessing the external monitoring data;

[0018] The fitting function of the enameled flat wire is constructed based on the pre-processed external monitoring data;

[0019] The ideal enameled flat wire height zi0 corresponding to the collection coordinates on any enameled flat wire surface is calculated by the fitting function of the enameled flat wire, and then the enameled flat wire height zi corresponding to the collection coordinates on any enameled flat wire surface is obtained. The convex-concave deviation value ATP corresponding to the collection coordinates on any enameled flat wire surface is calculated by the formula ATP=zi-zi0.

[0020] Preferably, the working process of the laser initial inspection unit further includes:

[0021] When the concave-convex deviation value of the acquisition coordinates of any enameled flat wire surface is less than zero, it is determined that the enameled flat wire surface is concave, and the acquisition coordinates of the enameled flat wire surface are marked as concave acquisition coordinates; when the concave-convex deviation value of the acquisition coordinates of any enameled flat wire surface is equal to zero, it is determined that the enameled flat wire surface is normal, and there is no abnormality on the enameled flat wire surface; when the concave-convex deviation value of the acquisition coordinates of any enameled flat wire surface is greater than zero, it is determined that the enameled flat wire surface is convex, and the acquisition coordinates of the enameled flat wire surface are marked as convex acquisition coordinates;

[0022] When there is an abnormality in the acquisition coordinates of the surface of the enameled flat wire, a defect image around the acquisition coordinates of the surface of the enameled flat wire is acquired, and then the defect image is converted into an abnormal grayscale image, and the grayscale value of any acquisition coordinate in the abnormal grayscale image is obtained;

[0023] Obtaining the grayscale value of the normal acquisition coordinates of the surface of the enameled flat wire, and then calculating the average grayscale value of the normal acquisition coordinates of the surface of the enameled flat wire;

[0024] If the grayscale value of any acquisition coordinate in the abnormal grayscale image is greater than the average grayscale value of the normal acquisition coordinates, the concave-convex deviation value of the acquisition coordinate is obtained and included in the convex acquisition coordinate set. Then, the convex acquisition coordinate set is searched to obtain the maximum concave-convex deviation value in the convex acquisition coordinate set. The search is radiated outward from the acquisition coordinate corresponding to the maximum concave-convex deviation value. When the grayscale value of any acquisition coordinate is equal to the average grayscale value of the normal acquisition coordinates, the search is stopped. The acquisition coordinate before the radiation stops is taken as the convex edge acquisition coordinate. Similarly, the concave edge acquisition coordinate is obtained.

[0025] Connect the coordinates collected from the raised edges one by one to obtain the raised area on the surface of the enameled flat wire; connect the coordinates collected from the recessed edges one by one to obtain the recessed area on the surface of the enameled flat wire, and count the number of pixels in the raised area or the recessed area;

[0026] If the number of pixels is less than the threshold, no processing is performed because the defect of the convex or concave area is small and therefore ignored; if the number of pixels is greater than or equal to the threshold, the convex or concave area is recorded as an abnormal concave-convex area;

[0027] Mark the concave edge acquisition coordinates and the convex edge acquisition coordinates as a first acquisition coordinate set;

[0028] The minimum value of the concave-convex deviation value in the concave area of ​​the enameled flat wire surface, the maximum value of the concave-convex deviation value in the convex area, and the first acquisition coordinate set corresponding to the abnormal concave-convex area are recorded as abnormal laser monitoring data.

[0029] Preferably, the fitting function is as follows:

[0030] zi0=axi 2 +byi 2 +cxiyi+dxi+eyi+f, where i is the number corresponding to the acquisition coordinate of any enameled flat wire surface, i=1, 2, ..., n, where n is a positive integer, zi0 is the ideal enameled flat wire height corresponding to the acquisition coordinate of any enameled flat wire surface, xi and yi are the values ​​of any acquisition coordinate on the enameled flat wire surface on the X-axis and the Y-axis in the three-dimensional coordinate system, and a, b, c, d, e, and f are fitting parameters;

[0031] The fitting parameters are calculated using the formula, which is as follows:

[0032] P=(X T X) -1 X T Z, where P, X, and Z are matrices, X T is the transpose of matrix X, (X T X) -1 for (X T X) inverse matrix;

[0033]

[0034] Substitute the matrix P, matrix X, and matrix Z into the above formula to calculate the values ​​of the fitting parameters.

[0035] Preferably, the surface image data of the enameled flat wire is obtained by simultaneously taking two surface images of the enameled flat wire when the conveyor belt runs a preset length.

[0036] Preferably, the working process of the image monitoring unit is as follows:

[0037] The surface image data of the enameled flat wire is preprocessed. The preprocessing process is as follows:

[0038] The initial surface image data of the enameled flat wire is input into the image recognition model. The image recognition model recognizes the enameled flat wire through the surface image data, and then crops the surface image data of the enameled flat wire along the edge of the enameled flat wire. The image recognition model then outputs the cropped surface image data. The horizontal coordinate of the cropped surface image data starts at xj and ends at xj+x0. The vertical coordinate of the cropped surface image data starts at yj and ends at yj+y0, where x0 is the length of the enameled flat wire and y0 is the width of the enameled flat wire.

[0039] The cropped surface image data of the enameled flat wire is converted into a grayscale image using the formula. The formula is as follows:

[0040] Hm(up, vp)=0.3Rm(up, vp)+0.59Gm(up, vp)+0.11Bm(up, vp), where m is the surface image number in the cropped surface image data of the enameled flat wire, m=1, 2, p is the number of any pixel point in the grayscale image of the enameled flat wire, Hm is the grayscale value corresponding to any surface image in the cropped surface image data of the enameled flat wire when the pixel position is (up, vp), Rm is the R value in the RGB value corresponding to any surface image in the cropped surface image data of the enameled flat wire when the pixel position is (up, vp), and Gm and Bm are similarly the G value and B value in the RGB value corresponding to any surface image in the cropped surface image data of the enameled flat wire when the pixel position is (up, vp);

[0041] Calculate the average grayscale value PJZ(up, vp) of the grayscale value of any surface image at any pixel position in the cropped surface image data, and then calculate the pixel standard deviation BZC(up, vp) of any pixel position in the two images by the formula. The specific formula is as follows:

[0042]

[0043] When the pixel standard deviation of any pixel position is greater than the pixel standard deviation threshold, no operation is performed; when the pixel standard deviation of any pixel position is equal to the pixel standard deviation threshold, the pixel type of the pixel is determined to be a color difference edge pixel; when the pixel standard deviation of any pixel position is less than the pixel standard deviation threshold, the pixel type of the pixel is determined to be a normal area pixel;

[0044] Connecting the color difference edge pixels one by one to obtain a contour map of the color difference area in the surface image data corresponding to the enameled flat wire, and then counting the number of pixels in the contour map of the color difference area in the surface image data as the area of ​​the color difference area;

[0045] The acquisition coordinates of the pixel points closest to each color difference edge pixel in the three-dimensional coordinate system are calculated by the formula, and the acquisition coordinates of the pixel points closest to the color difference edge pixel are marked as the second acquisition coordinate set. The specific formula is as follows:

[0046] Where, the value range of xi is [xj, xj+x0], and the value range of yi is [yj, yj+y0];

[0047] The second acquisition coordinate set and the area of ​​the color difference region are recorded as abnormal image monitoring data.

[0048] Preferably, the working process of the defect identification module is as follows:

[0049] Matching the first acquisition coordinate set in the abnormal laser monitoring data with the second acquisition coordinate set in the abnormal image monitoring data, and calculating the distance between all the acquisition coordinates in the first acquisition coordinate set and all the coordinates in the second acquisition coordinate set one by one;

[0050] If the distance between any acquisition coordinate in the first acquisition coordinate set and any acquisition coordinate in the second acquisition coordinate set is greater than the distance threshold, no operation is performed;

[0051] If the distance between any acquisition coordinate in the first acquisition coordinate set and any acquisition coordinate in the second acquisition coordinate set is less than or equal to the distance threshold, the two sets of corresponding acquisition coordinates are marked as similar coordinate pairs, and the number of similar coordinate pairs is counted;

[0052] When the number of similar coordinate pairs is greater than the number threshold, it is determined that the first acquisition coordinate set and the second acquisition coordinate set match; when the number of similar coordinate pairs is less than or equal to the number threshold, it is determined that the first acquisition coordinate set and the second acquisition coordinate set do not match;

[0053] When the first acquisition coordinate set and the second acquisition coordinate set match, the abnormality of the enameled flat wire is recorded as severe abnormality;

[0054] When the first acquisition coordinate set and the second acquisition coordinate set do not match, the abnormality degree of the enameled rectangular wire is recorded as a slight abnormality.

[0055] Preferably, the integration process of the integration module is as follows:

[0056] When the abnormality of the enameled flat wire is severe, the minimum value of the concave-convex deviation value in the concave area or the maximum value of the concave-convex deviation value in the convex area and the area of ​​the color difference area are integrated as the first alarm information of the enameled flat wire;

[0057] When the abnormality of the enameled flat wire is slight, the integration module integrates the minimum concave-convex deviation value in the concave area or the maximum concave-convex deviation value in the convex area, or the area of ​​the color difference area into the second alarm information of the enameled flat wire.

[0058] The present invention also proposes an online detection method for enameled flat wire, which is specifically as follows:

[0059] Step S1, collecting external monitoring data of the enameled flat wire, and performing detection and analysis on the external monitoring data of the enameled flat wire to obtain abnormal laser monitoring data of the enameled flat wire;

[0060] Step S2, then collecting surface image data of the enameled flat wire, and performing detection and analysis on the surface image data of the enameled flat wire to obtain abnormal image monitoring data of the enameled flat wire;

[0061] Step S3, analyzing the defect of the enameled flat wire based on the abnormal laser monitoring data and the abnormal image monitoring data to obtain the abnormality degree of the enameled flat wire;

[0062] Step S4 : integrating the alarm information corresponding to the enameled flat wire according to the abnormality degree, and checking the online detection result of the enameled flat wire in combination with the alarm information.

[0063] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are:

[0064] The present invention first collects external monitoring data of the enameled flat wire, and detects and analyzes the external monitoring data of the enameled flat wire to obtain abnormal laser monitoring data of the enameled flat wire. Then, the surface image data of the enameled flat wire is collected, and the surface image data of the enameled flat wire is detected and analyzed to obtain abnormal image monitoring data of the enameled flat wire. The defect situation of the enameled flat wire is analyzed based on the abnormal laser monitoring data and the abnormal image monitoring data to obtain the abnormality degree of the enameled flat wire. The alarm information corresponding to the enameled flat wire is integrated according to the abnormality degree, and the online detection result of the enameled flat wire is checked in combination with the alarm information. The present invention realizes comprehensive and accurate detection of the enameled flat wire. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] To facilitate understanding by those skilled in the art, the present invention is further described below with reference to the accompanying drawings.

[0066] Figure 1 is a block diagram of the overall system of the present invention;

[0067] Figure 2 is a schematic cross-sectional view of the enameled rectangular wire of the present invention;

[0068] Figure 3 A top view of the enameled rectangular wire of the present invention;

[0069] Figure 4 Flow chart of the method of the present invention. DETAILED DESCRIPTION

[0070] The technical solutions of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0071] Example 1: Please refer to Figure 1-Figure 3 As shown, the technical solution provided by the present invention is: an enameled flat wire online detection system, including a laser monitoring module, an image detection module, a database module, a defect identification module, an integration module and a management terminal;

[0072] In a specific implementation, the laser monitoring module includes a laser data acquisition unit and a laser initial inspection unit. The laser data acquisition unit is used to collect external monitoring data of the enameled flat wire and send the collected external monitoring data of the enameled flat wire to the laser initial inspection unit;

[0073] It should be specifically noted that the external monitoring data of the enameled flat wire is the acquisition coordinates of the surface of any enameled flat wire;

[0074] In this embodiment, the external monitoring data acquisition process of the enameled flat wire is as follows:

[0075] A three-dimensional coordinate system is established, with the starting point of the conveyor belt as the origin, the horizontal plane parallel to the ground as the plane formed by the x-axis and y-axis, the forward direction of the conveyor belt as the positive direction of the x-axis, the direction perpendicular to the x-axis in the plane formed by the x-axis and y-axis as the y-axis, and the direction corresponding to the plane perpendicular to the x-axis and y-axis as the z-axis. The collected coordinates of the surface of any enameled flat wire are then merged into the external monitoring data of the enameled flat wire;

[0076] It should be noted that the maximum value of x is the maximum length of the enameled rectangular wire;

[0077] The laser initial inspection unit is used to detect and analyze the external monitoring data of the enameled flat wire to obtain abnormal laser monitoring data of the enameled flat wire. The analysis process is as follows:

[0078] Step 1: Obtain external monitoring data of the enameled flat wire, and then pre-process the external monitoring data;

[0079] Specifically, the preprocessing of the external monitoring data of the enameled flat wire is as follows: using the median filtering technology to remove the noise in the external monitoring data;

[0080] It should be specifically noted that the median filtering technology is an existing technology. The median filtering technology sorts the z values ​​in the external monitoring data of the enameled flat wire and takes the middle value of the sorted z values ​​as the denoising result. The median filtering technology can retain the concave and convex features of the surface of the enameled flat wire.

[0081] Step 2: Construct a fitting function for the enameled flat wire based on the pre-processed external monitoring data. The fitting function for the enameled flat wire is as follows:

[0082] zi0=axi 2 +byi 2 +cxiyi+dxi+eyi+f, where i is the number corresponding to the acquisition coordinate of any enameled flat wire surface, i=1, 2, ..., n, where n is a positive integer, zi0 is the ideal enameled flat wire height corresponding to the acquisition coordinate of any enameled flat wire surface, xi and yi are the values ​​of any acquisition coordinate on the enameled flat wire surface on the X-axis and the Y-axis in the three-dimensional coordinate system, and a, b, c, d, e, and f are fitting parameters;

[0083] Furthermore, the fitting parameters are calculated by the formula, which is as follows:

[0084] P=(X T X) -1 X T Z, where P, X, and Z are matrices, X T is the transpose of matrix X, (X T X) -1 (X T X), Substitute the matrix P, matrix X and matrix Z into the above formula to calculate the values ​​of each fitting parameter;

[0085] Step 3: The ideal enameled flat wire height zi0 corresponding to the collection coordinates on any enameled flat wire surface is calculated by the fitting function of the enameled flat wire. Then, the enameled flat wire height zi corresponding to the collection coordinates on any enameled flat wire surface is obtained. The concave-convex deviation value ATP corresponding to the collection coordinates on any enameled flat wire surface is calculated by the formula. The specific formula is as follows:

[0086] ATP=zi-zi0;

[0087] Step 4: When the concave-convex deviation value of the collected coordinates of any enameled flat wire surface is less than zero, it is determined that the surface of the enameled flat wire is concave, and the collected coordinates of the enameled flat wire surface are marked as concave collection coordinates;

[0088] When the concave-convex deviation value of the collected coordinates of any enameled flat wire surface is equal to zero, the surface of the enameled flat wire is judged to be normal, and at this time, there is no abnormality on the surface of the enameled flat wire;

[0089] When the concave-convex deviation value of the collection coordinates on the surface of any enameled flat wire is greater than zero, it is determined that the surface of the enameled flat wire is convex, and the collection coordinates on the surface of the enameled flat wire are marked as convex collection coordinates;

[0090] Step 5: When there is an abnormality in the acquisition coordinates of the surface of the enameled flat wire, a defect image around the acquisition coordinates of the surface of the enameled flat wire is acquired, and then the defect image is converted into an abnormal grayscale image, and the grayscale value of any acquisition coordinate in the abnormal grayscale image is obtained;

[0091] Furthermore, the grayscale value of the normal acquisition coordinates of the surface of the enameled flat wire is obtained, and then the average grayscale value of the normal acquisition coordinates of the surface of the enameled flat wire is calculated;

[0092] If the grayscale value of any acquisition coordinate in the abnormal grayscale image is greater than the average grayscale value of the normal acquisition coordinates, the concave-convex deviation value of the acquisition coordinate is obtained and included in the convex acquisition coordinate set. Then, the convex acquisition coordinate set is searched to obtain the maximum concave-convex deviation value in the convex acquisition coordinate set. The search is radiated outward from the acquisition coordinate corresponding to the maximum concave-convex deviation value. When the grayscale value of any acquisition coordinate is equal to the average grayscale value of the normal acquisition coordinates, the search is stopped. The acquisition coordinate before the radiation stops is taken as the convex edge acquisition coordinate. Similarly, the concave edge acquisition coordinate is obtained.

[0093] Furthermore, the coordinates collected from the raised edges are connected one by one to obtain the raised area on the surface of the enameled flat wire; the coordinates collected from the recessed edges are connected one by one to obtain the recessed area on the surface of the enameled flat wire, and the number of pixels in the raised area or the recessed area is counted;

[0094] If the number of pixels is less than the threshold, no processing is performed because the defect of the convex or concave area is small and therefore ignored; if the number of pixels is greater than or equal to the threshold, the convex or concave area is recorded as an abnormal concave-convex area;

[0095] Step 6: Mark the concave edge acquisition coordinates and the convex edge acquisition coordinates as a first acquisition coordinate set;

[0096] Step seven: record the minimum value of the concave-convex deviation value in the concave area of ​​the surface of the enameled flat wire, the maximum value of the concave-convex deviation value in the convex area, and the first acquisition coordinate set corresponding to the abnormal concave-convex area as abnormal laser monitoring data.

[0097] The laser initial inspection unit sends abnormal laser monitoring data to the database module.

[0098] Furthermore, the image monitoring module includes an image acquisition unit and an image monitoring unit, wherein the image acquisition unit is used to acquire surface image data of the enameled flat wire and send the surface image data of the enameled flat wire to the image monitoring unit;

[0099] It should be specifically stated that the surface image data of the enameled flat wire is: two surface images of the enameled flat wire are taken simultaneously when the conveyor belt runs a preset length;

[0100] For example, every time the conveyor belt runs 20 centimeters, the enameled rectangular wire is photographed simultaneously using the first camera and the second camera, with the angle between the first camera and the conveyor belt being 45 degrees, and the angle between the second camera and the conveyor belt being 135 degrees;

[0101] The image monitoring unit is used to detect and analyze the surface image data of the enameled flat wire. The analysis process is as follows:

[0102] Step P1: pre-processing the surface image data of the enameled flat wire. The pre-processing process is as follows:

[0103] Inputting initial surface image data of the enameled flat wire into an image recognition model, the image recognition model recognizes the enameled flat wire through the surface image data, and then cutting the surface image data of the enameled flat wire along the edge of the enameled flat wire, and then the image recognition model outputs the cut surface image data;

[0104] It should be specifically noted that the horizontal coordinate of the cropped surface image data starts at xj and ends at xj+x0; the vertical coordinate of the cropped surface image data starts at yj and ends at yj+y0, where x0 is the length of the enameled flat wire and y0 is the width of the enameled flat wire.

[0105] Step P2: Convert the cropped surface image data of the enameled flat wire into a grayscale image using the formula. The formula is as follows:

[0106] Hm(up, vp)=0.3Rm(up, vp)+0.59Gm(up, vp)+0.11Bm(up, vp), where m is the surface image number in the cropped surface image data of the enameled flat wire, m=1, 2, p is the number of any pixel point in the grayscale image of the enameled flat wire, Hm is the grayscale value corresponding to any surface image in the cropped surface image data of the enameled flat wire when the pixel position is (up, vp), Rm is the R value in the RGB value corresponding to any surface image in the cropped surface image data of the enameled flat wire when the pixel position is (up, vp), and Gm and Bm are similarly the G value and B value in the RGB value corresponding to any surface image in the cropped surface image data of the enameled flat wire when the pixel position is (up, vp);

[0107] Step P3, calculate the average grayscale value PJZ(up, vp) of the grayscale value of any surface image at any pixel position in the cropped surface image data, and then calculate the pixel standard deviation BZC(up, vp) of any pixel position in the two images by the formula. The specific formula is as follows:

[0108]

[0109] Step P4: When the pixel standard deviation at any pixel position is greater than the pixel standard deviation threshold, no operation is performed; when the pixel standard deviation at any pixel position is equal to the pixel standard deviation threshold, the pixel type of the pixel is determined to be a color difference edge pixel; when the pixel standard deviation at any pixel position is less than the pixel standard deviation threshold, the pixel type of the pixel is determined to be a normal area pixel;

[0110] Step P5: Connecting the color difference edge pixels one by one to obtain a contour map of the color difference region in the surface image data corresponding to the enameled flat wire, and then counting the number of pixels in the contour map of the color difference region in the surface image data as the area of ​​the color difference region;

[0111] In step P6, the acquisition coordinates of the pixel points closest to each color difference edge pixel in the three-dimensional coordinate system are calculated by a formula, and the acquisition coordinates of the pixel points closest to the color difference edge pixel are marked as a second acquisition coordinate set. The specific formula is as follows:

[0112] Where, the value range of xi is [xj, xj+x0], and the value range of yi is [yj, yj+y0];

[0113] Step P7, recording the second acquisition coordinate set and the area of ​​the color difference region as abnormal image monitoring data;

[0114] The image monitoring unit sends abnormal image monitoring data of the enameled flat wire to the database module.

[0115] Furthermore, the defect identification module is connected to a database module, and the database module stores abnormal laser monitoring data and abnormal image monitoring data;

[0116] The defect identification module is used to analyze the defects of the enameled flat wire by combining abnormal laser monitoring data and abnormal image monitoring data. The analysis process is as follows:

[0117] Matching the first acquisition coordinate set in the abnormal laser monitoring data with the second acquisition coordinate set in the abnormal image monitoring data, and calculating the distance between all the acquisition coordinates in the first acquisition coordinate set and all the coordinates in the second acquisition coordinate set one by one;

[0118] If the distance between any acquisition coordinate in the first acquisition coordinate set and any acquisition coordinate in the second acquisition coordinate set is greater than the distance threshold, no operation is performed;

[0119] If the distance between any acquisition coordinate in the first acquisition coordinate set and any acquisition coordinate in the second acquisition coordinate set is less than or equal to the distance threshold, the two sets of corresponding acquisition coordinates are marked as similar coordinate pairs, and the number of similar coordinate pairs is counted;

[0120] When the number of similar coordinate pairs is greater than the number threshold, it is determined that the first acquisition coordinate set and the second acquisition coordinate set match; when the number of similar coordinate pairs is less than or equal to the number threshold, it is determined that the first acquisition coordinate set and the second acquisition coordinate set do not match;

[0121] When the first acquisition coordinate set and the second acquisition coordinate set match, the abnormality of the enameled flat wire is recorded as severe abnormality;

[0122] When the first acquisition coordinate set and the second acquisition coordinate set do not match, the abnormality degree of the enameled rectangular wire is recorded as a slight abnormality;

[0123] Specifically, the first acquisition coordinate set and the second acquisition coordinate set match, that is, the abnormal concave-convex area corresponding to the first acquisition coordinate set overlaps with the color difference area corresponding to the second acquisition coordinate set; the first acquisition coordinate set and the second acquisition coordinate set do not match, that is, the abnormal concave-convex area corresponding to the first acquisition coordinate set does not overlap with the color difference area corresponding to the second acquisition coordinate set.

[0124] The defect identification module sends the abnormality degree of the enameled flat wire to the integration module, and the integration module is used to integrate the alarm information corresponding to the enameled flat wire according to the abnormality degree. The integration process is as follows:

[0125] When the abnormality of the enameled flat wire is severe, the minimum value of the concave-convex deviation value in the concave area or the maximum value of the concave-convex deviation value in the convex area and the area of ​​the color difference area are integrated as the first alarm information of the enameled flat wire;

[0126] When the abnormality of the enameled flat wire is mild, the integration module integrates the minimum value of the concave-convex deviation value in the concave area or the maximum value of the concave-convex deviation value in the convex area, or the area of ​​the color difference area into the second alarm information of the enameled flat wire;

[0127] The integration module sends the alarm information of the enameled flat wire to the management terminal, and the management terminal is used to view the online detection result of the enameled flat wire.

[0128] In this application, if a corresponding calculation formula appears, the above calculation formula is dimensionless and its numerical calculation is performed. The weight coefficient, proportional coefficient and other coefficients in the formula are set to a result value obtained by quantifying each parameter. Regarding the size of the weight coefficient and the proportional coefficient, as long as it does not affect the proportional relationship between the parameter and the result value, it is acceptable.

[0129] Example 2: Figure 4 Based on another concept of the same invention, this embodiment provides an online detection method for enameled flat wire, comprising the following steps:

[0130] Step S1, collecting external monitoring data of the enameled flat wire, and performing detection and analysis on the external monitoring data of the enameled flat wire to obtain abnormal laser monitoring data of the enameled flat wire;

[0131] Step S2, then collecting surface image data of the enameled flat wire, and performing detection and analysis on the surface image data of the enameled flat wire to obtain abnormal image monitoring data of the enameled flat wire;

[0132] Step S3, analyzing the defect of the enameled flat wire based on the abnormal laser monitoring data and the abnormal image monitoring data to obtain the abnormality degree of the enameled flat wire;

[0133] Step S4 : integrating the alarm information corresponding to the enameled flat wire according to the abnormality degree, and checking the online detection result of the enameled flat wire in combination with the alarm information.

[0134] The preferred embodiments of the present invention disclosed above are intended only to help illustrate the present invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the present invention to specific embodiments. Obviously, many modifications and variations are possible based on the contents of this specification. These embodiments are selected and described in detail in this specification to better explain the principles and practical applications of the present invention, thereby enabling those skilled in the art to better understand and utilize the present invention. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. An enameled flat wire online detection system, characterized in that: Including laser monitoring module, image detection module, database module, defect identification module, integration module and management terminal; The laser monitoring module includes a laser data acquisition unit and a laser initial inspection unit. The laser data acquisition unit is used to collect external monitoring data of the enameled flat wire and send the collected external monitoring data of the enameled flat wire to the laser initial inspection unit. The laser initial inspection unit is used to detect and analyze the external monitoring data of the enameled flat wire and send abnormal laser monitoring data of the enameled flat wire obtained through analysis to the database module. The image monitoring module includes an image acquisition unit and an image monitoring unit. The image acquisition unit is used to acquire surface image data of the enameled flat wire and send the surface image data of the enameled flat wire to the image monitoring unit. The image monitoring unit is used to detect and analyze the surface image data of the enameled flat wire, and send abnormal image monitoring data of the enameled flat wire obtained through analysis to the database module. The database module is used to store abnormal laser monitoring data and abnormal image monitoring data; the defect identification module is used to analyze the defect conditions of the enameled flat wire in combination with the abnormal laser monitoring data and the abnormal image monitoring data, and the abnormal degree of the enameled flat wire is obtained by analysis and sent to the integration module. The integration module is used to integrate the alarm information corresponding to the enameled flat wire according to the abnormal degree, and send the alarm information of the enameled flat wire to the management terminal. The management terminal is used to view the online detection results of the enameled flat wire.

2. The online detection system for enameled rectangular wire according to claim 1, characterized in that: The external monitoring data of the enameled flat wire is the collected coordinates of the surface of any enameled flat wire. The external monitoring data acquisition process is as follows: A three-dimensional coordinate system is established, with the starting point of the conveyor belt as the origin, the horizontal plane parallel to the ground as the plane formed by the x-axis and the y-axis, the forward direction of the conveyor belt as the positive direction of the x-axis, and the direction perpendicular to the x-axis in the plane formed by the x-axis and the y-axis as the y-axis. The direction corresponding to the plane perpendicular to the x-axis and the y-axis is used as the z-axis to obtain the collection coordinates of the surface of any enameled flat wire, and then the collection coordinates of the surface of the enameled flat wire are merged into the external monitoring data of the enameled flat wire.

3. The online detection system for enameled flat wire according to claim 2, characterized in that: The process of the laser initial inspection unit is as follows: Obtaining external monitoring data of the enameled flat wire, and then preprocessing the external monitoring data; The fitting function of the enameled flat wire is constructed based on the pre-processed external monitoring data; The ideal enameled flat wire height zi0 corresponding to the collection coordinates on any enameled flat wire surface is calculated by the fitting function of the enameled flat wire, and then the enameled flat wire height zi corresponding to the collection coordinates on any enameled flat wire surface is obtained. The convex-concave deviation value ATP corresponding to the collection coordinates on any enameled flat wire surface is calculated by the formula ATP=zi-zi0.

4. The online detection system for enameled rectangular wire according to claim 3, characterized in that: The working process of the laser initial inspection unit also includes: When the concave-convex deviation value of the acquisition coordinates of any enameled flat wire surface is less than zero, it is determined that the enameled flat wire surface is concave, and the acquisition coordinates of the enameled flat wire surface are marked as concave acquisition coordinates; when the concave-convex deviation value of the acquisition coordinates of any enameled flat wire surface is equal to zero, it is determined that the enameled flat wire surface is normal, and there is no abnormality on the enameled flat wire surface; when the concave-convex deviation value of the acquisition coordinates of any enameled flat wire surface is greater than zero, it is determined that the enameled flat wire surface is convex, and the acquisition coordinates of the enameled flat wire surface are marked as convex acquisition coordinates; When there is an abnormality in the acquisition coordinates of the surface of the enameled flat wire, a defect image around the acquisition coordinates of the surface of the enameled flat wire is acquired, and then the defect image is converted into an abnormal grayscale image, and the grayscale value of any acquisition coordinate in the abnormal grayscale image is obtained; Obtaining the grayscale value of the normal acquisition coordinates of the surface of the enameled flat wire, and then calculating the average grayscale value of the normal acquisition coordinates of the surface of the enameled flat wire; If the grayscale value of any acquisition coordinate in the abnormal grayscale image is greater than the average grayscale value of the normal acquisition coordinates, the concave-convex deviation value of the acquisition coordinate is obtained and included in the convex acquisition coordinate set. Then, the convex acquisition coordinate set is searched to obtain the maximum concave-convex deviation value in the convex acquisition coordinate set. The search is radiated outward from the acquisition coordinate corresponding to the maximum concave-convex deviation value. When the grayscale value of any acquisition coordinate is equal to the average grayscale value of the normal acquisition coordinates, the search is stopped. The acquisition coordinate before the radiation stops is taken as the convex edge acquisition coordinate. Similarly, the concave edge acquisition coordinate is obtained. Connect the coordinates collected from the raised edges one by one to obtain the raised area on the surface of the enameled flat wire; connect the coordinates collected from the recessed edges one by one to obtain the recessed area on the surface of the enameled flat wire, and count the number of pixels in the raised area or the recessed area; If the number of pixels is less than the threshold, no processing is performed because the defect of the convex or concave area is small and therefore ignored; if the number of pixels is greater than or equal to the threshold, the convex or concave area is recorded as an abnormal concave-convex area; Mark the concave edge acquisition coordinates and the convex edge acquisition coordinates as a first acquisition coordinate set; The minimum value of the concave-convex deviation value in the concave area of ​​the enameled flat wire surface, the maximum value of the concave-convex deviation value in the convex area, and the first acquisition coordinate set corresponding to the abnormal concave-convex area are recorded as abnormal laser monitoring data.

5. The online detection system for enameled flat wire according to claim 3, characterized in that: The fitting function is as follows: zi0=axi 2 +byi 2 +cxiyi+dxi+eyi+f, where i is the number corresponding to the acquisition coordinate of any enameled flat wire surface, i=1, 2, ..., n, where n is a positive integer, zi0 is the ideal enameled flat wire height corresponding to the acquisition coordinate of any enameled flat wire surface, xi and yi are the values ​​of any acquisition coordinate on the enameled flat wire surface on the X-axis and the Y-axis in the three-dimensional coordinate system, and a, b, c, d, e, and f are fitting parameters; The fitting parameters are calculated using the formula, which is as follows: P=(X T X) -1 X T Z, where P, X, and Z are matrices, X T is the transpose of matrix X, (X T X) -1 for (X T X) inverse matrix; Substitute the matrix P, matrix X, and matrix Z into the above formula to calculate the values ​​of the fitting parameters.

6. The online detection system for enameled rectangular wire according to claim 4, characterized in that: The surface image data of the enameled flat wire is as follows: two surface images of the enameled flat wire are taken simultaneously when the conveyor belt runs a preset length.

7. The online detection system for enameled rectangular wire according to claim 6, characterized in that: The working process of the image monitoring unit is as follows: The surface image data of the enameled flat wire is preprocessed. The preprocessing process is as follows: The initial surface image data of the enameled flat wire is input into the image recognition model. The image recognition model recognizes the enameled flat wire through the surface image data, and then crops the surface image data of the enameled flat wire along the edge of the enameled flat wire. The image recognition model then outputs the cropped surface image data. The horizontal coordinate of the cropped surface image data starts at xj and ends at xj+x0. The vertical coordinate of the cropped surface image data starts at yj and ends at yj+y0, where x0 is the length of the enameled flat wire and y0 is the width of the enameled flat wire. The cropped surface image data of the enameled flat wire is converted into a grayscale image using the formula. The formula is as follows: Hm(up, vp)=0.3Rm(up, vp)+0.59Gm(up, vp)+0.11Bm(up, vp), where m is the surface image number in the cropped surface image data of the enameled flat wire, m=1, 2, p is the number of any pixel point in the grayscale image of the enameled flat wire, Hm is the grayscale value corresponding to any surface image in the cropped surface image data of the enameled flat wire when the pixel position is (up, vp), Rm is the R value in the RGB value corresponding to any surface image in the cropped surface image data of the enameled flat wire when the pixel position is (up, vp), and Gm and Bm are similarly the G value and B value in the RGB value corresponding to any surface image in the cropped surface image data of the enameled flat wire when the pixel position is (up, vp); Calculate the average grayscale value PJZ(up, vp) of the grayscale value of any surface image at any pixel position in the cropped surface image data, and then calculate the pixel standard deviation BZC(up, vp) of any pixel position in the two images by the formula. The specific formula is as follows: When the pixel standard deviation of any pixel position is greater than the pixel standard deviation threshold, no operation is performed; when the pixel standard deviation of any pixel position is equal to the pixel standard deviation threshold, the pixel type of the pixel is determined to be a color difference edge pixel; when the pixel standard deviation of any pixel position is less than the pixel standard deviation threshold, the pixel type of the pixel is determined to be a normal area pixel; Connecting the color difference edge pixels one by one to obtain a contour map of the color difference area in the surface image data corresponding to the enameled flat wire, and then counting the number of pixels in the contour map of the color difference area in the surface image data as the area of ​​the color difference area; The acquisition coordinates of the pixel points closest to each color difference edge pixel in the three-dimensional coordinate system are calculated by the formula, and the acquisition coordinates of the pixel points closest to the color difference edge pixel are marked as the second acquisition coordinate set. The specific formula is as follows: Where, the value range of xi is [xj, xj+x0], and the value range of yi is [yj, yj+y0]; The second acquisition coordinate set and the area of ​​the color difference region are recorded as abnormal image monitoring data.

8. The online detection system for enameled rectangular wire according to claim 7, characterized in that: The working process of the defect identification module is as follows: Matching the first acquisition coordinate set in the abnormal laser monitoring data with the second acquisition coordinate set in the abnormal image monitoring data, and calculating the distance between all the acquisition coordinates in the first acquisition coordinate set and all the coordinates in the second acquisition coordinate set one by one; If the distance between any acquisition coordinate in the first acquisition coordinate set and any acquisition coordinate in the second acquisition coordinate set is greater than the distance threshold, no operation is performed; If the distance between any acquisition coordinate in the first acquisition coordinate set and any acquisition coordinate in the second acquisition coordinate set is less than or equal to the distance threshold, the two sets of corresponding acquisition coordinates are marked as similar coordinate pairs, and the number of similar coordinate pairs is counted; When the number of similar coordinate pairs is greater than the number threshold, it is determined that the first acquisition coordinate set and the second acquisition coordinate set match; when the number of similar coordinate pairs is less than or equal to the number threshold, it is determined that the first acquisition coordinate set and the second acquisition coordinate set do not match; When the first acquisition coordinate set and the second acquisition coordinate set match, the abnormality of the enameled flat wire is recorded as severe abnormality; When the first acquisition coordinate set and the second acquisition coordinate set do not match, the abnormality degree of the enameled rectangular wire is recorded as a slight abnormality.

9. The online detection system for enameled rectangular wire according to claim 8, characterized in that: The integration process of the integration module is as follows: When the abnormality of the enameled flat wire is severe, the minimum value of the concave-convex deviation value in the concave area or the maximum value of the concave-convex deviation value in the convex area and the area of ​​the color difference area are integrated as the first alarm information of the enameled flat wire; When the abnormality of the enameled flat wire is slight, the integration module integrates the minimum concave-convex deviation value in the concave area or the maximum concave-convex deviation value in the convex area, or the area of ​​the color difference area into the second alarm information of the enameled flat wire.

10. An online detection method for enameled flat wire, characterized in that: Based on the enameled flat wire online detection system according to any one of claims 1 to 9, the method is as follows: Step S1, collecting external monitoring data of the enameled flat wire, and performing detection and analysis on the external monitoring data of the enameled flat wire to obtain abnormal laser monitoring data of the enameled flat wire; Step S2, then collecting surface image data of the enameled flat wire, and performing detection and analysis on the surface image data of the enameled flat wire to obtain abnormal image monitoring data of the enameled flat wire; Step S3, analyzing the defect of the enameled flat wire based on the abnormal laser monitoring data and the abnormal image monitoring data to obtain the abnormality degree of the enameled flat wire; Step S4 : integrating the alarm information corresponding to the enameled flat wire according to the abnormality degree, and checking the online detection result of the enameled flat wire in combination with the alarm information.

Images